Sport Training – The Sport Journal

A Longitudinal Cross-Sectional Analysis of Physical Fitness and Motor Competency for Intermediate School Students

Authors: Moez Baklouti¹

¹Full Professor, Human Sciences Department, Institut Superieur de Sport et de l’Education Physique, University of Mannouba, Tunisia

Editor’s Note: This article uses the pseudonym Nm.Wr.Qs. The Sport Journal has discussed this with the author. The acronym represents a school in North America, and The Sport Journal has confirmed that the school and district exist. This note serves to assure readers that reasonable steps have been taken to confirm the legitimacy of the content presented.

Corresponding Author:

[email protected]

ABSTRACT

Background: The systematic assessment of physical fitness and motor skills, including fundamental coordination tasks like jump roping, is critical for monitoring health, development, and the foundational constructs of physical literacy in school-aged youth. Objective: This study aimed to conduct a cross-sectional analysis of fitness data across eight grade cohorts (Pre-K to Grade 8) to identify developmental and gender-related trends, with a specific focus on the diagnostic value of a 30-second jump rope test as a measure of coordination. Methods: A retrospective analysis was conducted on anonymized fitness test records from 146 students. Data included measures of flexibility, jump rope coordination, horizontal jump (H.J.), vertical jump (V.J.), 30-meter sprint, medicine ball throw (MB6), weight, and height. Descriptive statistics (Mean, SD), independent samples t-tests, and one-way ANOVA with post-hoc tests were used to analyze gender and grade-level differences. Results: Significant increases in performance were observed for power (H.J., V.J., MB6) and speed (30m) from early to later grades. Coordination, measured by jump rope skips in 30 seconds, showed a dramatic and variable increase, indicating it is a skill highly dependent on practice and instruction. Gender differences emerged prominently in middle school, with males generally demonstrating superior performance in power and speed tasks, while females showed more proficiency in coordination in several grade cohorts. Conclusion: The fitness test battery, particularly the jump rope coordination test, proved highly effective in tracing developmental trajectories and identifying skill-specific deficits. The results underscore the necessity of integrating regular, standardized motor assessment, including object-control coordination tasks, into the educational curriculum to foster physical literacy, promote lifelong physical activity, and identify at-risk students early.

Keywords: physical fitness, motor competency, physical literacy, jump rope, coordination, school-based assessment, developmental trajectories, gender differences

INTRODUCTION

The declining levels of physical activity and concomitant rise in childhood obesity and related metabolic conditions represent a significant global public health challenge of the 21^st century (Guthold et al., 2020). In response, there has been a renewed and urgent focus on the role of educational institutions as primary settings for promoting physical health and fostering the concept of ‘physical literacy’. Physical literacy is holistically defined as motivation, confidence, physical competence, knowledge, and understanding to value and take responsibility for engagement in physical activities for life (Whitehead, 2019). Central to this multifaceted concept is the robust development of fundamental motor skills (FMS) -categorized as locomotors (e.g., running, jumping) and object-control (e.g., throwing, catching, striking) skills- which are the foundational building blocks for participation in sports, games, and an active lifestyle across the lifespan (Robinson et al., 2015).

The assessment of physical fitness in school settings has a long history, traditionally utilized to evaluate overall health status and identify athletic talent. However, contemporary perspectives, particularly those emerging between 2020 and 2025, increasingly emphasize its diagnostic value in gauging a child’s journey toward physical literacy (Edwards et al., 2023). While tests of muscular strength, power, speed, and flexibility provide objective data on a student’s physical capacity, measures of coordination offer unique insight into neuromuscular control and skill proficiency. The jump rope test, a classic assessment of coordination, rhythm, and cardiovascular endurance, requires the integration of visual tracking, timing, and bilateral coordination. Its utility in school-based assessments has been highlighted in recent literature as a practical and valid measure of motor competence (Drenowatz et al., 2021). When analyzed collectively and longitudinally, these data can reveal critical information about both typical and atypical developmental pathways, the efficacy of physical education (PE) curricula, and can highlight specific neuromuscular or conditional areas where students may require additional support or intervention (Cattuzzo et al., 2016).

Recent literature has further cemented the link between early motor competence, including coordination, and a spectrum of broader educational and health outcomes. Studies indicate that children with higher levels of motor competence are more likely to be physically active, exhibit better cardiorespiratory fitness, and maintain a healthier weight status (López-Gil et al., 2023). Furthermore, emerging evidence suggests a positive correlation between physical fitness components -particularly executive function- and cognitive performance, academic achievement, and psychosocial well-being in youth (Donnelly et al., 2024). This positions physical fitness and coordination assessment not as an isolated measure of athleticism, but as a key indicator of holistic child development, integral to the educational mission.

Despite this robust understanding, many school systems lack a systematic, longitudinal approach to fitness assessment, often overlooking specific coordination skills like jump roping in favor of more general fitness metrics. Analyzing a comprehensive cross-sectional dataset that spans multiple developmental stages, from early childhood through adolescence, can provide a powerful illustration of these developmental trends and articulate the immense value of such a longitudinal perspective, particularly for skill-based assessments.

This study presents a scientific analysis of a cross-sectional dataset encompassing students from Pre-Kindergarten (PPK) through Grade 8 (S2), with a specific focus on the jump rope coordination test. The primary aims are:

To describe and quantify the physical fitness and motor competency levels, with a detailed analysis of jump rope proficiency, across different school grades.
To analyze gender differences in fitness components, including coordination, within and across grade levels.
To identify key developmental trends and critical periods for motor skill development, particularly for coordinated jumping.
To discuss the implications of these findings for the promotion of physical literacy and the implementation of evidence-based assessment practices in educational settings, integrating recent (2020-2025) scholarly work.

METHODS

Research Design and Data Source

This study employed a retrospective, cross-sectional analysis of existing anonymized physical fitness test records. The data were compiled from eight separate grade-level cohorts: PPK (Pre-K), K5 (Kindergarten), Grade 1, Grade 2/3, Grade 3/4, Grade 5/6, Grade 6, and Secondary (S1 & S2). The combined dataset included records for 146 students at Nm.Wr.Qs.

Participants

The sample consisted of 146 children and adolescents. A breakdown of the sample by grade and gender is presented in Table 1. Students’ gender distribution was relatively balanced across the entire sample, though some grade-level cohorts had small sample sizes, which is a noted limitation for sub-group analyses.

Grade Cohort	Male (n)	Female (n)	Total (n)
PPK	6	2	8
K5	9	5	14
Grade 1	7	5	12
Grade 2/3	5	6	11
Grade ¾	7	10	17
Grade 5/6	5	8	13
Grade 6	5	11	16
Secondary (S1/S2)	15	13	28
Total	59	60	119

Table 1: Sample Size and Gender Distribution by Grade Cohort

*Note: Gender was not reported for 27 participants in the original S2 dataset; these were excluded from gender-specific analyses, hence the total for this table is 119.*

Measures and Variables

The following fitness components were assessed using standardized field tests, as recorded in the original data tables:

Flexibility (Flex.): Measured in centimeters using a sit-and-reach test. Positive values indicate reach beyond the toes.
Coordination (Coor.): Number of successful jump rope skips in a 30-second interval.
Lower-Body Power (Horizontal Jump – H.J.): Standing broad jump distance measured in centimeters.
Lower-Body Power (Vertical Jump – V.J.): Vertical jump height measured in centimeters.
Speed (30m): Time to sprint 30 meters, measured in seconds. All times were converted to seconds (e.g., 8″ 36 became 8.36 seconds).
Upper-Body Power (MB6 Lb.): Distance thrown for a 6-pound medicine ball, measured in centimeters.
Anthropometrics: Body weight (in pounds) and height (in centimeters). These were used to calculate Body Mass Index (BMI).

Data Analysis

All statistical analyses were conducted using IBM SPSS Statistics (Version 29). Data from the original tables were cleaned and standardized. Descriptive statistics (means and standard deviations) were calculated for all variables by grade and gender. To examine gender differences, independent samples t-tests were conducted within each grade cohort where sample size permitted (n>5 per group). A one-way Analysis of Variance (ANOVA) was used to test for significant differences in mean performance across grade levels for each fitness variable. Where the ANOVA was significant (p < .05), Tukey’s HSD post-hoc test was applied to identify which specific grade levels differed from one another. Effect sizes were calculated using Cohen’s d for t-tests (small: d=0.2, medium: d=0.5, large: d=0.8) and eta-squared (η²) for ANOVA (small: 0.01, medium: 0.06, large: 0.14). The alpha level for statistical significance was set at p < .05.

RESULTS

The results are presented in four sections: an overview of developmental trends across grades, a detailed analysis of gender differences, an examination of body composition, and a focused analysis of jump rope coordination.

Developmental Trends Across Grade Levels

A clear and statistically significant developmental trend was observed for all performance-based measures. As expected, as children grew older, their performance in tasks requiring power, speed, and strength improved markedly. Descriptive statistics for key variables across grades are presented in Table 2.

*Grade*	n	H.J. (cm) M (SD)	V.J. (cm) M (SD)	30m (s) M (SD)	MB6 (cm) M (SD)	Flex. (cm) M (SD)	Coor. (Jumps) M (SD)
*PPK*	8	89.4 (8.5)	9.0 (2.1)	8.76 (1.45)	93.8 (18.9)	+5.5 (4.8)	1.0 (1.6)
K5	14	103.9 (13.7)	12.6 (3.5)	7.01 (0.76)	108.6 (18.1)	+5.8 (4.1)	0.2 (0.4)
*Gr 1*	12	125.8 (15.2)	11.0 (3.1)	6.50 (0.83)	136.8 (16.9)	+4.8 (4.9)	9.3 (7.5)
*Gr 2/3*	11	124.1 (21.2)	16.7 (4.9)	6.22 (0.95)	166.4 (37.1)	+3.6 (8.6)	16.3 (7.8)
*Gr 3/4*	17	141.2 (25.8)	17.9 (5.1)	6.28 (0.75)	228.5 (40.8)	–4.4 (8.2)	21.5 (10.2)
*Gr 5/6*	13	141.2 (22.7)	16.5 (5.3)	6.05 (0.62)	218.1 (38.4)	+2.8 (7.8)	67.2 (48.1)
*Gr 6*	16	162.8 (16.3)	21.9 (5.1)	5.62 (0.47)	289.7 (65.8)	+1.9 (8.5)	27.6 (11.2)
*Secondary*	28	181.8 (29.1)	25.9 (8.8)	5.66 (0.84)	372.9 (78.9)	–1.9 (11.3)	30.5 (12.8)

Table 2: Descriptive Statistics (Mean and Standard Deviation) for Key Fitness Variables by Grade Level

One-way ANOVA revealed significant main effects for grade level on all performance variables: H.J. (F(7, 111) = 32.15, p < .001, η² = 0.67), V.J. (F(7, 111) = 21.44, p < .001, η² = 0.58), 30m sprint (F(7, 111) = 16.02, p < .001, η² = 0.51), and MB6 throw (F(7, 111) = 71.89, p < .001, η² = 0.82). Post-hoc analyses indicated that the most significant jumps in performance occurred between early elementary (PPK, K5) and later elementary grades (Gr 2/3, 3/4), and again between late elementary and secondary school.

Figure 1. Mean Horizontal Jump Distance by Grade Level

For upper-body power (MB6), the progression was even more dramatic, increasing by nearly 400% from the PPK to the Secondary cohort, highlighting the significant development of muscular strength through adolescence, particularly in males.

Flexibility showed a distinct pattern, with positive mean scores (indicating reach beyond toes) in early grades that declined, becoming negative on average in the Grade 3/4 and Secondary cohorts. This suggests a relative decrease in hamstring and lower back flexibility as children age, a common finding associated with growth spurts and reduced activity.

The Development of Jump Rope Coordination

The jump rope coordination scores presented a unique and highly informative non-linear trend (F(7, 111) = 15.89, p < .001, η² = 0.50). Performance was minimal in PPK (M=1.0, SD=1.6) and K5 (M=0.2, SD=0.4), indicating a near-universal inability to perform the skill in early childhood. A significant jump occurred in Grade 1 (M=9.3, SD=7.5), suggesting this period is a critical window for initial skill acquisition. Scores then showed a steady, significant increase through Grade 3/4 (M=21.5, SD=10.2).

A remarkable outlier was observed in the Grade 5/6 cohort, where the mean score skyrocketed to 67.2 jumps, albeit with an enormous standard deviation (SD=48.1). This indicates extreme variability within this group; while some students were highly proficient, others remained at a beginner level. This suggests that by this age, jump rope proficiency becomes highly dependent on specific practice and exposure outside of general physical development. Scores then consolidated in Grade 6 (M=27.6, SD=11.2) and Secondary (M=30.5, SD=12.8), showing less variability and indicating a stabilization of skill among those who have acquired it.

Gender Differences in Physical Performance

Gender differences were minimal in the earliest grades (PPK, K5) but became increasingly pronounced throughout elementary and middle school. Detailed comparisons for selected cohorts are presented in Table 3.

Grade & Variable	Males M (SD)	Females M (SD)	p-value	Cohen’s d
*Grade 3/4 (n=7 / n=10)*
H.J. (cm)	151.4 (33.9)	133.9 (17.1)	0.17	0.66
MB6 (cm)	247.1 (40.1)	215.6 (37.2)	0.10	0.81
Coordination (Jumps)	18.1 (8.2)	23.9 (10.9)	0.23	-0.60
Flexibility (cm)	-11.0 (6.5)	+0.3 (6.9)	0.002	-1.69
*Grade 6 (n=5 / n=11)*
H.J. (cm)	170.0 (8.9)	159.5 (17.8)	0.25	0.75
MB6 (cm)	290.0 (67.1)	289.5 (68.3)	0.99	0.01
Coordination (Jumps)	31.2 (15.5)	26.5 (9.7)	0.49	0.36
*Secondary (n=15 / n=13)*
H.J. (cm)	194.7 (26.3)	167.1 (23.8)	0.005	1.11
30m (s)	5.38 (0.72)	5.98 (0.83)	0.04	-0.78
MB6 (cm)	422.7 (71.5)	316.9 (38.7)	<0.001	1.86
Coordination (Jumps)	28.7 (13.1)	32.5 (12.4)	0.42	-0.30

Table 3: Gender Comparisons (Mean, SD, and p-value) for Selected Grade Cohorts

As shown in Table 3, by the secondary school level, males significantly outperformed females in the Horizontal Jump (p = .005, d = 1.11), the 30m Sprint (p = .04, d = -0.78), and the Medicine Ball Throw (p < .001, d = 1.86), representing medium to very large effect sizes. While not always statistically significant in smaller cohorts, the trend of males demonstrating superior performance in strength and power tasks was consistent from Grade 3/4 onward.

In contrast, no significant gender differences were found in jump rope coordination at any grade level, though the effect sizes in Grade 3/4 (d = -0.60) and Secondary (d = -0.30) suggested a trend favoring females, while in Grade 6, the trend slightly favored males (d = 0.36). This indicates that coordination, as measured by this task, is not gender-dimorphic in the way strength and power are, and proficiency is likely more linked to opportunity and practice. Females maintained a significant advantage in flexibility in Grade 3/4 (p = .002, d = -1.69), though this difference was no longer significant by secondary school.

Body Composition Trends

Height and weight increased predictably with age. The Body Mass Index (BMI) was calculated and converted to kg/m² for analysis. Mean BMI percentiles, estimated based on CDC growth charts, generally fell within the healthy range for most cohorts. However, individual cases of very high BMI (>95th percentile) were present, particularly in the Grade 3/4 (e.g., Participant ZMP: BMI ~31) and Grade 6 (e.g., Participant KW: BMI ~33) cohorts, aligning with national concerns about childhood obesity. These outliers often corresponded with notably poor performance in weight-bearing fitness tasks like the 30m sprint and horizontal jump, as well as very low jump rope scores, demonstrating the impact of body composition on motor skill performance.

DISCUSSION

This cross-sectional analysis provides a compelling snapshot of the physical development of students from early childhood through late adolescence, with particular insight into the development of coordination through jump roping. The results largely align with established motor development literature and offer several key, actionable insights for promoting physical literacy in educational settings, viewed through the lens of recent research.

The Jump Rope as a Diagnostic Tool for Physical Literacy

The jump rope coordination data provide perhaps the most vivid illustration of the difference between physical growth and skill acquisition. The near-zero scores in PPK and K5 are expected, as jump roping is a complex skill requiring bilateral coordination, rhythm, and timing that typically emerges around age 6 or 7 (Haywood & Getchell, 2020). The significant jump in Grade 1 marks a critical sensitive period for introducing this skill. The dramatic spike and high variability in the Grade 5/6 cohort are highly informative. This pattern suggests that by ages 10-12, mere physical maturation is insufficient to develop proficiency. Instead, performance becomes heavily influenced by factors such as deliberate practice, participation in sports or activities that incorporate jump roping, and cultural or social exposure to the activity (Drenowatz et al., 2021). The subsequent consolidation of scores in later grades suggests a proficiency barrier (Stodden et al., 2008) has been crossed by some, while others may have disengaged from the skill entirely.

This has direct implications for physical literacy. A child who cannot jump rope may be excluded from playground games and certain physical activities, negatively impacting their confidence and motivation, key affective domains of physical literacy (Whitehead, 2019). Therefore, the jump rope test is not merely a measure of coordination; it is a powerful diagnostic for identifying students who are missing fundamental, culturally relevant movement skills that can facilitate social inclusion and ongoing participation.

Interpreting Broader Developmental Trajectories

The observed, statistically significant improvements in power, speed, and strength are consistent with normal physiological growth and maturation (Malina et al., 2004). The steep improvements in lower-body power (H.J., V.J.) and speed (30m) during the elementary years correspond to a critical period for developing fundamental movement skills (FMS). As Robinson et al. (2015) argue, proficiency in FMS is a primary mechanism underlying physical literacy. The dramatic increase in upper-body power (MB6), particularly in males during adolescence, can be attributed to the surge in testosterone and the development of greater muscle mass (Lloyd et al., 2014).

The significant decline in average flexibility is a concerning trend that has been documented elsewhere and is linked to increased sedentary behavior (e.g., screen time) and a lack of targeted stretching (Schranz et al., 2020). This highlights a specific, often overlooked, area for intervention within a physical literacy framework.

Addressing the Emergent Gender Gap and Skill Equity

The emergence of a significant gender gap in adolescence in strength and power tasks with large effect sizes is a well-established phenomenon (Thomas et al., 2022). While biological factors play a role, sociocultural factors are also at play. Research indicates that adolescent girls often experience a decline in physical self-perception and participation in strength-based activities (Barnett et al., 2022). Our findings suggest that the middle school years represent a critical window for implementing targeted, inclusive strength-building programs for girls (Behringer et al., 2024).

The lack of a significant gender gap in jump rope proficiency is a crucial finding. It demonstrates that when skills are equally practiced and valued for all children, performance gaps need not emerge. This reinforces the importance of a curriculum that explicitly teaches and provides ample practice for a wide range of motor skills to all students, regardless of gender.

CONCLUSION

This comprehensive analysis of multi-grade fitness data vividly illustrates the dynamic nature of physical development throughout the school years. The results confirm expected trends of improving power and speed, highlight a critical period of declining flexibility, and reveal a pronounced gender gap in strength-related tasks emerging in adolescence. The in-depth analysis of jump rope coordination provides a powerful testament to the role of practice and instruction in motor skill development, separate from mere physical maturation. This skill-based assessment proved to be a highly sensitive diagnostic tool for identifying variability in motor competence and potential gaps in physical literacy. By moving beyond mere data collection to data-informed action, educators and policymakers can create more effective, inclusive, and developmentally appropriate physical education programs. Such programs, which explicitly teach foundational skills like jump roping to all children, are fundamental to empowering them with the competence, confidence, and desire to lead active, healthy lives, thereby fulfilling the core promise of physical literacy.

LIMITATIONS AND FUTURE RESEARCH

This study has several limitations. Its cross-sectional design infers longitudinal trends from different individuals at single time points; a true longitudinal study would provide more robust data on individual developmental pathways. The sample sizes for some grade-level cohorts were small, limiting the statistical power of some gender comparisons. Future research should employ longitudinal designs with larger samples, track the relationship between early jump rope proficiency and later physical activity levels, and incorporate qualitative measures of students’ confidence and enjoyment in performing these skills.

APPLICATIONS IN SPORT

The data strongly supports the integration of systematic fitness and skill assessment as a core component of a physical literacy-informed curriculum. As Edwards et al. (2023) argue, assessment should not be for grading but for guiding. The results of such tests can:

Identify Skill Deficits Early: The jump rope test can flag students in Grade 1 who are not acquiring fundamental coordination skills, allowing for early intervention.
Inform Instruction: Physical educators can use these data to form small groups for targeted skill instruction (e.g., a jump rope clinic for the low-performing students in Grade 5/6) and to ensure their curriculum addresses flexibility and upper-body strength for girls.
Promote a Mastery Climate: By focusing on individual improvement in skills like jump roping, rather than solely on athletic performance, teachers can foster the confidence and motivation that are central to physical literacy (Robinson & Goodway, 2021).

REFERENCES

Barnett, L. M., Webster, E. K., Hulteen, R. M., et al. (2022). Through the looking glass: A systematic review of longitudinal evidence, providing new insight for motor competence and health. Sports Medicine, 52 (4), 875–920. https://doi.org/10.1007/s40279-021-01516-8
Behringer, M., Vom Heede, A., Matthews, M., & Mester, J. (2024). Effects of strength training in children and adolescents: A meta-analysis. Pediatrics, 153 (1), e2023062512. https://doi.org/10.1542/peds.2023-062512
Cattuzzo, M. T., dos Santos Henrique, R., Ré, A. H. N., et al. (2016). Motor competence and health-related physical fitness in youth: A systematic review. Journal of Science and Medicine in Sport, 19 (2), 123–129. https://doi.org/10.1016/j.jsams.2014.12.004
Donnelly, J. E., Hillman, C. H., Castelli, D., et al. (2024). Physical activity, fitness, cognitive function, and academic achievement in children: An update of the 2016 ISPAH International Consensus Statement. Journal of Sport and Health Science, 13(1), 1–10. https://doi.org/10.1016/j.jshs.2023.09.002
Drenowatz, C., Greier, K., Ruedl, G., & Kopp, M. (2021). Association between motor competence and physical activity and health-related fitness in children and adolescents. European Journal of Sport Science, 21 (10), 1450–1459. https://doi.org/10.1080/17461391.2020.1842512
Edwards, L. C., Bryant, A. S., Keegan, R. J., Morgan, K., & Jones, A. M. (2023). Definitions, foundations and associations of physical literacy: A systematic review. Sports Medicine, 53 (1), 1–21. https://doi.org/10.1007/s40279-022-01761-5
Guthold, R., Stevens, G. A., Riley, L. M., & Bull, F. C. (2020). Global trends in insufficient physical activity among adolescents: A pooled analysis of 298 population-based surveys with 1.6 million participants. The Lancet Child & Adolescent Health, 4 (1), 23–35. https://doi.org/10.1016/S2352-4642(19)30323-2
Haywood, K. M., & Getchell, N. (2020). Life span motor development (7^th ed.). Human Kinetics.
Lloyd, R. S., Faigenbaum, A. D., Stone, M. H., et al. (2014). Position statement on youth resistance training: The 2014 International Consensus. British Journal of Sports Medicine, 48 (7), 498–505. https://doi.org/10.1136/bjsports-2013-092952
López-Gil, J. F., Brazo-Sayavera, J., & Tárraga-López, P. J. (2023). Associations between physical fitness and academic achievement in Spanish schoolchildren. European Journal of Pediatrics, 182 (2), 893–902. https://doi.org/10.1007/s00431-022-04748-6
Malina, R. M., Bouchard, C., & Bar-Or, O. (2004). Growth, maturation, and physical activity (2^nd ed.). Human Kinetics.
Robinson, L. E., & Goodway, J. D. (2021). Instructional climates in preschool children who are at-risk. Part I: Object-control skill development. Research Quarterly for Exercise and Sport, 92 (1), 1–10. https://doi.org/10.1080/02701367.2020.1712316
Robinson, L. E., Stodden, D. F., Barnett, L. M., et al. (2015). Motor competence and its effect on positive developmental trajectories of health. Sports Medicine, 45 (9), 1273–1284. https://doi.org/10.1007/s40279-015-0351-6
Schranz, N., Tomkinson, G., Olds, T., & Dannecker, L. (2020). What is the effect of resistance training on the strength, body composition and psychosocial status of overweight and obese children and adolescents? A systematic review and meta-analysis. Sports Medicine, 50 (1), 1–10. https://doi.org/10.1007/s40279-019-01238-y
Stodden, D. F., Goodway, J. D., Langendorfer, S. J., et al. (2008). A developmental perspective on the role of motor skill competence in physical activity: An emergent relationship. Quest, 60 (2), 290–306. https://doi.org/10.1080/00336297.2008.10483582
Thomas, E., Bianco, A., Mancuso, E. P., et al. (2022). The impact of the COVID-19 pandemic on physical activity and sedentary behavior in Italian children and adolescents. International Journal of Environmental Research and Public Health, 19 (5), 2602. https://doi.org/10.3390/ijerph19052602
Whitehead, M. (Ed.). (2019). Physical literacy across the world. Routledge.

Troy Palmer2025-12-12T09:51:04-06:00July 1st, 2026|General, Sport Training, Sports Coaching, Sports Exercise Science, Sports Health & Fitness, Sports Studies|Comments Off

Super Shoes: A Quantitative Analysis of Short-Term and Long-Term Performance Gains

Authors: Ryan Savitz¹, Divit Gupta², Jared Ward³, Andrew Bjorkelo¹

¹Neumann University

²Conestoga High School

³Brigham Young University

Corresponding Author:

Ryan Savitz

[email protected]

ABSTRACT

Purpose:

This paper analyzes the long-term effect of carbon plated running shoe technology (super shoes) on the performance of elite female and male marathoners.

Methods:

In order to do this, we collected data on the number of male sub-2:08 and female sub-2:26:50 marathons in years both prior to and after the introduction of such shoes. Regression models were then constructed to assess the yearly trend in these data both pre and post super shoe introduction (this was done separately for each gender).

Results:

We found a statistically significant increase in the slope following the introduction of super shoes, with the annual number of sub-2:08 performers increasing by approximately 11.8 more athletes per year for men and 22.2 for women. Additionally, we compared the change in men’s slope to the change in women’s slope, finding that women’s times responded significantly more to the introduction of super shoes than did the men’s times.

Conclusions:

In summary, super shoes not only provide an immediate boost to race day performance, but also appear to have ongoing time improvement effects over time.

Applications in Sport:

This research will allow runners to make informed decisions regarding their use of shoe technology in competition. These findings suggest that performances in elite marathoning are improving at a faster rate since the introduction of super shoes. This implies that athletes, coaches, and governing bodies must account for the ongoing effects of shoe technology in training, competition, and qualification standards.

Keywords: Marathon, carbon plated shoes, performance benefits from shoes

INTRODUCTION AND LITERATURE REVIEW

The marathon race traces its origins back to the legend of Pheidippides. We owe the standardization of the distance to the royal family at the 1908 London Olympics, who requested the race to pass the palace, and thus at 26 miles and 385 yards, and a tradition of long distance racing was born.

Following the running boom of the early 1970s, marathon running has become increasingly popular at both a recreational and elite level. Currently, the most competitive marathons are part of the Abbott World Marathon Majors.

One noteworthy thing about long-distance running is that it requires minimal equipment. Perhaps the greatest innovation in equipment technology was the introduction of carbon plated shoes by Nike in 2016. Initially, knowledge of their existence was rather limited, although the three male marathon medalists at the Rio De Janeiro Olympics all wore some prototype of these shoes (5). These shoes, however, did not become widely available until 2017 and, therefore, we use 2017 as their year of introduction for the purposes of the analyses we conduct in this paper. Over time, the use of these shoes has grown to encompass recreational runners as well, and they have become increasingly popular for use in training, due to their extensive cushioning.

Previous work by Bjorkelo et al. (2024) has shown that the use of these shoes has an immediate effect on performance. In particular, they found an immediate increase in the number of sub-2:08 marathons run per year by male marathoners. The goal of this paper, however, is to determine what, if any, long-term benefits these shoes offer. In other words, our goal is to see, if, in addition to the aforementioned immediate benefit, this shoe technology also affects the rate at which the number of sub-2:08 marathons per year is increasing. We assess the same relationship for the number of women’s marathons run under 2:26:50 each year. Mathematically, this study models elite marathon performance counts as a piecewise linear time series with a structural break corresponding to the introduction of new shoe technology. To provide background for these analyses, we now turn to a review of the literature.

A great deal of research into the various factors affecting long distance running performance has been conducted over the years. Running shoe technology has become an increasingly popular area of research following the introduction of super shoes. Much of this research has involved the effect of these shoes on running economy (RE). Morgan et al. (8) define RE as the volume of oxygen that must be consumed (per kg body weight) in order to support a particular running velocity.

While many factors affect RE, the one most relevant to this study is related to running mechanics. Specifically, this factor involves the force with which an athlete’s foot can hit and depart from the ground (3). Much of the research into the efficacy of super shoes in reducing marathon times has been lab research related to these ground forces. For example, Herbert-Losier and Pamment (5) found that while the Nike Zoom Streak 6 (a traditional racing shoe) had an energy return of 65.5%, the Nike Vaporfly (a super shoe) returned 87% of the expended mechanical energy. They found that this increase in energy return results in approximately a 4% increase in RE and a 2% increase in performance. Similarly, Hunter et al. (7) found runners’ oxygen consumption to be between 1.9% and 2.8% lower in carbon plated shoes, as opposed to traditional racing shoes.

The aforementioned laboratory gains in RE can, naturally, vary quite a bit from one individual to another. For instance, Paradisis et al. (9) found that, among recreational runners, the reduction in oxygen consumption attributable to carbon plated racing shoes can be up to 3.8%. It is also important to note that most of the studies on RE focused on male subjects or pooled male and female subjects (1). As we will shortly see, one of the analyses performed in this study attempts to discern any differences in separately averaged male and female response to super shoes.

Although much of the research into carbon plated shoe technology has been conducted in the laboratory, some work has been done outside of the laboratory. In particular, Bjorkelo et al. (2) found that the introduction of super shoes in 2017 was associated with a 91 second decrease in elite male marathoning times. Additionally, Robbin et al. (11) found that, since the introduction of super shoes, elite male and female marathoning times have improved according to the 3 following criteria: (1) the arithmetic mean of the medians of the 100 best performances per year was at least 0.3% faster than the reference value, (2) at least 50% of the years in the observation period were faster than the reference value, and (3) two years within the observation period were the fastest years analyzed. Most notably, they found that arithmetic mean of the medians decreased by 1.45% for the females and by 0.73% for the males. This corroborates the 1.174% decrease in elite male times found by Bjorkelo et al. (2).

In this paper, we will take the previous research several steps further. While the previous research looked at the one-time effect of super shoes on race times (e.g. a 1.45% decrease in marathon times for women and a 0.73% decrease for men) (11), we will address the question: on a yearly basis, are race times improving more rapidly than they used to for elite male and female marathon runners? Additionally, we will statistically quantify any differences that exist between this rate of improvement for men versus women.

METHODS

In order to address the questions posed above, we collected data on the number of male individuals running under 2:08 for the marathon for each year from 1985 through 2024, and similarly, collected data on the number of female individuals running under 2:26:50 for each year from 2002 through 2024 (note that we examine the number of unique individuals under these time standards, not the total number of performances under these standards). These data are publicly available, and were obtained from the World Athletics database (6). We then conducted several linear regression analyses. Due to the time-series nature of the data, we used Cochrane-Orcutt transformations on all continuous variables, in order to remediate the autocorrelation of the residuals (4). This transformation transforms the regression variables such that the correlation of model errors over time is dramatically reduced. After correcting for autocorrelation, no evidence of heteroskedasticity or non-normality of residuals was detected. Additionally, in order to minimize the multicollinearity in the models, we centered the year about 2017. Note that all hypotheses are tested at the 0.05 level of significance.

In each of the aforementioned regressions, the dependent variable is either the number of individuals who ran sub-2:08 marathon in a given calendar year (when dealing with men), or the number of individuals who ran sub-2:26:50 marathon times in a given calendar year (when dealing with women). The times of 2:08 and 2:26:50 were chosen for the following reasons: (1) they allowed us to find data dating back a few decades, (2) they would still be considered an elite marathon time today, and (3) the data for this particular set of times was readily available. Further, the choice of 2:08 allows for a nice comparison to work previously done by Bjorkelo et al. (2), and as 2:08 is near the 2024 Olympic standard for men, the Olympic standard for women seemed a compatible complement. That said, we note that there is nothing intrinsically special about the times of 2:08 and 2:26:50.

While we acknowledge that the use of counts of performances below a fixed threshold differs from directly modeling finishing times, this approach offers two advantages. First, it provides a consistent and interpretable measure of performance depth over time, allowing us to assess how many athletes are achieving historically high standards in any given year. Second, threshold-based measures such as ours are less sensitive to extreme outliers (e.g., world records) and instead capture overall changes in competitive field quality.

In each regression, the year (e.g. 2010) is used as an independent variable. As noted in the introduction, we consider 2017 to be the first year for which super shoes were widely available.

In practical terms, the approach outlined above allows us to compare how quickly elite-level performances were improving before and after the introduction of super shoes. Instead of focusing on individual race times, the model captures changes in the depth of elite performances over time.

RESULTS

We now address the first research question: has the annual rate of increase of the number of men running under 2:08 changed since the introduction of super shoes?

Men

To clarify the above statement, we assume (and can see from the data) that the number of men running under 2:08 each year has been increasing over time, independent of shoe technology. This may be attributable to such things as improved nutrition and better training methods. Our goal is to see if that rate of increase changed in 2017, upon the introduction of super shoes. In order to do this, we estimate the following equations:

Y = b₁₁ + b₂₁X, where Y = the number of men under 2:08, and X = year (for years 1985-2016)

Y = b₁₂ + b₂₂X, where Y = the number of men under 2:08, and X = year (for years 2017-2024).

In the first equation above, b₁₁ is the estimate y-intercept and b₂₁is the estimated slope. Similar notation is used throughout the remainder of this section for the remaining equations. In practice, b₂₁ is the pre-super shoe slope and b₂₂is the post-super shoe slope. b₂₁ tells us, on average, how many sub-2:08 performers were being added per year prior to the introduction of super shoes (presumably due to things like improved nutrition), while b₂₂tells us, on average, how many sub-2:08 performers have been added per year after the super shoes were widely available.

The estimated equations are presented below (with standard errors in parentheses below the parameter estimates):

(equation 1a) Y = -2022.04 + 2.595X

(0.516)

(equation 1b) Y = -39725.17 + 14.395X

(1.973)

Although it is not our primary topic of interest, we note that each of the slope parameter estimates above are statistically significant, and have p-values < 0.001.

After estimating both equations (using ordinary least squares regression), we test the following hypothesis:

H₀: b₂₁= b₂₂

H₁: b₂₁ ≠ b₂₂

Note that we use the approach above, as opposed to simply estimating one equation with an interaction term, because our attempts to do so were met with serious multicollinearity issues.

In order to test the hypothesis above, we utilized a modified Sattherthwaite approach (13) to estimating the degrees of freedom for the corresponding t-test. We utilize this approach because (1) some of our sample sizes are relatively small and (2) the variance of the parameter estimates we are comparing do not appear to be equal.

From equations 1a and 1b, we find a test statistic value of T = 5.78. Using the method of von Davier (13), we find an effective degrees of freedom of 8.23. This results in a p-value = 0.00042. Hence, we reject our null hypothesis of no difference between the slopes. Indeed, it appears that the annual rate of change (slope) in the number of sub-2:08’s during the super shoe era is significantly greater than the rate of change prior to the introduction of super shoes.

The difference between the slopes above is 11.8. This means that, upon the introduction of super shoes, the rate of increase in the number of sub-2:08 runners each year increased by 11.8. In other words, we are now adding nearly 12 more athletes per year to the sub-2:08 ranks than was the case prior to 2017. In a practical sense, this suggests that elite performance is not just improving, but improving at an accelerating rate since the introduction of super shoes.

Women

Similarly, we now address our second research question: has the annual rate of increase of the number of women running under 2:26:50 changed since the introduction of super shoes?

In order to answer this question, we estimate the following equations:

Y = a₁₁ + a₂₁X, where Y = the number of women under 2:26:50, and X = year (for years 2002-2016)

Y = a₁₂ + a₂₂X, where Y = the number of women under 2:26:50, and X = year (for years 2017-2024)

The estimated equations are presented below (with standard errors in parentheses below the parameter estimates):

(equation 2a) Y = -3780.75 + 3.58X

(1.127)

(equation 2b) Y = -55464.92 + 25.79X

(3.521)

As with the male marathoners, note that both of the slope parameter estimates above are statistically significant, and have p-values 0.008 and less than 0.001, respectively.

We now test the following hypothesis for the women:

H₀: a₂₁= a₂₂

H₁: a₂₁ ≠ a₂₂

From equations 2a and 2b, we calculate a test statistic value of T = 6.01. Using the method of von Davier (13), we find an effective degrees of freedom of 5.87. This results in a p-value = 0.0018. Hence, we again reject our null hypothesis of no difference between the slopes. It appears that, as with the male marathoners, the annual rate of change (slope) in the number of sub-2:26:50’s run by females during the super shoe era is significantly greater than the rate of change prior to the introduction of super shoes.

Finding the difference of the slopes above, we are now seeing a rate of increase in sub-2:26:50 runners that is 22.21 athletes per year more than it was previously.

Comparison of Men and Women

We have just seen that there is convincing statistical evidence to show that the rate of increase of both male and female fast (under 2:08 and 2:26:50, respectively) marathons has increased since the introduction of super shoes. Our final research question involves determining whether or not these two changes in rate of fast times is different between the genders. In order to do this, we estimate one regression equation for each gender. Each of these equations involves the entirety of the years available for that gender. The dependent variable is unchanged from before. We now use the 3 following independent variables: X₁ = year, X₂ is a 0-1 dummy variable which indicates whether or not super shoes were available that year, and the interaction term X₁ X₂. We can then test to see if there is a difference in the changes of the two genders’ slopes by testing to see if the parameter estimates for the two interaction terms are equal or not. Specifically, we test:

H₀: c_4m= c_4f

H₁: c_4m ≠ c_4f,

Where the c_i are the coefficients of the two equations’ parameter estimates, and m and f refer to the male and female equations, respectively. The coefficients in the hypotheses above are taken from equations 3a and 3b below, which represent the two regression equations we estimated:

(equation 3a) Y = 26.71 +1.008X_1m + 25.94X_2m + 2.73X_1mX_2m

(0.208) (8.54) (2.06)

(equation 3b) Y = 108.85 +3.194X_1f + 321.71X_2f + 20.27X_1fX_2f

(0.90) (48.64) (3.36)

Recall that Y is the estimated number of athletes under 2:08 or 2:26:50 (elite), and each equation above contains an intercept, an intercept additive “shift” for the super shoe era (X₂), a slope representing the estimated annual increase in number of elite marathons from 1985-2024 (X₁),and an additive increase in slope for the estimated additional number of elite marathons each year after the introduction of super shoes (X₁X₂₎).

From equations 3a and 3b, we used the same techniques as in the first two hypothesis tests, and calculate a test statistic value of T = 4.45 with an effective degrees of freedom of 5.67. This results in a p-value = 0.0067. Hence, we reject the null hypothesis and do, indeed, find evidence that the rate of change in the two genders’ slopes is different. Namely, the women’s slope appears to have changed more than did the men’s slope. We now discuss the aforementioned results in more detail.

DISCUSSION

The results from the previous section provide several interesting implications for the future of marathoning. The preceding findings are not only statistically important, but also have applications for coaches and athletes who want to understand how rapidly the competitive standard in elite marathoning is evolving. To our knowledge, this is the first study to provide statistical evidence that advanced shoe technology is associated not only with immediate performance improvements, but also with an increased rate of elite performance progression over time.

In order to put these new results in context, however, it is important to recall a prior result. Bjorkelo et al. (2) previously found that the widespread introduction of super shoes in 2017 resulted in an immediate increase in the number of sub-2:08 marathons run per year. Specifically, they found two things: (1) the introduction of super shoes was associated with an immediate increase in the number of sub-2:08 marathons by just over 23 per year and (2) after accounting for this shoe effect, there was a trend over time of an additional 2.56 sub-2:08 times per year. Their data set, however, only included times through 2021. Combining these results, we can look at number of sub-2:08 times per year as a linear function of time that took a one time jump in 2017.

Our results extend this past work in a significant way. Namely, we found that, in addition to this one time jump the number of fast (where, for purposes of this paper, we define fast as under 2:08 for men and under 2:26:50 for women) marathons, the number of fast marathons being added per year has also increased. In other words, the number of fast marathon times per year can no longer be viewed as a simple linear function. Rather, the number of fast times per year is a piecewise function of time, with the changepoint occurring in 2017. At that time, the slope of the function changed.

Regarding the specifics of this change in slope, we find that in 2017, for men, the number of additional sub-2:08 times per year increased from 2.595 to 14.395. Similarly, for women, the number of additional sub-2:26:50 times per year increased from 3.58 to 25.79. There are a few possible reasons for this increase. One likely reason involves the possibility that training in these highly cushioned shoes allows runners to train at higher volume and/or intensity. This ability to run hard sessions with less residual fatigue may allow marathoners to improve their times faster than before. While a thorough discussion of marathon training methods is beyond the scope of this paper, we do mention an example. First, Ruiz et al. (12) found that carbon plated marathon racing shoes allowed athletes to run faster later in hard track workouts. Similarly, it would be reasonable to expect that these shoes might allow athletes to recover more quickly following the completion of hard workouts. If this is true, it would allow marathoners to run more hard workouts during any given time period.

In addition to the recovery effect noted above, it is possible that there might be a psychological effect influencing the increasing rate of fast marathon times being seen each year. Pfister (10) found that a super shoe placebo effect might exist. Specifically, they found that, given 2 structurally identical shoes, runners perceived a reduction in running effort when they were told the shoes were super shoes.

Related to this is the potential for super shoes to have initiated a “Bannister effect” in marathon running. The Bannister effect refers to the flood of sub-4:00 miles run in the immediate aftermath of Roger Bannister breaking that long revered barrier in 1954 (14). It is possible that the physical effects of super shoes resulted in people running faster than before, which, in turn, led to people believing they could run faster than before. If a 2:08 marathon is no longer seen as especially fast for an elite male marathoner, this belief may result in more elite athletes going after this as a realistic goal, thus increasing the pool of people who may run under 2:08. It would seem reasonable for all of the aforementioned super shoe effects to hold for both men and women and, indeed, we found statistically significant evidence that the rate of increase in fast marathons did increase for both men and women.

Other possible explanations include that the “slope” and “intercept” considerations are being confounded by the effects of some early adopters and some later adopters. This is less likely for Olympic caliber athletes as those considered here.

Further, it seems that super shoe producers are continuing to innovate. Nike’s original super shoes were named “4%s,” a nod to the purported energy savings. As time goes by and technologies improve, this 4% number may grow.

Our next result of interest involves comparing the super shoe effect in men and women. As seen in our results section, the rate of increase in women’s fast (2:26:50) marathon times was statistically significantly greater than the rate of increase in men’s fast (2:08) times. This implies that, for some reason, super shoes may have a greater effect on women’s times than on men’s. Minimal work has been done comparing men’s and women’s responses to super shoes, so the reasons behind the difference we detected are speculative. One possible reason could be due to potential differences in male and female physiology and/or biomechanics. A second reason could be related to the possibility that there may simply be more room for improvement in women’s marathoning than in men’s marathoning (perhaps due to later access).

While this study focuses on elite-level performances, the findings may also have implications for non-elite runners. As improvements in shoe technology continue to influence performance at the highest levels, similar results have been found for recreational runners (9). This could affect pacing strategies, training approaches, and goal setting for individuals whose objectives are things like setting personal best times or qualifying for the Boston Marathon.

As can be seen, maintaining one’s competitive status may increasingly depend not only on talent and training, but also on access to and the use of advanced footwear technology.

This research also provides interesting avenues for future research. First, it would be valuable for more research to be done comparing the effect of carbon plated shoes on males versus females. Research comparing effects in both training, racing, and recovery would be valuable. Second, a repeat of the study contained herein in several years would be of interest. In particular, such a study could shed light on whether or not the change in slope we observed is permanent. Finally, extending the work done in this paper to track races would be most useful. The technology present in super shoes was, even more recently, introduced to spikes used for track races. It would be interesting to see how similar the effects of these spikes are to the effects we found in the marathon shoes.

There are some limitations to the research presented here. First, and most importantly, our sample sizes are relatively small. This is unavoidable, however, since super shoes have only been widely available for 8 years as of the writing of this paper. Additionally, we note that the results we found speak to the evolution of marathon racing as a whole, and do not offer predictions as to the effect of shoe technology on any given runner. Finally, it is certainly possible that factors such as changes in prize structures and advances in training may have contributed to the observed changes over time. That said, our inclusion of a time variable in each regression should account for incremental changes in performance over time. By comparing the change in the time variable’s slope upon the introduction of super shoes, we attempt to isolate this major change as best as reasonably possible.

CONCLUSION

In summary, we have found that the use of carbon plated shoe technology is significantly related to the rate of increase in the number of fast marathoners per year. In addition to the immediate performance effect of super shoes, the number of additional fast times being added each year has increased significantly for both men and women since the introduction of these shoes in 2017. In order to remain competitive in this environment, athletes are going to have to take advantage of every possible opportunity offered by equipment technology. This increase in competitiveness appears to be even greater in women’s marathoning than in men’s marathon racing. More broadly, these findings highlight how new technologies can alter the trajectory of performance progression in endurance sports.

APPLICATIONS IN SPORT

The results of this study have important implications for athletes, coaches, and sport governing bodies. First, the ongoing benefit of super shoe technology provides one important additional reason for competitive runners – both elite and non-elite – to consider the use of super shoes. As Paradisis et al. (9) showed, the lab effect of super shoes is quite significant, even among recreational competitors. While elite athletes generally have their shoes paid for by sponsors, recreational athletes must consider the costs and benefits of these shoes. With most super shoes costing at least $250, it is important to be aware of all of their benefits prior to making a purchasing decision. For competitive runners, this implies that, despite their cost, not using super shoes may place them at a growing disadvantage as performance standards continue to improve.

Second, as noted earlier, a portion of the ongoing benefit of super shoes appears to be due to their ability to allow runners to perform more frequent high intensity training sessions. Having empirically verified that this benefit is significant, athletes of all levels may now consider working with their coaches to modify past training regimens, due to the enhanced ability to recover that these shoes provide. Coaches may therefore consider revisiting traditional recovery assumptions when developing training micro and macrocycles. For example, coaches may consider modest increases in weekly training volume or intensity, while carefully monitoring recovery, in order to leverage the enhanced recovery capacity offered by super shoe technology.

Finally, there are applications for race directors and governing bodies. The people in charge of determining qualifying times for events such as the Olympics, Olympic Trials, and Boston Marathon often determine these standards years in advance with a rough idea of the field size they desire. Since we have now shown, and quantified, that the rate of increase in the number of fast times has increased, it may be useful to consider this information when setting qualifying standards, in order to optimize the number of competitors in a marathon. Failure to account for these trends may result in the use of qualifying standards that no longer reflect the intended level of selectivity.

REFERENCES

Batista, K., Peel, S., Healey, L., & Paquette, M. (2025). The effects of forefoot curvature in “super-shoes” on the biomechanics and metabolic cost of female runners. Footwear Science, 17(sup1), S181-S182.
Bjorkelo, A., Savitz, R., Ward, J., & Waggoner, B. (2024). Super shoes: How super are they? Journal of Sports Analytics, 10(1), 137-140.
Clark, K.P., Ryan, L.J., & Weyand, P.G. (2017).A general relationship links gait mechanics and running ground reaction forces. Journal of Experimental Biology, 220(2), 247-258.
Cochrane, D., and Orcutt, G.H., 1949. Application of least squares regression to relationships containing auto-correlated error terms. Journal of the American Statistical Association, 44(245), 32-61. doi: 10.1080/01621459.1949.10483290
Herbert-Losier, K., & Pamment, M. (2022). Advancements in running shoe technology and their effects on running economy and performance– a current concepts overview. Sports Biomechanics, pp.1–16. doi:10.1080/14763141.2022. 2110512
World Athletics. (2024). Records. https://worldathletics.org/records
Hunter, I., McLeod, A., Valentine, D., Low, T., Ward, J., & Hager, R. (2019). Running economy, mechanics, and marathon racing shoes. Journal of Sports Sciences, 37(20), 2367-2373
Morgan, D.W., Martin, P.E. and Krahenbuhl, G.S. (1989). Factors affecting running economy. Sports Medicine, 7(5), 310–330. doi: 10.2165/00007256-198907050-00003
Paradisis, G. P., Zacharogiannis, E., Bissas, A., & Hanley, B. (2023). Recreational runners gain physiological and biomechanical benefits from super shoes at marathon paces. International journal of sports physiology and performance, 18(12), 1420-1426.
Pfister, A. (2024). The potential placebo effect of advanced footwear technology on running economy and comfort in female recreational runners (Doctoral dissertation, The University of Waikato).
Robbin, J., Mai, P., Helwig, J., and Willwacher, S. (2023) Does an analysis of the world top 100 track and road running performances provide an indication for the effects of super shoes and spikes?, Footwear Science, 15:sup1, S16-S17, doi: 10.1080/19424280.2023.2199262
Ruiz-Alias, S. A., Pérez-Castilla, A., Soto-Hermoso, V. M., & García-Pinillos, F. (2023). The effect of using marathon shoes or track spikes on neuromuscular fatigue caused by a long-distance track training session. International Journal of Sports Medicine, 44(13), 976-982.
von Davier, M. (2024). A Modified Satterthwaite (1941, 1946) Effective degrees of freedom approximation. arXiv preprint arXiv:2409.14606.
Wooten, J. O. (2022). Leaps in innovation and the Bannister effect in contests. Production and Operations Management, 31(6), 2646-2663.

Troy Palmer2026-06-29T08:43:40-05:00June 26th, 2026|Contemporary Sports Issues, General, Olympics, Sport Training, Sports Coaching, Sports Exercise Science, Sports Marketing|Comments Off

Total Goalkeeper Performance (TGP): A Comprehensive Metric for Evaluating Modern Soccer Goalkeepers

Authors: Daniel J. Marcolongo¹, Bret R. Myers²

¹Graduate of Sports Industry Management Program, Georgetown University, Washington DC, USA

²Department of Management and Operations, Villanova University, Villanova, PA, USA

Corresponding Author:

Bret R, Myers, Ph.D.

Department of Management and Operations

Villanova School of Business

800 E Lancaster Avenue

Villanova, PA 19085

[email protected]

Daniel J. Marcolongo is a 2025 graduate of Georgetown University’s Sports Industry Management masters program. His focus is in soccer analytics, which he developed as a collegiate soccer player and lifelong student of the game.

Bret R. Myers, Ph.D. is a Professor of Practice in the Department of Management and Operations in the Villanova School of Business. His research interests focus on sports analytics, specifically, in the areas of team evaluation and managerial decision-making. He also is an active Analytics Consultant with 10+ years of experience working with professional teams and other sports organizations.

ABSTRACT

The purpose of this study was to develop and validate a comprehensive metric for evaluating modern soccer goalkeepers that accounts for both defensive and offensive responsibilities. Total Goalkeeper Performance (TGP) was constructed using publicly available data from the English Premier League, incorporating shot-stopping, cross-stopping, sweeping, and distribution metrics. Analysis of 70 observations of goalkeeper performance revealed a moderate positive correlation between TGP and team success (r = 0.474, p < 0.001), with TGP explaining 22.5% of variance in expected team points per game (3 points for win/1 point for draw/0 points for loss). A one-unit increase in TGP corresponded to 1.75-4.64 additional expected points over a 38-match season. Year-over-year analysis showed moderate consistency in goalkeeper performance as measured by TGP. These findings suggest TGP effectively captures goalkeeper contribution to team success while accounting for the evolving multidimensional nature of the position. TGP provides a data-driven framework for recruitment, talent identification, and tactical planning that aligns with the demands of modern soccer.

Key Words: soccer analytics, goalkeeper metrics, player evaluation, player development

INTRODUCTION

The goalkeeper is a unique position in sports. The player is often isolated from the rest of the team as the last line of defense. They even have different equipment from the rest of the team. In hockey, soccer, and more, one can immediately distinguish a goalkeeper from their teammates (3, 5). It leads to a feeling that the goalkeeper is isolated from the rest of the team. But no one is an island. There have been times when goalkeepers have done more than just protect their goal, which has been seen prominently in soccer (13).

This has been supported by several studies into the position. A study using machine learning algorithms showed that the difference between what they called elite and sub-elite goalkeepers was their ability with their feet (11). This suggests that the position has evolved so much that shot stopping is not even a goalkeeper’s main priority at the world’s best clubs. This was far from the only study to suggest that goalkeepers have seen an increase in their responsibilities. Goalkeepers are asked to do a lot more than just save shots in today’s game (13, 19). Soccer is not the only sport where this has occurred.

From roughly the mid-1990s to the mid-2000s, certain hockey goaltenders had a similar task. The most notable was Martin Brodeur. The New Jersey Devils, Brodeur’s team, employed a strategy called a neutral zone trap. The trap utilized a goalkeeper’s ability outside of the net to limit their opponent’s scoring chances. The Devils won three Stanley Cup titles from 1994 to 2003 with this strategy before the NHL introduced a new rule which severely limited what a goalkeeper could do outside of making saves (5).

In soccer, a goalkeeper is the only player on the team who can touch the ball with their hands and the player can only do this inside the 18-yard box (2). Historically, this led to goalkeepers not playing with their feet at all. But as the game evolved, this began to change, helped along by a rule change after the 1990 World Cup. The 1990 World Cup is considered one of the worst World Cups of all time due to the boring play. Part of the reason for the dullness was the goalkeepers who would waste time by holding the ball as long as legally allowed (14).

To combat this, the back-pass rule was introduced. This established the rule that goalkeepers could not pick up the ball if a player on their team passed it to them (14). With the change introduced to the game, it made goalkeepers’ ability to play with their feet more important (1). Following the success of goalkeepers like Manuel Neuer and Ederson, a goalkeeper’s distribution has become an essential skill (16-17). It is to the point that in some development teams, such as Chelsea, goalkeepers are judged more for their passing than their saves (4). Goalkeepers need to do so much more than just stop shots. However, that idea still hasn’t taken hold.

There is no way to rank soccer goalkeepers in a way that accounts for what they do with the ball. There isn’t even a statistic to accurately rank a goalkeeper defensively. Some statistics individually look at saves, cross-stopping, and sweeping, but no statistics takes all those aspects into account (6). In fact, some awards recognize players for a single performance statistic. For example, the Premier League Golden Glove award is given to the goalkeeper who has had the most ‘clean sheets’ (i.e., a game where they did not allow a goal) (10). The way goalkeepers are ranked has not kept up with the times. There should be a new statistic that accurately rates a goalkeeper based on everything they have to do, their Total Goalkeeper Performance.

The purpose of this study is to develop and advance a comprehensive metric for evaluating modern soccer goalkeepers that captures both their defensive and offensive responsibilities. First, the study outlines the methodology, including the acquisition of player performance data from the English Premier League for men’s professional soccer. Second, the offensive and defensive statistics used to construct the Total Goalkeeper Performance (TGP) metric are defined, and the procedures for calculating these statistics are detailed. Third, the data analysis plan—featuring correlation analysis and regression modeling—is described. The results are then presented and interpreted, followed by the study’s conclusions. Finally, the practical applications of this metric within sports analytics, particularly in organized soccer, are discussed.

METHODS

Dataset and Sampling

The data used in the paper spans eight seasons from the Premier League (2017-2018 through 2024-2025), a period where modern goalkeeper statistics are publicly available. All of the data comes from FBRef.com, with the exception of the data on punches which came from the Premier League’s website. In all, 70 goalkeepers were examined.

TGP features several different parts of a goalkeeper’s responsibilities. These can be divided into defensive and offensive statistics. Based on the data available the following performance statistics are used to construct TGP.

Defensive Statistics

Defensively, the main responsibility a goalkeeper has is shot stopping, making saves to prevent goals, but that’s not their only task. Goalkeepers also must defend the goal when balls come into their area. This can be from either crosses that a goalkeeper must deal with inside the 18-yard box or passes that force a goalkeeper to leave the box (18). These skills will be called cross-stopping and sweeping.

Shot Stopping

Shot stopping will be measured with expected goals (xG). xG tracks how likely a goal is to be scored from the moment it is struck on a scale of 0-1 with a better shot being ranked closer to 1 (8). To make this stat useful for a goalkeeper, one must track how much xG a goalkeeper faces and then subtract the total number of goals allowed to get the post-shot (PS) expected goals minus goals allowed (GA). In order to standardize this for all goalkeepers, the statistic will be converted into a per 90 minutes through using the minutes played by each goalkeeper (PSxG-GA/90).

2. Cross Stopping

Cross Stopping is another skill needed to be quantified. Goalkeepers typically face several crosses being put into their box during a game. The best way to stop a cross is to claim it, catching the cross before the opposition can get to it. Punching a cross away can also be beneficial but is not preferable to catching as the ball could go back to the opposition but it is still preferable to leaving the cross to the opposition (9). When making cross-stopping into a statistic, one must factor in both crosses claimed (CC), crosses punched (CP), and total crosses faced (TC), but claiming and punching crosses are not equal.. Because of that, TGP weighs a punch as half of a claim when measuring cross-stopping. The final stat to measure cross stopping for TGP is: Cross Stopping = (CC+.5xCP)/TC.

3. Sweeping

Sweeping is the easiest of the three defensive skills to quantify. The statistic – defensive actions outside of the penalty area – measures sweeping well. This tracks how often a goalkeeper comes outside of his goal to help his team (6). The more often a goalkeeper does this, the better they are at sweeping. To fairly measure these statistics in comparison, it must be looked at on a per-90-minute basis as well. TGP will use defensive actions outside the penalty area per 90 (DAOP/90) minutes to measure sweeping.

From an interview with US International goalkeeper Tyler Miller, shot-stopping is the most important, followed by cross-stopping, then sweeping (12). TGP will weigh the skills 3:2:1 in that order. The stats will then be added together to make a defensive score.

Offensive Statistics

Days 1-4 focused on primary lift progression (Front Squat, Bench Press, Deadlift, Overhead Press) with integrated plyometric, conditioning, and movement quality components. Day 5 emphasized pulling strength and unilateral work. Day 6 focused on coordination, explosive power, and metabolic conditioning. All days included Tabata rowing (20 seconds work/10 seconds rest × 8 rounds) for conditioning stimulus and mental toughness development.

Pass Completion Percentage (in buildup)

Offensive skills are more difficult to track due to the limitations of data on goalkeepers’ offensive abilities. One skill to track is a goalkeeper’s ability in buildup, making short passes to help his team up the field. In addition, a goalkeeper’s ability to make decisive passes that can start an attack on his team is important as well. The best widely available statistic to track a goalkeeper’s ability in buildup is completion percentage (PC) , how accurate they are as a passer.

2. Long Pass Completion Percentage

Similarly, long pass completion percentage (LP) shows how effective a goalkeeper is with long passes that are more likely to lead to an attack (6). Both statistics, completion percentage and long pass completion percentage, will be weighed equally to make an offensive score.

Component Weighting and Possession-Based Adjustments

The TGP metric is a weighted average of offensive and defensive scores. However, weights are conditionally applied based on possession characteristics of the team. Teams with more possession tend to take more advantage of the offensive skills of the goalkeeper through more time with the ball. Meanwhile, teams with less of the ball have much less of a use for an offensively minded goalkeeper. (18). Because of that, possession, the statistic for how much of the ball a team has per game, is a way to weigh how important a goalkeeper’s offensive skills are for a team (6).

To ensure that the statistic is applicable across different seasons, a player’s score in different statistics will be weighed against the league’s average score. This includes the league average scores on shot-stopping (μPSxG – GA/90), cross stopping (μ(CC + .5CP)/TC)), sweeping (μDAOP/90), pass completion percentage(μPC), and long pass completion percentage (μLP).

For a team with 62.5 percent possession (P) or more, a number chosen for ease of calculations though it is a number only the most ball dominant teams can reach, the offensive and defensive scores will be weighted equally. For a team with 37.5 percent possession or below, a number chosen for the same reasons but for the least ball dominant teams, it will be weighted 3:1 defensive score to the offensive score (7). For example a player on a team with 62.5% possession or more his defensive and offensive scores would remain the same for calculations. In a team with 37.5% possession or below his defensive score would be multiplied by 1.5 and his offensive score multiplied by 0.5. For a team with between 37.5 and 62.5 percent possession the weight would slide between those ratios. For example, in a team with .531 percent possession a goalkeeper would have his defensive score multiplied by 1.188 and his offensive score multiplied by .812.

The overall TGP formula for a goalkeeper per match can be expressed as follows:

TGP = (DS*(2 – (2P – 0.25))) + (OS*(2P – 0.25))) / 2

where:
DS = 1.47*(((PSxG – GA/90 + 0.52)/(μPSxG – GA/90 + 0.52)*5.09) + 0.97*(((CC + .5CP)/TC)/μ(CC + .5CP)/TC)*5.15) + 0.62*((DAOP/90)/μ(DAOP/90)*4.02) OS = (0.90*((PC*/μPC*)/10.12) + 1.01*((LP*/μLP*)/9.92)/(4/3)

*μ represented the mean levels of the performance metric by league.

Here is a sample calculation for a high performing goalkeeper with the following measures:

Nick Pope 2023-24: TGP=(19.37*1.206 + 16.05*0.794) / 2 = 18.03

DS=1.47*(.58/0.44)*5.09 + (0.97*0.98)/0.83)*5.15) + 0.62*(1.87/1.29)*4.02=19.37

OS=(0.99(76.9/72.9)*10.12+1.01*(47.9/44.27)*9.92)/(4/3)=16.05

Here is a sample calculation for a low performing goalkeeper with the following measures:

James Trafford 2023-24: TGP (14*1.302 + 12.09*.698) / 2 = 13.36

DS=1.47*(0.31/0.44)*5.09 + 0.97*(0.95/0.83)*5.15) + 0.62*(1.54/1.29)*4.02=14.00

OS=(0.99(65.5/72.9)*10.12+1.01*(32/44.27)*9.92)/(4/3)=12.16

The formula is created with a score of 15 to be the league average for every season. The numbers each individual statistic is multiplied by is there to ensure that no stat is weighted more than any other.

Analyses and Visualizations

Three key areas in this study are explored: 1) Relationship between TGP and Team Success, 2) Individual TGP Rankings and Year-Over-Year Repeatability, 3) TGP vs. Player Market Value. In order to evaluate Team Success, Team EPL Points per Game (PPG) will be used (3 points for team 1, 1 point for team draw, 0 points for team loss). Python is used to carry out correlation and regression analyses exploring the relationship between TGP and Team Success based on n = 70 qualifying goalkeepers from the EPL. Specifically, the scipy library is used for correlation analysis and statsmodels library is used for regression analysis. Furthermore, data visualization is carried out using Python’s matplotlib library. Correlation analysis and data visualization (also from Python) are also used to explore year-over-year repeatability of TGP scores based on n = 10 qualifying goalkeepers, Similar methods are also used to help examine the relationship between TGP and Player Market value.

RESULTS

Relationship between TGP and Team Success

In order to assess the relationship between TGP and team success, a Pearson correlation analysis was performed comparing TGP to PPG for 70 observations across the 2022-2023, 2023-2024, and 2024-2025 English Premier league seasons. The data set is representative of 39 distinct goalkeepers that qualify by having played at least 10 matches.

The analysis revealed a moderative positive correlation between TGP and PPG (r = 0.474 and p < 0.001). This indicates that goalkeepers with higher TGP scores tend to play for teams that earn more points per match. Figure 1 displays the scatterplot with a fitted regression line which demonstrates the positive, linear trend between TGP and team performance. While correlation does not imply causation, the statistically significant relationship suggests that the multidimensional TGP metric captures aspects of goalkeeper performance that contributes directly to winning.

Figure 1

Note. Scatterplot depicting relationship Points per Game and TGP across 2022-2023, 2023-2024, and 2024-2025 seasons in the English Premier League.

Furthermore, a simple linear regression was performed to help understand the magnitude of the contribution to team success. The analysis was performed using the statsmodels library in Python and the results are included in Figure 2.

Figure 2

Note. Ordinary Least Squared Regression Results for TGP vs. Team Performance

The model was statistically significant, F(1,68)=19.74, p<0.001, and explained 22.5% of the variance in PPG (R² = 0.225). The resulting regression equation was:

PPG=0.143+0.084×TGP

The TGP coefficient was positive and significant (β=0.084, t=4.44, p1<0.001), with a 95% confidence interval ranging from 0.046 to 0.122. To put it more in practical terms, every 1 unit increase in TGP is expected to increase points per game from 0.046 to 0.122. In the context of a 38 match EPL season, a 1 unit increase in TGP exhibited by GKs would lead to 1.75 to 4.64 additional points.

Individual TGP Rankings and Year-over-Year Analysis

The Total Goalkeeper Performance (TGP) results for the 2024–2025 Premier League season are summarized in Table 2 below. The top-performing goalkeeper was Guglielmo Vicario of Tottenham Hotspur, who achieved a TGP score of 19.93 across 24 league appearances. Based on the established regression model, this corresponds to an expected points-per-game (PPG) value of approximately 1.81. In contrast, Alphonse Areola of West Ham recorded the lowest TGP score of 11.50 over 26 matches, translating to an expected PPG of roughly 1.10. When extrapolated over a full 38-match season, the difference in expected point contribution between a high-performing and low-performing goalkeeper equates to 26.98 points (68.78 vs. 41.80). While overall team success depends on multiple factors—including defensive structure and attacking capabilities—this analysis demonstrates that goalkeeper performance, as captured by TGP, is a significant independent driver of team outcomes.

Table 2

2024-2025 TGP Rankings in the EPL

Ranking	Player	Club	Effective Matches played (per 90)	TGP
1	Guglielmo Vicario	Tottenham	20	19.93
2	Ederson	Man City	21.8	18.95
3	Nick Pope	Newcastle	23	18.28
4	Robert Sánchez	Chelsea	27	17.92
5	Arijanet Muric	Ipswich	18	16.93
6	David Raya	Arsenal	33	16.78
7	Alisson	Liverpool	22.9	16
8	Mark Flekken	Brentford	31.4	15.87
9	Kepa Arrizabalaga	Bournemouth	26	15.7
10	Emiliano Martínez	Aston Villa	31	15.57
11	Jordan Pickford	Everton	33	14.98
12	Łukasz Fabiański	West Ham	11.9	14.54
13	Mads Hermansen	Leicester	25.5	14.43
14	Stefan Ortega	Man City	11.2	14.12
15	Bart Verbruggen	Brighton	31	14.1
16	André Onana	Man United	32	13.53
17	Bernd Leno	Fulham	33	13.45
18	Dean Henderson	Crystal Palace	33	13.38
19	José Sá	Wolves	25	12.37
20	Aaron Ramsdale	Southampton	25	12
21	Matz Sels	Nottingham Forest	32	11.68
22	Alphonse Areola	West Ham	21.1	11.5

There is also good evidence of the repeatability of TGP ratings year over year. That is – there is slight to moderate positive correlation between seasons. Table 3 represents the TGP performance of 10 GK who had qualifying minutes in the 2022-2023, 2023-2024, and 2024-2025 seasons.

Table 3

Year-over year TGP performances of qualifying Goalkeepers

Player	24-25 TGP	23-24 TGP	22-23 TGP
Emiliano Martínez	15.57	20.98	19.42
Ederson	18.95	19.53	16.57
Nick Pope	18.28	18.03	17.41
Alisson	16.00	16.06	20.72
David Raya	16.78	16.67	17.72
Robert Sánchez	17.92	17.33	13.96
Bernd Leno	13.45	14.47	18.67
Jordan Pickford	14.98	16.55	14.88
José Sá	12.37	17.43	12.96
Dean Henderson	13.38	11.70	12.90

To accompany this table, Figure 3 below shows a correlation matrix that summarizes the strength of the pairwise association between each of the last three seasons in terms of TGP performance.

Figure 3

Note. Correlation matrix of TGP performance for 2022-2023, 2023-2024, and 2024-2025 seasons

TGP vs. Player Market Value

Player evaluators and scouts need to be in tune with the market value of players. One common method is to use Transfermkt (https://www.transfermarkt.com/), a highly reputable site used to estimate player market value based on performance, potential, age, and other market trends. Accordingly, the player market values from the recent 2024-2025 season were collected and paired against TGP values. The relationship between the two variables is depicted in Figure 4.

Figure 4

Note. TGP vs. Player Market Value for the 2024-2025 season.

As you can see, there is a baseline positive relationship between TGP and player market value. The scatterplot also labels the player with a color-coding system such that players above the expectation of performance by salary are in green, while those at expectation are in yellow, and those below expectation in red. Given the typical club operates on player budgets for wages, it is a common goal to try to acquire players that deliver at or above expectations in terms of performance.

DISCUSSION

Interpretation of Findings

The results of this study provide compelling evidence for the utility of the Total Goalkeeper Performance (TGP) metric as a comprehensive evaluation tool for modern soccer goalkeepers. The correlation (r = 0.474, p < 0.001) between TGP and PPG is evidence of a moderate, positive association between goalkeeping performance (as measured by TGP) and team performance. Furthermore, it can be said that 22.5% of the variation in PPG can be explained by the TGP metric. Given that there are 11 players on the field that contribute to team performance, 22.5% in the goalkeeping position signifies how critical the position is to team success.

The regression model also indicates that a single unit increase in TGP corresponds to an additional 1.75 to 4.64 points over a 38-match season. This finding quantifies the tangible impact a high-performing goalkeeper can have on a team’s league position. The substantial 26.98-point difference in expected contribution between the highest and lowest TGP scores in our sample (Vicario at 19.93 vs. Areola at 11.50) underscores the potential competitive advantage teams can gain through goalkeeper selection and development.

The year-over-year analysis reveals moderate consistency in goalkeeper performance as measured by TGP, suggesting that while the metric captures some stable aspects of goalkeeper ability, performance also fluctuates due to contextual factors such as team defensive structure, managerial approach, and opposition quality. This temporal stability adds credibility to TGP as a metric that identifies genuine skill rather than merely capturing random variation.

Tactical or Practical Implications

The TGP metric offers several practical applications for soccer professionals. First, it provides a data-driven framework for recruitment and talent identification that aligns with the multidimensional demands of the modern goalkeeper position. Clubs can use TGP to identify goalkeepers whose specific skill profiles match their tactical approach, rather than relying on traditional metrics that may not capture relevant abilities.

For teams with high possession percentages, our findings suggest that investing in goalkeepers with strong distribution skills yields tangible benefits. Conversely, teams that typically have less possession might prioritize shot-stopping and cross-claiming abilities. This contextual approach to goalkeeper evaluation enables more nuanced decision-making in the transfer market. Our analysis shows how TGP can be paired with player valuations, which can enable front offices to make smarter decisions.

The year-over-year analysis provides insights for player development specialists. The moderate temporal stability of TGP scores suggests that while goalkeeper performance has a skill component that persists across seasons, there is also room for improvement through targeted training. Development programs could use TGP component scores to identify specific areas for improvement in young goalkeepers.

From a tactical perspective, managers can use TGP to inform game strategy. Understanding the relative strengths of opposition goalkeepers across different dimensions could influence pressing approaches, crossing strategies, and shot selection. Similarly, awareness of one’s own goalkeeper’s TGP profile might influence defensive organization and build-up patterns.

Limitations

Several limitations must be acknowledged when interpreting these results. First, while our dataset includes three seasons of Premier League data, it represents only one league.. Goalkeeper requirements may differ substantially across leagues with different tactical tendencies, and the TGP weightings established here may not generalize perfectly to other contexts.

Second, our reliance on publicly available data limits the granularity of our analysis. More sophisticated tracking data could provide additional insights into goalkeeper positioning, command of area, and communication—aspects that are difficult to quantify with event data alone. The offensive component of TGP is particularly constrained by data availability, as metrics like pass completion percentage do not fully capture the quality and tactical significance of goalkeeper distribution.

Third, while we adjusted for team possession, other contextual factors like defensive structure, opposition quality, and score state may influence goalkeeper performance in ways not fully accounted for in the TGP metric. A goalkeeper playing behind a well-organized defense may face fewer high-quality shots, potentially affecting their PSxG-GA/90 component.

Finally, our weighting system, while informed by intuitive insight, introduces a subjective element to the metric. Different experts might propose alternative weightings based on their philosophical approach to the position. Future research could explore the sensitivity of TGP to different weighting schemes or develop data-driven approaches to component weighting. Despite these limitations, TGP represents a significant advancement in goalkeeper evaluation methodology and provides a foundation for future refinements as data availability and analytical techniques continue to evolve.

CONCLUSION

This study demonstrates that the Total Goalkeeper Performance (TGP) metric is a robust and comprehensive tool for evaluating modern goalkeepers. By integrating both defensive and offensive contributions into a single, possession-adjusted framework, TGP captures the multidimensional nature of the position more effectively than existing measures. The results show a clear and statistically meaningful relationship between TGP and team success, as well as moderate year-over-year consistency, establishing TGP as a credible and practical benchmark for goalkeeper performance.

TGP should be recognized as a new standard for goalkeeper evaluation. It provides clubs, coaches, and analysts with a powerful framework for recruitment, player development, and tactical decision-making. The metric moves beyond traditional, outdated statistics such as clean sheets and instead delivers a data-driven, holistic assessment that reflects the modern demands of the position.

While future refinements—particularly improved offensive data, expanded league coverage, and longitudinal tracking—will further strengthen its utility, the evidence presented here is clear: TGP represents a decisive advancement in goalkeeper analytics. By adopting this framework, the soccer industry can better align evaluation practices with the realities of today’s game and gain a competitive edge in identifying and developing top goalkeepers.

APPLICATIONS IN SPORT

TGP provides practical value for multiple stakeholders in professional soccer. For technical directors and recruitment teams, it offers a multidimensional framework for goalkeeper evaluation that aligns with modern tactical demands, enabling more informed transfer decisions by identifying goalkeepers whose specific skill profiles match a team’s playing style. For coaches and tactical analysts, TGP components can inform game strategy by highlighting opposition goalkeeper weaknesses across different dimensions. Teams might adjust pressing approaches against goalkeepers with poor distribution or increase crossing volume against those who struggle with aerial control. Player development specialists can utilize TGP component scores to create targeted training programs addressing specific goalkeeper weaknesses, allowing youth academies to track development progress across all relevant goalkeeper skills rather than focusing exclusively on traditional shot-stopping metrics.

This type of expanded analysis has proven transformative in other sports. In American football, quarterback evaluation has evolved far beyond simple counting statistics such as touchdowns or interceptions. Advanced metrics like Expected Points Added (EPA), Completion Percentage Over Expectation (CPOE), and QBR now provide a multidimensional assessment of quarterback decision-making, efficiency, and contextual performance. In baseball, the introduction of Wins Above Replacement (WAR) revolutionized how players are valued, combining offensive, defensive, and baserunning contributions into a single comprehensive number. These examples illustrate the power of moving past one-dimensional measures to holistic frameworks that better reflect player impact.

Soccer goalkeepers are a natural candidate for this type of approach, but they are not alone. Other sports positions that blend defensive and offensive responsibilities—such as catchers in baseball, liberos in volleyball, or goaltenders in lacrosse and hockey—could benefit from similar metrics that capture their multifaceted roles. Expanding evaluation frameworks in this way allows teams across sports to more accurately quantify player value, align talent acquisition with tactical systems, and design targeted development programs that reflect the true demands of the position.

REFERENCES

Alencar, M. (2024, Mar 20). Higuita: Scorpion kick changed football forever: Rene Higuita’s acrobatic clearance helped usher in new era of ball-playing goalkeepers, he tells Mauricio Alencar. City A.M. https://proxy.library.georgetown.edu/login?url=https://www.proquest.com/newspapers/higuita-scorpion-kick-changed-football-forever/docview/2968727496/se-2?accountid=11091
Bate, A. (2021, May 26). Stanley Menzo interview: Goalkeeping pioneer who changed the game under Johan Cruyff at Ajax. Sky Sports. https://www.skysports.com/football/story-telling/11946/12311915
Bugda, G., & Swann, S. (2024, July 9). Who is the “goalkeeper” for your organization? Medium. https://twodummies.medium.com/who-is-the-goalkeeper-for-your-organization-aa6b53838f9f
Carmichael-Brown, S. [Hashtag United]. (2024, January 9). CHELSEA GOALKEEPER TED CURD RECALLED! [Video]. Youtube. https://www.youtube.com/watch?v=BLCTnNQfWPY
Diamos, J. (2005, September 16). Hockey; new rule will take a weapon away from Brodeur. HOCKEY – New Rule Will Take a Weapon Away From Brodeur – NYTimes.com. http://web.archive.org/web/20131101025522/https://select.nytimes.com/gst/abstract.html?res=F10616F835550C758DDDA00894DD404482
FBRef. (2025). Premier League goalkeeper stats. FBref.com. https://fbref.com/en/comps/9/keepersadv/Premier-League-Stats#all_stats_keeper_adv
FBRef. (2025). Premier League stats. FBRef.com https://fbref.com/en/comps/9/Premier-League-Stats
FBRef. (2019). XG explained. FBref.com. https://fbref.com/en/expected-goals-model-explained/
FIFA. (2025, January 7). Goalkeeping: Dealing with crosses from corners. FIFA Training Centre. https://www.fifatrainingcentre.com/en/game/tournaments/fu17wwcfu17wwc-24/2024/goalkeeping-dealing-with-crosses.php
Garrick, O. (2024, May 19). Arsenal’s David Raya wins 2023-24 premier league golden glove award. The New York Times .https://www.nytimes.com/athletic/5495358/2024/05/19/david-raya-arsenal-golden-glove/
Jamil, M., Phatak, A., Mehta, S., Beato, M., Memmert, D., & Connor, M. (2021, November 22). Using multiple machine learning algorithms to classify elite and sub-elite goalkeepers in professional men’s football. Nature News. https://www.nature.com/articles/s41598-021-01187-5#Sec7
Marcolongo, D., & Miller, T. (2025, March 27). Interview with Tyler Miller. personal.
Obetko, M., Peráček, P., Mikulič, M., & Babic, M. (2022). Technical–tactical profile of an elite soccer goalkeeper. Journal of Physical Education and Sport, 22(1), 38-46. https://doi.org/10.7752/jpes.2022.01005
Patrikarakos, D. (2009). Defining Moment: The back-pass rule livens up football, 1992. FT.Com, https://proxy.library.georgetown.edu/login?url=https://www.proquest.com/trade-journals/defining-moment-back-pass-rule-livens-up-football/docview/229164255/se-2?accountid=11091
Premier League. (2025). Most punches – premier league goalkeeper stats. https://www.premierleague.com/stats/top/players/punches
Rose, G. (2025, January 8). Manchester City News: How Pep Guardiola has influenced goalkeeper evolution in English football. BBC Sport. https://www.bbc.com/sport/football/articles/clyvw8xlv4lo
Schmidt, C., & Hopkins, O. (2021, April 7). Manuel Neuer: Record-chaser and revolutionary. Opta Analyst. https://theanalyst.com/2021/04/manuel-neuer-record-chaser-and-revolutionary
Smith, A. (2023, August 15). Best goalkeeper in the Premier League revealed: Andre Onana, David Raya and Robert Sanchez transfers analysed. Sky Sports. https://www.skysports.com/football/news/11661/12933578/best-goalkeeper-in-the-premier-league-revealed-andre-onana-david-raya-and-robert-sanchez-transfers-analysed
Yam, D. (2019). A data driven goalkeeper evaluation framework. MIT Sloan Sports Analytics Conference. https://www.sloansportsconference.com/research-papers/a-data-driven-goalkeeper-evaluation-framework

Troy Palmer2025-12-03T09:26:11-06:00May 20th, 2026|General, Sport Training, Sports Coaching, Sports Exercise Science|Comments Off

The Role of Sport Relationships in Positive Youth Development

Authors: Jim P. Arnold¹ and William V. Massey¹

¹Department of Kinesiology, College of Health, Oregon State University

Corresponding Author:

Jim P. Arnold

[email protected]

Jim P. Arnold https://orcid.org/0009-0004-2282-1915
William V. Massey, Ph.D. https://orcid.org/0000-0002-4002-3720
We have no known conflicts of interest to disclose.

ABSTRACT

Purpose. Research on positive youth development (PYD) through sport remains unclear and speculative (Whitley et al., 2019). It has been suggested that sport-based PYD can occur implicitly through positive relationships (Holt et al., 2017). The present pilot study examined the impact of changes in the coach-athlete relationship, peer cohesion, and parental involvement on PYD outcomes across a sport season in a sample of youth soccer participants (N = 41, M_age= 11.85, 61% boys).

Methods. Athletes responded to surveys rating their relationships with coaches, parents, and peers at two time points, and additionally reported their perceptions of developmental skills gained across the sport season. A difference score was calculated for each relationship variable to measure change across the season. Four developmental outcomes (i.e., personal and social skills, cognitive skills, goal setting skills, initiative) were regressed on changes in relationship quality across the season, while controlling for age, race, and gender.

Results. Changes to the coach-athlete relationship (b= 0.482, p = 0.002) and parental involvement (b= 0.326, p = 0.022) were significant predictors of perceptions of social skill development (R² = 0.454, F(5,34) = 5.664, p < 0.001), supporting a relationship-based model of PYD in sport. Significant age and gender differences in ratings of the coach-athlete relationship were also discovered.

Conclusions. The present study not only offers partial support to a Holt and colleagues’ (2017) theory of implicit PYD through sport but also highlights the need important developmental role of relationship building in in the sport context.

Applications in Sport. Organizations should prioritize positive sport relationships through education, training, and programming, as poor or absent relationships may undermine the envisioned benefits of sport. In particular, the present study highlights the need for positive parental involvement, which may require stakeholders to work with parents to define their role expectations.

KEYWORDS: youth sport, positive youth development, sport relationships, coaching, parental involvement

INTRODUCTION

The goal of supporting positive outcomes for younger people (i.e., generativity; Erikson, 1950) is one that is both widely and cross-culturally relevant, yet despite this, the understanding for how to best support young people and the strategies employed to do so are still in flux. Only recently have developmental psychology and social research begun to place an emphasis on fostering positive outcomes for youth, as opposed to the prevention of negative outcomes and problematic behaviors (Larson, 2000). Within the areas of social and developmental research, this emphasis has led to the creation of diverse approaches to and philosophies of developmental youth programming (Lerner et al., 2011), which often provide opportunities for life skill development (i.e., explicit positive youth development). That said, the translation of such knowledge to spaces where youth development is view as a secondary priority, such as sport, tends to be challenging (Jones et al., 2011). The primary aim of the present pilot study was to test a grounded theory of implicit positive youth development through sport by examining the impact of peer, coach, and parental relationships on youth sport experiences in a small, single-organization sample. In doing so, the present study offers a novel examination of the collective social climate (i.e., PYD climate) and its relationship to athlete developmental outcomes. We hypothesized the following:

Athletes’ perceptions of positive outcomes obtained through sport participation (e.g., social skills, goal setting skills) will be predicted by positive changes to the ratings of the coach-athlete relationship, peer cohesion, and parental involvement across a sport season.

At two time points (e.g., beginning of the season, end of the season), athletes’ ratings of their relationships with their coach, peer cohesion, and parental involvement were collected. Subsequently, athletes’ perceptions of skill development across four areas (e.g., personal and social skills, cognitive skills, goal setting, initiative) were regressed on changes to the relationship variables. Both the coach-athlete relationship and parental involvement were shown to significantly predict social skill development, not only offering partial support for a theory of implicit PYD through sport and underscoring the critical developmental role of relationship building in sport but also pointing to the need for stakeholders to prioritize a high-quality social climate in the sport context to better support youth development.

LITERATURE REVIEW

Historically, adolescence and adolescent development has been regarded as a period during which youth are at risk and laden with problematic behaviors (Benson et al., 2006), therefore implying that the role of adults was to manage and prevent the problems that arise from adolescent development, also known as a deficit-focused approach to youth development (Clonan et al., 2004; Lerner, 2005). However, preventing such problems through a focus on treatment or intervention often failed to yield positive results (Catalano et al., 2008). Appearing concurrently with positive psychology’s focus on human strengths and flourishing, positive youth development theory offered that youth are “resources to be developed,” presenting a path toward positive youth outcomes through youth enrichment and the promotion of adolescent strengths (Lerner, Almerigi, et al., 2005). Positive youth development is a broad term, but generally refers to “processes, approaches, and instances” (Lerner et al., 2011) which seek to optimally prepare young people for adulthood, with the targeted outcomes being well-being and the fulfillment of their potential (Catalano et al., 2008). Contexts which aim to support positive youth development vary widely, to include agricultural programming (Lerner, Lerner, et al., 2005), volunteer and service programming (McBride et al., 2011), tutoring (Worker et al., 2019), aquatics (Storm et al., 2017), adventure-based programming (Sibthorp & Morgan, 2011), and sport (Bruner et al., 2021).

Youth sports are generally touted as tools for healthy and positive development, yet research aimed at validating this claim or understanding the processes by which it occurs is ambiguous (Holt et al., 2017). PYD theory was developed outside of the sport context (Lerner, Lerner, et al., 2005) and researchers have struggled to apply PYD models and measures to sporting contexts (Jones et al., 2011). One reason for this may be that PYD researchers have failed to acknowledge keyfeatures of the sport environment (Holt et al., 2017). In a systematic review of qualitative data, Holt and colleagues (2017) proposed that PYD through sport occurs via two distinct pathways. In the first, programs offer explicit education to youth sport participants aimed at life skill development. In the second pathway, PYD occurs implicitly via positive relationships with coaches, peers, and parents (i.e., the creation of a ‘PYD climate’). Holt and colleagues concluded that further research is needed to not only investigate the validity of this framework but also understand additional nuances for when and how PYD may occur through explicit and implicit factors. The need for further research was bolstered by a systematic review of sport-based PYD programming, conducted by Whitley and colleagues (2019), who concluded the benefit of explicit PYD programming in sport is not clear enough to support the implementation of a standardized intervention. Therefore, while the field’s understanding of how to best implement explicit PYD programming through sport is still evolving, there also exists a need to test the proposed model of implicit PYD through positive relationships within sport. While the specific role positive relationships play in supporting PYD within sport is unclear, it is generally accepted that these relationships are all valuable, if not necessary, for positive athlete outcomes (Burns et al., 2019).

Coach-Athlete Relationship

Arguably the primary relationship in the sporting context (Jowett, 2017), the dyadic relationship between coach and athlete has been shown to be instrumental to numerous athlete outcomes. In a systematic review of the coach-athlete relationship literature, Nikolina and Đorić (2023) reported that a positive coach-athlete relationship was not only predictive of increased motivation, satisfaction, and performance, but also protective from athlete stress, burnout, and negative affect. Davis and Jowett (2014) have reported that the quality of the coach-athlete relationship is directly related to athlete positive and negative affect. Furthermore, in a systematic review of the literature, McShan and Moore (2023) found that a positive coach-athlete relationship, as reported by coaches, was associated with coach’s beliefs of fostering an environment supportive of athlete life skill development. In Holt and colleague’s (2017) grounded theory of implicit PYD, the authors posit that strong, positive relationships between athletes and coaches can create a developmentally supportive social environment.

Peer Cohesion

Paralleling the coach-athlete relationship research, research on the role of peer relationships in the sport environment have shown these relationships to be highly influential on athlete experiences and outcomes (Smith & Ullrich-French, 2020). Peer support has been shown to be related to elite sport participation, athlete motivation, and reduced withdrawal from sport (Sheridan et al., 2014). Additionally, researchers have shown that peer cohesion is not only associated with performance (Carron et al., 2002; Filho et al., 2014), but also athlete need satisfaction and learning (Erikstad et al., 2018). Furthermore, Smith and Ulrich-French (2020) have posited that peer relationships in the sport context are likely to be influential to individual athlete development, to include character, moral, social, and life skill development. In proposing strong peer relationships as influential of an implicit PYD climate, Holt and colleagues (2017) highlighted how strong peer relationships in the sport context often result in feelings of belongingness and support, which may provide developmental benefit.

Parental Involvement

While not always directly involved in the training environment, researchers have shown that parents are highly influential to youth athletes’ experiences and outcomes in sport. Youth who perceive their parents as satisfied with their performance and who experience low parental pressure are more likely to report sport enjoyment and positive affect (Dorsch et al., 2021). Additionally, parental involvement has also been associated with youth sport enjoyment, perceptions of competence, and self-esteem (Dorsch et al., 2021). Parental involvement in sport has also been found to be associated with youth athlete need satisfaction (Felber Charbonneau & Camiré, 2020). Furthermore, parental involvement in sport has also been connected to athletes’ development, to include socialization and value adoption (Danioni et al., 2017). In their grounded theory model, Holt and colleagues (2017) highlighted the reinforcing role that parental involvement plays to creating a PYD climate; while coaches may be responsible for delivering lessons and values to athletes in the sport context, the authors noted that it is important that parents support, not contradict, these messages.

Study Aims

In their grounded theory model, Holt and colleagues (2017) posited that these three relationships (i.e., coaches, peers, parents) collectively create a social climate supportive of implicit positive youth development. Therefore, the primary aim of the present study was to examine the impact of peer, coach, and parental relationships on youth sport experiences and youth athletes’ perceptions of developmental skills gained, thereby piloting a test of Holt and colleagues’ (2017) grounded theory model. Should these relationships be predictive of positive youth development, it could be expected that athletes who experience positive changes to these relationships (e.g., increased peer cohesion, increased parental involvement) across a sport season should also receive increased benefit from their participation compared to athletes whose relationships did not improve. As such, we hypothesized that athletes’ perceptions of positive outcomes obtained through sport participation (e.g., social skills, goal setting skills) would be predicted by positive changes to the ratings of their peer relationships, coach-athlete relationships, and parental involvement across a sport season.

METHODS

Participants

Participants included 67 youth athletes from a competitive soccer club in the northwest region of the United States. In total, 41 athletes (M_age= 11.85) completed data collection at both time points. Participants represented 13 teams from four separate age categories. Additionally, 65.9% of the athletes identified as white and 61.0% of the athletes identified as boys.

Measures

Coach-Athlete Relationship Questionnaire (CART-Q)

To measure athlete perceptions of their relationship with their coach, the Coach-Athlete Relationship Questionnaire (CART-Q; Jowett & Ntoumanis, 2004) was utilized. The 11-item scale measured the nature of the athlete’s relationship with their coach (a = 0.97). Using a seven-point Likert scale, athletes rated their agreement with statements such as, “I trust my coach.”

Youth Sport Environment Questionnaire (YSEQ)

Athletes’ perceptions of their relationship with teammates were measured utilizing the Youth Sport Environment Questionnaire (YSEQ; Eys et al., 2009). The scale, which has been shown to be both valid and reliable, measured group cohesion and peer relationship quality. The YSEQ contains 16 statements, such as, “I am happy with my team’s level of desire to win” (a = 0.93). Athletes rated their agreement with these statements utilizing a seven-point Likert scale.

Parental Involvement in Sport Questionnaire (PISQ)

The Parental Involvement in Sport Questionnaire (PISQ; Lee & MacLean, 1997) is a valid and reliable 19-item scale (a = 0.87), which captures athletes’ perceptions of parental involvement across three subscales: directive behavior, praise and understanding, and active involvement. Utilizing a five-point Likert scale, athletes rated their level of agreement with statements such as, “Do your parents push you to practice harder?”

Youth Experience Survey for Sport (YES-S)

Employed only at the second time point, the short form Youth Experience Survey for Sport (YES-S; MacDonald et al., 2012; Sullivan et al., 2015) is 16-item scale that measured the perceptions of athletes’ experiences participating in sport across the previous season, and was utilized in the present study to operationalize PYD. The scale measures whether athletes perceived any benefit to their participation across four subscales: personal and social skills (a = 0.78), cognitive skills (a = 0.78), goal setting (a = 0.81), and initiative (a = 0.71). Athletes rated their agreement with statements such as, “I learned to push myself” on a five-point Likert scale.

Procedure

Ahead of the start of the summer season, the first author attended the club’s tryouts and parent meetings to share information about the study and recruit participants. During this time, parental consent was obtained through the completion of a written consent form and household demographic survey. The first survey was completed electronically one month into the summer season. Subsequently, 14 weeks later, the research team returned to conduct the second survey during the final week of the fall season. At both time points, the surveys collected demographic information, athlete perceptions of relationships with their coach, peer cohesion, and parental involvement. At the second time point, the survey collected measurements of athletes’ perceptions of their experiences playing sport across the previous season, particularly focused on skills gained.

The dataset contained 0.3% missingness, and results of an MCAR test were not significant (X²(1386) = 0.00, p = 1.00), suggesting data was missing at random. For cases with missingness, scales were prorated based on completed items. Descriptive statistics were calculated for each scale and notable demographic differences are reported in Table 1. For each of the relationship variables (i.e., CART-Q, PISQ, YSEQ), a difference score was calculated (M_T2– M_T1) to measure changes in these relationships across the season. While the utilization of difference scores has been criticized for its negative, summative impact on reliability (Edwards, 1994), researchers have noted that difference scores can be an appropriate choice in research, particularly for nonrandomized, theory-driven analyses (Castro-Schilo & Grimm, 2018). Assumptions testing revealed issues regarding multicollinearity as there was a high correlation between coach-athlete relationship and the peer cohesion change scores (r = 0.801), which resulted in unstable beta coefficients. This instability indicated that the presence of the peer cohesion variable in the model was distorting the estimation of other predictors, undermining the reliability and interpretability of the model. As such, the peer cohesion variable was removed from primary analyses. Following this, we regressed the four subscales of the YES-S (i.e., personal and social skills, cognitive skills, goal setting skills, initiative) on changes in relationship quality across the season, while controlling for age, race, and gender.

Table 1

Sample Characteristics and Descriptive Statistics

			CART-Q		YSEQ		PISQ		YES-S Social Skills	YES-S Cog. Skills	YES-S Goal Setting	YES-S Initiative
Variable	n	%	T1 – M(SD)	T2 – M(SD)	T1 – M(SD)	T2 – M(SD)	T1 – M(SD)	T2 – M(SD)	T2 – M(SD)	T2 – M(SD)	T2 – M(SD)	T2 – M(SD)
Age
10	3	7.3	5.61(1.24)*	5.97(1.47)*	4.25(2.01)*	5.08(1.98)*	2.39(0.18)	2.91(0.45)	3.58(0.52)	3.67(0.58)	4.25(0.58)	4.58(0.52)
11	9	22.0	5.46(1.73)	6.36(0.39)	4.74(1.52)	5.53(0.83)	3.02(0.60)	3.13(0.52)	4.00(0.60)	3.69(1.05)	4.00(0.85)	4.50(0.45)
12	20	48.8	6.10(0.40)	5.96(0.85)	5.10(0.75)	5.30(0.91)*	2.92(0.60)	3.25(0.74)*	4.17(0.75)	3.53(1.16)	3.93(0.90)	4.25(0.59)
13	9	22.0	5.71(1.04)*	5.15(1.26)*^†	4.69(1.40)	4.89(1.18)	3.16(0.69)	3.30(0.58)	4.03(0.57)	3.56(0.69)	4.25(0.57)	4.43(0.66)
Gender
Boy	25	61.0	6.03(0.61)	6.17(0.69)^†	4.91(1.03)*	5.26(0.89)*	2.92(0.62)*	3.24(0.66)*	4.12(0.66)	3.72(0.85)	4.11(0.71)	4.43(0.41)
Girl	16	39.0	5.56(1.43)	5.41(0.99)^†	4.81(1.42)	5.22(1.25)	3.01(0.62)	3.17(0.62)	3.96(0.89)	3.35(1.19)	3.90(0.94)	4.27(0.76)
Race
White	27	65.9	5.77(1.13)	5.97(0.83)	4.75(1.21)*	5.24(1.03)*	2.95(0.61)	3.14(0.57)	4.07(0.69)	3.52(1.04)	3.99(0.85)	4.43(0.54)
Black	1	2.4
Asian	4	9.8	5.50(1.38)	5.41(1.85)	4.77(1.85)*	5.30(1.64)*	2.74(0.90)	3.29(0.90)	4.00(0.35)	3.94(0.43)	3.94(0.43)	3.94(0.66)
Hispanic	4	9.8	6.27(0.45)	5.86(1.12)	5.50(0.89)	5.55(0.74)	2.99(0.57)	3.41(0.83)	4.50(0.41)	4.25(0.54)	4.69(0.47)	4.63(0.32)
Other	5	12.2	6.13(0.31)	5.65(1.20)	5.05(0.83)	4.99(1.04)	2.99(0.45)	3.31(0.52)	3.80(0.89)	3.15(1.29)	3.80(0.94)	4.15(0.74)
Total	41	100.0	5.84(1.02)	5.87(0.99)	4.87(1.18)*	5.24(1.03)*	2.96(0.62)*	3.21(0.64)*	4.06(0.67)	3.58(1.00)	4.03(0.80)	4.37(0.57)

Notes. n = 41; CART-Q = Coach-Athlete Relationship; PISQ = Parental Involvement; YSEQ = Ratings of Peer Cohesion; YES-S = Perceptions of Developmental Experiences, *Difference is significant between time points; ^†Difference is significant between groups.

RESULTS

The model examining personal and social skills was significant and explained 45.4% of variance in the outcome (R² = 0.454, F(5,34) = 5.664, p < 0.001).

Regression Results for Perceptions of Social Skills Gained by Athletes

				95% CI
Variable	b	b	SE	LL	UL	p
Intercept		0.774	1.268	-1.802	3.350	0.546
Gender	-0.129	-0.175	0.184	-0.550	0.199	0.348
Age	0.369	0.291	0.110	0.067	0.515	0.012
Race	-0.024	-0.008	0.042	-0.094	0.078	0.858
DCART-Q	0.482	0.250	0.074	0.099	0.400	0.002
DPISQ	0.326	0.382	0.160	0.580	0.707	0.022

Notes. n = 41; R²= 0.454, F(5,34) = 5.664, p < 0.001; DCART-Q = Change in Coach-Athlete Relationship; DPISQ = Change in Parental Involvement.

**When ran independently due to the existing multicollinearity, change to peer cohesion was also a significant predictor of personal and social skills (R² = 0.317, F(4,35) = 4.063, p = 0.008).

Within this model, both changes to the coach-athlete relationships (b= 0.482, p = 0.002) and changes to parental involvement (b= 0.326, p = 0.022) across the season were significant predictors of personal and social skills. Additionally, the covariate age was also a significant predictor of personal and social skills (b = 0.369, p = 0.012). The model examining cognitive skills explained 25.1% of the variance, however was only marginally significant (R² = 0.251, F(5,34) = 2.275, p = 0.069). Within this model the change in coach-athlete relationship was a statistically significant predictor (b= 0.403, p = 0.022), whereas changes to parental involvement was not (b= 0.158, p = 0.330).

Table 3

Regression Results for Perceptions of Cognitive Skills Gained by Athletes

				95% CI
Variable	b	b	SE	LL	UL	p
Intercept		2.048	2.221	-2.465	6.561	0.363
Gender	-0.155	-0.315	0.323	-0.972	0.342	0.337
Age	0.143	0.169	0.193	-0.224	0.561	0.389
Race	-0.066	-0.032	0.074	-0.182	0.119	0.670
DCART-Q	0.403	0.312	0.130	0.048	0.576	0.022
DPISQ	0.158	0.277	0.280	-0.292	0.845	0.330

Notes. n = 41; R²= 0.251, F(5,34) = 2.275, p = 0.069; DCART-Q = Change in Coach-Athlete Relationship; DPISQ = Change in Parental Involvement.

** When ran independently due to the existing multicollinearity, change to peer cohesion was not a significant predictor of cognitive skills.

The models predicting goal setting skills (R² = 0.183, F(5,34) = 1.528, p = 0.207) and initiative (R² = 0.185, F(5,34) = 1.542, p = 0.203) were not statistically significant.

DISCUSSION

The present study provides partial support to Holt and colleague’s (2017) proposition that there is an implicit pathway of PYD in sport that takes place through positive relationships. In particular, changes to the coach-athlete relationship significantly predicted youth athletes’ perceptions of social skills and cognitive skills gained; and changes to perceptions of parental involvement also predicted social skills gained. Additionally, when analyzed separately due to issues of multicollinearity, changes to peer cohesion also significantly predicted social skill perceptions. As such, data in the current study reinforce the importance of relationships within the sport environment, and extend previous research by highlighting their value to the specific area of PYD through sport.

While research has shown the coach-athlete relationship to be associated with motivation (Adie & Jowett, 2010), collective-efficacy (Hampson & Jowett, 2014), and team cohesion (Turman, 2003), its role in the social and cognitive development of athletes is less understood. That said, research has shown that coaches seem to intuitively understand the developmental value of a positive coach-athlete relationship as coaches have reported a positive relationship with their athletes led to social and emotional development and resilience (White & Bennie, 2015). Furthermore, Davis and colleagues (2019) proposed a bidirectional relationship between communication skills and the coach-athlete relationship, where communication skills not only helped to improve the relationship, but also improved as a product of a high-quality coach-athlete relationship. When examining the more expansive literature on the impact of a high-quality relationships, researchers have documents that teacher-student relationships can promote cognitive development (Davis, 2003) and social adjustment (Dong et al., 2021) through positive and trusting learning environments. Data in the current study suggest coaches hold a responsibility to ensure the development and sustainment of positive relationships in the sport environment to support similarly positive developmental outcomes for youth athletes. This is particularly important as social skills have been shown to be associated with academic performance (Sung & Chang, 2010), increased mental health (Greenberg et al., 2003), wellbeing (Sancassiani et al., 2015), and self-esteem (Riggio et al., 1990).

The present study also highlights the important yet specific role that parents play in positive youth development through sport. Parental styles have been shown to be associated with social skill development; youth with democratic and permissive parents have been shown to score higher on social skills measures than those with neglectful or authoritative parents (Salavera et al., 2022). As such, it could be hypothesized that parents with more developmentally supportive parenting styles are more likely to be involved in their child’s sport and supportive of their child’s social skills. That said, data in the current study suggests the need to delineate the roles of parents and coaches, as these relationships may provide different benefits for youth. For example, Knight and colleagues (2011) reported that athletes consistently prefer parents to fill a supportive and encouraging role, as opposed to a coaching role. This is supported by data in the current study in that while change to parental involvement predicted athletes’ perceptions of social skill development, it did not predict their cognitive skill perceptions.

Finally, it is important to note that girls rated their relationship with their coach significantly lower than their peers who identified as boys; and older athletes were also significantly less likely to rate their coach-relationships higher than younger athletes. As such, should there exist any developmental benefit to high-quality, coaching relationships, the present findings would suggest that girls and older youth athletes are less likely to receive those benefits. Given that a positive coach-athlete relationship can be protective from poor mental health outcomes for girl athletes specifically (Massey et al., 2024), it is important that positive coach-athlete relationships are prioritized for female athletes, particularly adolescent female athletes. Furthermore, it is generally accepted that as athletes get older, the sporting environment shifts from a focus on fun to a focus on competition. Be that as it may, research has shown that the true shift lies within how athletes are treated; Kipp and Bolter (2020) found that while both older and younger athletes equally perceived their sporting environments to be focused on effort and learning, older athletes were more likely to report being punished or disciplined for mistakes. It is possible that such climates explain the decreasing trend of the coach-athlete relationship observed in the present study. Speaking strictly to the proposed developmental role of the coach-athlete relationship within sport, the present findings would offer that sports become less beneficial and developmentally supportive over time.

Despite the present study’s value to the literature base on PYD through sport, its small, homogenous sample limits its generalizability. In addition to being predominantly white, the sample derived from a singular, pay-to-play soccer organization within an affluent community. Additionally, the present sample predominantly identified as boys, which may parallel youth sport participation trends, but limits the generalizability of the findings to non-boy athlete populations. The age rage of the sample was also limited, clustered into the soccer organizations U11 and U13 age groupings, and as such, the findings may be in part reflective of the natural development occurring in this age range.

Furthermore, most athletes in the present study were satisfied with their relationship with their coach and peers, and the mean parental involvement score was slightly above the midpoint of the scale. Depending on sport or community context, it is possible that more athletes would report more dissatisfaction with these relationships or less parental involvement, thereby affecting the nature of the findings. With respect to age and gender differences, it is possible that these differences could be explained by confounding variables, such as coach gender, competition level, or position, which could not be differentiated in the present study due to the small sample size. Lastly, while multicollinearity necessitated the removal of the peer cohesion variable from the analyses, it should be acknowledged that doing so also limits the completeness of the model by excluding a theoretically important dimension of the sport environment, and one which should continue to be examined in this line of research. As such, future studies should not only continue to examine the nuanced roles of parents and coaches in sport-based PYD, but also peer relationships, and doing so in larger and more diverse samples.

CONCLUSION

The social context of the sport environment, which includes coaches, parents, and peers, plays a significant role in shaping athletes’ perceived development through sport. In the present study, athletes’ perceived social skill development was significantly predicted by positive changes to the coach-athlete relationship and parental involvement. The quality of the coach-athlete relationship also emerged as a meaningful predictor of athletes’ perceived cognitive development, highlighting the broader developmental impact of adult figures in the sport context. Furthermore, while peer cohesion was omitted in analyses due to multicollinearity, its interconnectedness with the coach-athlete relationship should be acknowledged, and researchers should continue to utilize it as a variable of interest as theory would dictate. Taken together, these findings underscore the importance of considering the full network of sport-based relationships when seeking to support athletes’ development through sport participation.

APPLICATIONS IN SPORT

In addition to providing support for Holt and colleagues’ (2017) theory of implicit PYD through sport, the present study highlights the interconnected nature of youth sport’s social context. We offer the following recommendations to stakeholders seeking to utilize these findings to develop their youth sport organization’s PYD climate:

Provide coaches with education and training that supports their development of communication and relationship-building skills (see Barnett et al., 1992; Jowett & Cockerill, 2003).
Provide education and clear expectations for parents’ involvement in the organization, as well as opportunities for involvement (see Knight et al., 2011).

Prioritize relationship building and psychological safety at the outset of the season, to include team-building activities and the development of team norms, rituals, and goals (see Carron et al., 1997; Senécal et al., 2008).

REFERENCES

1. Adie, J. W., & Jowett, S. (2010). Meta‐perceptions of the coach–athlete relationship, achievement goals, and intrinsic motivation among sport participants. Journal of Applied Social Psychology, 40(11), 2750-2773. https://doi.org/10.1111/j.1559-1816.2010.00679.x

2. Barnett, N. P., Smoll, F. L., & Smith, R. E. (1992). Effects of enhancing coach-athlete relationships on youth sport attrition. The Sport Psychologist, 6(2), 111-127. https://doi.org/10.1123/tsp.6.2.111

3. Benson, P. L., Scales, P. C., Hamilton, S. F., & Sesma Jr., A. (2006). Positive youth development so far: Core hypotheses and their implications for policy and practice. Search Institute Insights & Evidence 3(1), 1-13.

4. Bruner, M. W., McLaren, C. D., Sutcliffe, J. T., Gardner, L. A., Lubans, D. R., Smith, J. J., & Vella, S. A. (2021). The effect of sport-based interventions on positive youth development: A systematic review and meta-analysis. International Review of Sport and Exercise Psychology, 16(1), 368-395. https://doi.org/10.1080/1750984X.2021.1875496

5. Burns, L., Weissensteiner, J. R., & Cohen, M. (2019). Supportive interpersonal relationships: a key component to high-performance sport. British Journal of Sports Medicine, 53(22), 1386-1389. https://doi.org/10.1136/bjsports-2018-100312

6. Carron, A. V., Colman, M. M., Wheeler, J., & Stevens, D. (2002). Cohesion and performance in sport: A meta analysis. Journal of Sport and Exercise Psychology, 24(2), 168–188. https://doi.org/10.1123/jsep.24.2.168

7. Carron, A. V., Spink, K. S., & Prapavessis, H. (1997). Team building and cohesiveness in the sport and exercise setting: Use of indirect interventions. Journal of Applied Sport Psychology, 9(1), 61–72. https://doi.org/10.1080/10413209708415384

8. Castro-Schilo, L., & Grimm, K. J. (2018). Using residualized change versus difference scores for longitudinal research. Journal of Social and Personal Relationships, 35(1), 32-58.

9. Catalano, R. F., Toumbourou, J. W., & Hawkins, J. D. (2008). Positive youth development in the United States: History, efficacy, and links to moral and character education. In L. Nucci & T. Krettenauer (Eds.) Handbook of moral and character education. Routledge. https://doi.org/10.4324/9780203114896

10. Christiansen, E. D., & von Eye, A. (2005). Positive youth development, participation in community youth development programs, and community contributions of fifth-grade adolescents: Findings from the first wave of the 4-H study of positive youth development. The Journal of Early Adolescence, 25(1), 17-71. https://doi.org/10.1177/0272431604272461

10. Clonan, S. M., Chafouleas, S. M., McDougal, J. L., & Riley-Tillman, T. C. (2004). Positive psychology goes to school: Are we there yet? Psychology in the Schools, 41(1), 101–110. https://doi.org/10.1002/pits.10142

11. Danioni, F., Barni, D., & Rosnati, R. (2017). Transmitting sport values: The importance of parental involvement in children’s sport activity. Europe’s Journal of Psychology, 13(1), 75–92. https://doi.org/10.5964/ejop.v13i1.1265

12. Davis, H. A. (2003). Conceptualizing the role and influence of student-teacher relationships on children’s social and cognitive development. Educational Psychologist, 38(4), 207-234. https://doi.org/10.1207/S15326985EP3804_2

13. Davis, L., & Jowett, S. (2014). Coach–athlete attachment and the quality of the coach–athlete relationship: Implications for athlete’s well-being. Journal of Sports Sciences, 32(15), 1454–1464. https://doi.org/10.1080/02640414.2014.898183

14. Davis, L., Jowett, S., & Tafvelin, S. (2019). Communication strategies: The fuel for quality coach-athlete relationships and athlete satisfaction. Frontiers in Psychology, 10, 2156. https://doi.org/10.3389/fpsyg.2019.02156

15. Dong, Y., Wang, H., Luan, F., Li, Z., & Cheng, L. (2021). How children feel matters: Teacher–student relationship as an indirect role between interpersonal trust and social adjustment. Frontiers in Psychology, 11, 581235. https://doi.org/10.3389/fpsyg.2020.581235

16. Dorsch, T. E., Wright, E., Eckardt, V. C., Elliott, S., Thrower, S. N., & Knight, C. J. (2021). A history of parent involvement in organized youth sport: A scoping review. Sport, Exercise, and Performance Psychology, 10(4), 536. https://doi.org/10.1037/spy0000266

17. Edwards, J. R. (1994). Regression analysis as an alternative to difference scores. Journal of Management, 20(3), 683-689.

18. Erikson, E. H. (1950). Childhood and society. W. W. Norton & Co.

19. Erikstad, M. K., Martin, L. J., Haugen, T., & Høigaard, R. (2018). Group cohesion, needs satisfaction, and self-regulated learning: A one-year prospective study of elite youth soccer players’ perceptions of their club team. Psychology of Sport and Exercise, 39, 171–178. https://doi.org/10.1016/j.psychsport.2018.08.013

20. Eys, M., Loughead, T., Bray, S. R., & Carron, A. V. (2009). Development of a cohesion questionnaire for youth: The Youth Sport Environment Questionnaire. Journal of Sport and Exercise Psychology, 31(3), 390-408. https://doi.org/10.1123/jsep.31.3.390

21. Felber Charbonneau, E., & Camiré, M. (2020). Parental involvement in sport and the satisfaction of basic psychological needs: Perspectives from parent–child dyads. International Journal of Sport and Exercise Psychology, 18(5), 655–671. https://doi.org/10.1080/1612197X.2019.1570533

22. Filho, E., Dobersek, U., Gershgoren, L., Becker, B., & Tenenbaum, G. (2014). The cohesion–performance relationship in sport: A 10-year retrospective meta-analysis. Sport Sciences for Health, 10(3), 165–177. https://doi.org/10.1007/s11332-014-0188-7

23. Greenberg, M. T., Weissberg, R. P., O’Brien, M. U., Zins, J. E., Fredericks, L., Resnik, H., & Elias, M. J. (2003). Enhancing school-based prevention and youth development through coordinated social, emotional, and academic learning. American Psychologist, 58(6-7), 466. https://doi.org/10.1037/0003-066X.58.6-7.466

24. Hampson, R., & Jowett, S. (2014). Effects of coach leadership and coach–athlete relationship on collective efficacy. Scandinavian Journal of Medicine & Science in Sports, 24(2), 454-460. https://doi.org/10.1111/j.1600-0838.2012.01527.x

25. Holt, N. L., Neely, K. C., Slater, L. G., Camiré, M., Côté, J., Fraser-Thomas, J., MacDonald, D., Strachan, L., & Tamminen, K. A. (2017). A grounded theory of positive youth development through sport based on results from a qualitative meta-study. International Review of Sport and Exercise Psychology, 10(1), 1-49. https://doi.org/10.1080/1750984X.2016.1180704

26. Jones, M. I., Dunn, J. G., Holt, N. L., Sullivan, P. J., & Bloom, G. A. (2011). Exploring the ‘5Cs’ of positive youth development in sport. Journal of Sport Behavior, 34(3). https://psycnet.apa.org/record/2011-17095-003

27. Jowett, S. (2017). Coaching effectiveness: The coach–athlete relationship at its heart. Current Opinion in Psychology, 16, 154–158. https://doi.org/10.1016/j.copsyc.2017.05.006

28. Jowett, S., & Cockerill, I. M. (2003). Olympic medalists’ perspective of the athlete–coach relationship. Psychology of Sport and Exercise, 4(4), 313–331. https://doi.org/10.1016/S1469-0292(02)00011-0

29. Jowett, S., & Ntoumanis, N. (2004). The Coach–Athlete Relationship Questionnaire (CART‐Q): Development and initial validation. Scandinavian Journal of Medicine & Science in Sports, 14(4), 245-257. https://doi.org/10.1111/j.1600-0838.2003.00338.x

30. Kipp, L. E., & Bolter, N. D. (2020). Motivational climate, psychological needs, and personal and social responsibility in youth soccer: Comparisons by age group and competitive level. Psychology of Sport and Exercise, 51. https://doi.org/10.1016/j.psychsport.2020.101756

31. Knight, C. J., Neely, K. C., & Holt, N. L. (2011). Parental behaviors in team sports: How do female athletes want parents to behave? Journal of Applied Sport Psychology, 23(1), 76-92. https://doi.org/10.1080/10413200.2010.525589

32. Larson, R. W. (2000). Toward a psychology of positive youth development. American Psychologist, 55(1), 170–183. https://doi.org/10.1037/0003-066X.55.1.170

33. Lee, M., & MacLean, S. (1997). Sources of parental pressure among age group swimmers. European Journal of Physical Education, 2(2), 167-177. https://doi.org/10.1080/1740898970020204

34. Lerner, R. M. (2005). Promoting positive youth development: Theoretical and empirical bases. Washington, DC: National Academies of Sciences.

35. Lerner, R. M., Almerigi, J. B., Theokas, C., & Lerner, J. V. (2005). Positive youth development A view of the issues. The Journal of Early Adolescence, 25(1), 10–16. https://doi.org/10.1177/0272431604273211

36. Lerner, R. M., Lerner, J. V., Almerigi, J. B., Theokas, C., Phelps, E., Gestsdottir, S., Naudeau, S., Jelicic, H., Alberts, A., Ma, L., Smith, L. M., Bobek, D. L., Richman-Raphael, D., Simpson, I., 37.

37. Lerner, R. M., Lerner, J. V., Lewin-Bizan, S., Bowers, E. P., Boyd, M. J., Mueller, M. K., Schmid, K. L., & Napolitano, C. M. (2011). Positive youth development: Processes, orograms, and problematics. Journal of Youth Development, 6(3), 38–62. https://doi.org/10.5195/JYD.2011.174

38. MacDonald, D. J., Côté, J., Eys, M., & Deakin, J. (2012). Psychometric properties of the youth experience survey with young athletes. Psychology of Sport and Exercise, 13(3), 332-340. https://doi.org/10.1016/j.psychsport.2011.09.001

39. Massey, W. V., Veliz, P. T., Zarrett, N., & Farello, A. (2024). Thriving through sport: The transformative impact on girls’ mental health. Women’s Sport Foundation.

40. McBride, A. M., Johnson, E., Olate, R., & O’Hara, K. (2011). Youth volunteer service as positive youth development in Latin America and the Caribbean. Children and Youth Services Review, 33(1), 34–41. https://doi.org/10.1016/j.childyouth.2010.08.009

41. Nikolina, K., & Đorić, V. (2023). Relations of coach-athlete relationship quality and athlete psychosocial functioning: A systematic review. SportLogia, 19(1). https://doi.org/10.7251/SGIA2319002K

42. Riggio, R. E., Throckmorton, B., & Depaola, S. (1990). Social skills and self-esteem. Personality and Individual Differences, 11(8), 799-804. https://doi.org/10.1016/0191-8869(90)90188-W

43. Salavera, C., Usán, P., & Quilez-Robres, A. (2022). Exploring the effect of parental styles on social skills: The mediating role of affects. International Journal of Environmental Research and Public Health, 19(6), 3295. https://doi.org/10.3390/ijerph19063295

44. Sancassiani, F., Pintus, E., Holte, A., Paulus, P., Moro, M. F., Cossu, G., Angermeyer, M. C., Carta, M. G., & Lindert, J. (2015). Enhancing the emotional and social skills of the youth to promote their wellbeing and positive development: A systematic review of universal school-based randomized controlled trials [Special issue]. Clinical Practice and Epidemiology in Mental Health, 11, 21-40. https://doi.org/10.2174/1745017901511010021

45. Senécal, J., Loughead, T. M., & Bloom, G. A. (2008). A season-long team-building intervention: Examining the effect of team goal setting on cohesion. Journal of Sport and Exercise Psychology, 30(2), 186-199. https://doi.org/10.1123/jsep.30.2.186

46. Sheridan, D., Coffee, P., & Lavallee, D. (2014). A systematic review of social support in youth sport. International Review of Sport and Exercise Psychology, 7(1), 198-228. https://doi.org/10.1080/1750984X.2014.931999

47. Sibthorp, J., & Morgan, C. (2011). Adventure‐based programming: Exemplary youth development practice. New Directions for Youth Development, (130), 105-119. https://doi.org/10.1002/yd.400

48. Smith, A. L., & Ullrich-French, S. (2020). Peers and the sport experience. In G. Tenenbaum & R. C. Eklund (Eds.) Handbook of sport psychology (pp. 410–428). John Wiley & Sons, Inc. https://doi.org/10.1002/9781119568124.ch19

49. Storm, M. K., Williams, K. N., Shetter, E. M., Kaminsky, J., Lowery, C. M., Caldas, S. V., & Winch, P. J. (2017). Sink or swim: Promoting youth development through aquatics programs in Baltimore, Maryland. Journal of Park & Recreation Administration, 35(1), 66-79. https://doi.org/10.18666/JPRA-2017-V35-I1-7295

50. Sullivan, P. J., LaForge-MacKenzie, K., & Marini, M. (2015). Confirmatory factor analysis of the Youth Experiences Survey for Sport (YES-S). Open Journal of Statistics, 5(05), 421. https://doi.org/10.4236/ojs.2015.55044

51. Sung, Y. Y., & Chang, M. (2010). Which social skills predict academic performance of elementary school students. Journal on Educational Psychology, 3(3), 23-34. https://doi.org/10.26634/jpsy.3.3.1078

52. Turman, P. D. (2003). Coaches and cohesion: The impact of coaching techniques on team cohesion in the small group sport setting. Journal of Sport Behaviour, 26(1), 86-104. https://psycnet.apa.org/record/2003-01497-007

53. White, R. L., & Bennie, A. (2015). Resilience in youth sport: A qualitative investigation of gymnastics coach and athlete perceptions. International Journal of Sports Science & Coaching, 10(2-3), 379-393. https://doi.org/10.1260/1747-9541.10.2-3.379

54. Whitley, M. A., Massey, W. V., Camiré, M., Boutet, M., & Borbee, A. (2019). Sport-based youth development interventions in the United States: A systematic review. BMC Public Health, 19, 1-20. https://doi.org/10.1186/s12889-019-6387-z

55. Worker, S. M., Iaccopucci, A. M., Bird, M., & Horowitz, M. (2019). Promoting positive youth development through teenagers-as-teachers programs. Journal of Adolescent Research, 34(1), 30-54. https://doi.org/10.1177/0743558418764089

Appendix A
Supplemental Materials

Table 4

Correlation Matrix of Study Variables

Variables	1	2	3	4	5	6	7
1. Age
2. CART-Q	-0.34*
3. PISQ	0.15	0.23
4. YSEQ	-0.14	0.66**	0.31*
5. Social Skills	0.14	0.62**	0.35*	0.47**
6. Cognitive Skills	-0.05	0.40*	0.16	0.16	0.66**
7. Goal Setting	0.03	0.43**	0.13	0.42**	0.57**	0.70**
8. Initiative	-0.10	0.53**	0.18	0.47**	0.51**	0.40*	0.70**

Notes. * Correlation is significant at the 0.05 level (two-tailed). ** Correlation is significant at the 0.01 level (two-tailed); CART-Q = Coach-Athlete Relationship; PISQ = Parental Involvement, YSEQ = Peer Relationships

Table 5

Regression Results for Perceptions of Goal Setting Skills Gained by Athletes

			95% CI for B
Variable	b	SE	LL	UL	b	p
Intercept	2.053	1.862	-1.731	5.836		0.278
Gender	-0.228	0.271	-0.779	0.322	-0.140	0.405
Age	0.186	0.162	-0.143	0.515	0.196	0.259
Race	0.011	0.062	-0.115	0.137	0.028	0.863
DCART-Q	0.230	0.109	0.008	0.451	0.369	0.042
DPISQ	0.175	0.235	-0.302	0.651	0.124	0.462

Notes. R²= 0.183, p = 0.207; DCART-Q = Change in Coach-Athlete Relationship; DPISQ = Change in Parental Involvement

** When ran independently due to the existing multicollinearity, change to peer cohesion was not a significant predictor of cognitive skills.

Table 6

Regression Results for Perceptions of Initiative Gained by Athletes

			95% CI for B
Variable	b	SE	LL	UL	b	p
Intercept	4.000	1.315	1.328	6.671	-0.120	3.043
Gender	-0.138	0.191	-0.527	0.250	0.062	-0.723
Age	0.042	0.114	-0.191	0.274	0.035	0.365
Race	0.010	0.044	-0.079	0.099	0.389	0.221
DCART-Q	0.171	0.077	0.015	0.327	0.087	2.224
DPISQ	0.086	0.166	-0.251	0.423	-0.120	0.520

Notes. R²= 0.185, p = 0.203; DCART-Q = Change in Coach-Athlete Relationship; DPISQ = Change in Parental Involvement

** When ran independently due to the existing multicollinearity, change to peer cohesion was not a significant predictor of cognitive skills.

Troy Palmer2025-10-01T13:40:31-05:00March 18th, 2026|Leadership, Research, Sport Education, Sport Training, Sports Coaching, Sports Studies and Sports Psychology|Comments Off

Positional Differences and Workload Requirements of a 3-5-2 Formation in Women’s Soccer

Authors: Asher L. Flynn¹, Joanne Spalding², and Sellena Dixon³

¹Department of Sport & Exercise Science, Lincoln Memorial University, Harrogate, TN, USA

²Department of Health Sciences, Georgia College & State University, Milledgeville, GA, USA

³Athletics Department, Lincoln Memorial University, Harrogate, TN, USA

Corresponding Author:

Asher L. Flynn, PhD, CSCS

6965 Cumberland Gap Parkway

[email protected]

423.869.6828

Asher L Flynn, PhD, CSCS is an Assistant Professor of Sport and Exercise Science at Lincoln Memorial University, TN. His research interests focus on fatigue and athlete monitoring in colligate athletes, and aspects of women’s soccer performance.

Joanne Spalding, PhD, is an Assistant Professor in Exercise Science at Georgia College & State University, GA. Her research interests include athlete monitoring and long term athlete development in female athletes.

Sellena Dixon, BS, is the graduate assistant for the women’s soccer team at Lincoln Memorial University, TN.

ABSTRACT

The purpose of this research was to investigated the match demands of 24 female soccer players over one season (15 conference matches) playing in a 3-5-2 formation to determine the match demands and work rates of each position. In order to determine formation specific workload, positions were grouped as center-backs (CB), wingbacks (WB), midfielders (MF), and forwards (FW). Velocity bands used to compare distances and work rates were total distance (TD), 2-3 m/s, 3-4 m/s, 4-5 m/s, 5-6 m/s, and >6 m/s. Results revealed that MF covered significantly more total distance (TD; 10743 ±1520 m) and distance between 2-3 m/s (3444 ± 423 m) than all other positions (CB TD: 8549 ± 1106 m, 2-3: 2497 ± 276 m; WB TD: 8860 ± 1216 m, 2-3: 2324 ± 344 m; FW TD: 8069 ± 1286 m) and greater distance between 3-4 m/s (2515 ± 382m) compared to CB (1574 ± 214 m) and FW (1270 ± 281 m), but not WB (1814 ± 258 m). There were no significant differences between any of the higher velocity bands (>4 m/s). These results can be useful for the coaching staff as descriptive data for the expected distances covered and work rates of Division 2 Women’s soccer teams playing a 3-5-2 formation. These data can be used to determine practice plans in season and for off-season training plans, to prepare athletes for reporting for preseason in appropriate condition.

KEYWORDS: GPS, Sport Science, Athlete Monitoring, Fitness, NCAA

Abbreviations: CB: Center-back, WB: Wingback, MF: Midfield, FW: Forward, GPS: Global Positioning System, MD: Match Day, vVO_2max: Velocity at VO_2max

INTRODUCTION

There is a growing interest in the physical demands of women’s soccer, but the majority of research has focused on the top levels of competition (International and Professional; 23). Although this information is important, it is likely to have limited use for lower levels of competition. It has been reported that “top class” females perform more high intensity running compared to “high-level” females (16, 17, 19). Currently, there is a lack of research investigating match demands at other levels of women’s soccer, especially at lower levels of collegiate soccer in the USA.

The majority of the literature investigating the demands of lower levels of women’s soccer has focused on the match demands of NCAA Division I (2, 5, 12, 20, 21), two articles focused on NCAA Division II athletes (9, 11), and one article focused on NCAA DIII (22), with no studies investigating the match demands of women’s NAIA soccer. Gentles et al (2018) observed that NCAA DII female players covered approximately 5480 m in 45 minutes of match play, while NCAA Division I players covered between 9486m and 9930 m in 90 minutes (20, 22). With little evidence exploring the activity profiles of the lower divisions of collegiate women’s soccer, further investigation is needed.

Many of the studies to date, at any level, have been conducted using only a few matches (3, 17), which, due to the high standard deviations (approximately 10 – 40%; 3, 11, 13, 20, 21-23), raise the question of whether a small number of matches truly reflects the average match demands. Previous studies have also shown differences in match demands based on position (20, 21). On average, defenders covered less distance than midfielders and attackers (forwards), and forwards covered more distance than midfielders (20, 21). Another complication when extrapolating information from current research is that the formation type may also alter the match demands (4, 6, 7). These limitations can lead to complications when inferring information from current research for use in a practical setting with different teams.

Factors that alter expected match demands, level of play and formation used, imply that it would be improper for the coaching staff to use data from a different level of play and unknown formation to determine the expected demands for their specific situation; that is, a high school coach should not be implementing a training plan based on data from college, professional, or international level teams. As such, the purpose of this research was to observe the match demands (distances and work rates) of an NCAA Division II women’s soccer team playing in a 3-5-2 formation, and to determine if there are significant differences in distances covered and/or work rates between positions.

METHODS

Participants

Retrospective data from 24 field players (age: 19.90 ± 1.56 years old; height: 163.43 ± 6.18 cm, weight: 59.35 ± 7.20 kg,) on the same team were included in this study. A total of 194 observations were included in the analysis (center back (CB), n = 49; wingback (WB), n = 36; midfield (MF), n = 61; and forward (FW), n = 48). This study was approved by the institutional review board of Lincoln Memorial University.

Procedures

Match data from an NCAA division II women’s soccer team playing a 3-5-2 formation were collected during a single competitive season. Only conference regular season matches, conference tournament, and national tournament matches were included, with 15 matches used for analysis.

Global Positioning System (GPS) devices (TITAN sports, Houston, TX, USA), sampling at 10 Hz, were used to track player movement duringcompetition. GPS units were activated at least 20-minutes prior to kick-off and were worn in a provided chest halter that secured the device between the shoulder blades under their game uniform. GPS devices were provided to all athletes (starters and substitutes) during the 10-minute window after the team warm-up and before the start of the match, and were collected again after the final whistle. All data from the duration of the match (substitute warm-up and half-time warm-up) were included in the analysis.

For positional analysis, distances accumulated by all athletes who played in a specific position for a match were summed and then divided by the number of positions in the formation. For example, in the forward position, if one athlete was substituted for a portion of the match, the total distance covered by all three athletes playing in the forward position was summed and then divided by two for the two forward positions in the 3-5-2 system. For the work rate analysis, meters covered per minute at each threshold (distances divided by total match time) were averaged for all players in that position (20). Due to the different substitution rules in college soccer, this analysis was deemed optimal to determine the distances and work rate of each position, instead of specific players. Variables of interest included total distance and distances covered in velocity thresholds; 2 – 3 m/s (7.2 – 10.8 km·h^-1), 3 – 4 m/s (10.8 – 14.4 km·h^-1), 4 – 5 m/s (14.4 – 18.0 km·h^-1), 5 – 6 m/s (18 – 21.6 km·h^-1), and >6 m/s (>21.6 km·h^-1).

All matches analyzed were official NCAA matches consisting of two 45-minute halves with a 15-minute half-time period, with a maximum of two 10-minute extra-time periods with a 2-minute intermission, with extra time being stopped in the event of a goal if the competition was tied at the end of the normal 90-minute match.

Data Analyses

Two Repeated Measures ANOVA analyses with Bonferroni corrections were performed to determine whether there was a significant difference in distances and work rates for each position at each threshold (TD, 2 – 3 m/s, 3 – 4 m/s, 4 – 5 m/s, 5 – 6 m/s, and >6 m/s). Statistical analyses were performed using JASP (Version 0.16.2). The alpha level was set at 0.05.

RESULTS

The first RM-ANOVA revealed significantly higher distances (F(3.37, 62.82) = 11.96, p < 0.001) covered in the MF position compared to all other positions for TD and 2 – 3 m/s. Midfielders also covered significantly more distances than CBs and FWs, but not WBs, at 3 – 4 m/s. The only other significant difference observed was between the WB and FW positions in TD covered. There were no significant differences between any other positions at any other threshold. Descriptive statistics are provided in Table 1.

Table 1

Mean distance covered by position in each velocity zone (meters).

	CB	WB	MF	FW
TD	8549 ± 1106 (6467 – 11066)	8860 ± 1216# (5443 – 10854)	10743 ± 1520* (7060 – 12864)	8069 ± 1286# (6204 – 10537)
7.2 – 10.8 km·h-1	2497 ± 276 (1827 – 2972)	2324 ± 344 (1389 – 2842)	3444 ± 423* (2180 – 3916)	2301 ± 400 (1668 – 2942)
10.8 – 14.4 km·h-1	1574 ± 214 ǂ (1191 – 1959)	1814 ± 258 (1068 – 2218)	2515 ± 382# ǂ (1771 – 3129)	1270 ± 281# (759 – 1688)
14.4 – 18.0 km·h-1	607 ± 113 (449 – 769)	878 ± 135 (639 – 1124)	1092 ± 211 (792 – 1453)	587 ± 109 (395 – 734)
18.0 – 21.6 km·h-1	246 ± 58 (155 – 342)	404 ± 86 (252 – 552)	381 ± 79 (241 – 533)	272 ± 60 (149 – 357)
>21.6 km·h-1	96 ± 35 (45 – 175)	200 ± 102 (68 – 426)	120 ± 55 (68 – 295)	180 ± 112 (45 – 530)

Note: Data are presented as mean ± Standard Deviation (Range).
TD: Total distance, CB: Center back, WB: Wingback, MF: Midfield, FW: Forward
* = p < 0.05 compared to all other positions in that velocity range
# = p < 0.05 compared to the other indicated positions in that velocity range
ǂ = p < 0.05 compared to the other indicated positions in that velocity range

The RM-ANOVA performed to determine differences in work rates at each threshold revealed significantly higher work rates (F(3.47, 64.82) = 6.05, p < 0.001) over the entire match (TD) for the MF position compared to all other positions, and a significantly higher work rate from the MF position in the 3 – 4 m/s velocity range compared to the FW position. There were no other significant differences between any of the other positions at any other threshold. The results are presented in Table 2.

Table 2

Mean work rate by position in each velocity zone (meters per minute).

	CB	WB	MF	FW
TD	96 ± 8 (88 – 123)	101 ± 7 (85 – 117)	119 ± 20 * (88 – 167)	101 ± 24 (72 – 156)
7.2 – 10.8 km·h^-1	28 ± 2 (25 – 33)	26 ± 3 (21 – 35)	36 ± 3 (29 – 40)	27 ± 5 (19 – 35)
10.8 – 14.4 km·h^-1	18 ± 2 (14 – 21)	21 ± 2 (18 – 26)	27 ± 3 # (21 – 32)	16 ± 4 # (8 – 21)
14.4 – 18.0 km·h^-1	7 ± 1 (5 – 10)	10 ± 1 (8 – 12)	12 ± 3 (9 – 19)	8 ± 3 (4 – 14)
18.0 – 21.6 km·h^-1	3 ± 1 (2 – 4)	5 ± 1 (4 – 6)	4 ± 1 (2 – 7)	4 ± 2 (2 – 9)
>21.6 km·h-1	1.3 ± 1 (1 – 3)	2.3 ± 1 (1 – 4)	1.3 ± 1 (1 – 3)	3.4 ± 4 (1 – 13)

Note. Data are presented as mean ± Standard Deviation (Range).
TD: Total distance, CB: Center back, WB: Wingback, MF: Midfield, FW: Forward

* = p < 0.05 compared to all other positions in that velocity range
# = p < 0.05 compared to the other indicated positions in that velocity range

DISCUSSION

The purpose of this study was to examine the external demands of NCAA DII women’s college soccer playing in a 3-5-2 formation. The most interesting finding from these data was that the WB position was only significantly higher in total distance covered compared to the FW position (p = 0.044), but there were no other significant differences in distances or work rates compared to other positions. This was interesting, given that the WB position is commonly accepted as the most demanding position. Another interesting result was that the MF position had significantly higher distances at lower speeds (2-3 m/s, p < 0.001; 3-4 m/s, p < 0.001) only when compared to FW position. Other studies have reported that different positions have significantly different total distances covered in a match (1, 14). Abbot et al.(2018) reported that central midfielders covered the greatest distance (11,570 ± 469 m), followed by wide attackers (10,918 ± 353 m), wide defenders (10,747 ± 420 m), strikers (10,320 ± 420 m), and central defenders covering the least amount of distance (9,830 ± 428 m), while Lago-Peñas (2009) reported significant differences between nearly every position for each threshold, except for total distance (11.1 – 14 km·h^-1, 14.1 – 19 km·h^-1, 19.1 – 23 km·h^-1, > 23 km·h^-1). Both these studies observed high-level male soccer teams (U23 English Premier League, Professional Spanish Premier League) and as such they would not be expected to mimic the results of this study and highlight the importance of sex- and level-specific research.

In a more direct comparison with other women’s college soccer research, Sausaman et al(2019) and Alaxander et al (2014) reported significant differences in distances covered by position, whereas Corrales (2020) reported no differences, regardless of position. Sausaman et al (2019) reported that attackers covered more high-speed (> 15 km·h^-1) and sprint (> 18 km·h^-1) distances than midfielders and defenders, with no difference between midfielders and defenders. Alexander et al (2104) reported that central defenders covered the least amount of total distance (8041.2 ± 371.0 m), followed by central attacking midfielders (9236.1 ± 491.3 m), fullbacks (9306.2 ± 367.8 m), and wide midfielders (9500.4 ± 847.0 m), with central defensive midfielders (9947.4 ± 577.9 m) covering the greatest amount of total distance. Fullbacks (1321.5 ± 173.7 m) and wide midfielders (1208.2 ± 314.1 m) covered significantly more distance at high speed (> 15 km·h^-1) compared to central defensive midfielders (847.7 ± 234.9 m), central attacking midfielders (747.64 ± 196.5 m), and central defenders (614.1 ± 98.9 m). The difference in results between these studies and the current investigation could be due to a different level of competition (NCAA Division 1), a possible difference in formation (not reported), or due to the current studies banding velocity zones (i.e. 4 -5 m/s, 5 – 6 m/s) instead of summing distances above thresholds (> 15 km·h^-1).

CONCLUSION

This present study provides information on the expected work and work rates of division 2 women’s soccer. Data analysis revealed minimal differences based on position, with the midfield position being the only position with significant differences and only at low intensity thresholds (TD, 2-3 m/s). All other positions and intensities were not statistically different, highlighting the possibility that training for these positions likely does not need to be modified to fit each position but rather each athlete.

APPLICATIONS IN SPORT

The primary application of this research is to allow the coaching staff to determine the appropriate fitness, conditioning, and practice workloads for their team with respect to their level of competition, formation, and style of play. Using typical tactical periodization plans for match-day preparation (Match day (MD) +1, MD -2, MD -1, etc.), position-specific workloads can be determined and monitored to ensure optimal loading during each practice session. This information can also be used to determine fitness testing requirements. Since there was a significant drop (43.7%) in distance covered and work rate (approximately 43% decrease) at intensities above 4 m/s (14.4 km·h^-1) observed from this study, minimum criteria for aerobic fitness tests could be set at 15.5 km·h^-1, which would allow players to do the majority of the work expected below their estimated lactate threshold (85% vVO2max; 8, 18)

In addition, these data can be used to determine the appropriate time requirements for different conditioning drills. For example, making a time criterion of 3:40 for an 800 m run would meet the Long Interval definition for an individual with a vVO2max of 15.5 km·h^-1, but if a team had higher/lower requirements, adjusting time cut offs would be suggested (15). These data can also be used to create game-specific conditioning drills, such as creating an interval training exercise (100m active running, 100m recovery jog; Table 3) that would provide about 30 – 35% of game distances at the higher end of expected game work rates.

Table 3

Interval conditioning exercise.

Speed Level	Reps	Time	Work Rate	Accumulated Distance
8.0 km·h^-1	22	45s	89 m/min	2200
12.0 km·h^-1	10	30s	36 m/min	1000
15.0 km·h^-1	5	24s	20 m/min	500
20.0 km·h^-1	2	18s	8 m/min	200
>21.6 km·h^-1	1	<16s	4 m/min	100

Note: Work Rate was calculated as the total distance covered in each velocity threshold divided by the total exercise time (24.5 min).

When retroactively analyzing team distances and work rates to create a training plan, it is important to note that since the majority of the total distance covered during a match is at low intensities (<3 m/s; 11), focusing on “running” the observed TD is likely unnecessary. Lower speed distances (<4 m/s) would be expected to be accumulated through daily technical drills and exercises. Higher speed distances (>4 m/s) could be accumulated in any manner chosen by the coach, such as a mixture of soccer technical drills and/or conditioning drills.

When using workload data in this manner, this would allow the coaching staff to create training plans that develop physical characteristics in a manner appropriate to the athletes level and expected match play requirements instead of arbitrarily spending time and effort developing a specific characteristic beyond projected usefulness. For example, since the majority of work is performed below 14.4 kph, spending the time and training effort for a vVO2max above 16 kph would be counterproductive. The extra focus could be better spent on improving other training targets to improve performance (technical, tactical, sprint, acceleration, change of direction)

ACKNOWLEDGMENTS

The authors declare no conflicts of interest, and no funding was received for this research.

REFERENCES

1. Abbott, W., Brickley, G., & Smeeton, N. J. (2018). Positional differences in GPS outputs and perceived exertion during soccer training games and competition. The Journal of Strength & Conditioning Research, 32(11), 3222-3231.

2. Alexander, RP. (2014) Physical and Technical Demands of Women’s Collegiate Soccer (Publication No. 2421). [Doctoral dissertation, East Tennessee State University], Digital Commons.

3. Andersson, H. Å., Randers, M. B., Heiner-Møller, A., Krustrup, P., & Mohr, M. (2010). Elite female soccer players perform more high-intensity running when playing in international games compared with domestic league games. The Journal of Strength & Conditioning Research, 24(4), 912-919.

4. Aquino, R., Vieira, L. H. P., Carling, C., Martins, G. H., Alves, I. S., & Puggina, E. F. (2017). Effects of competitive standard, team formation and playing position on match running performance of Brazilian professional soccer players. International Journal of Performance Analysis in Sport, 17(5), 695-705.

5. Bozzini, B. N., McFadden, B. A., Walker, A. J., & Arent, S. M. (2020). Varying demands and quality of play between in-conference and out-of-conference games in division i collegiate women’s soccer. The Journal of Strength & Conditioning Research, 34(12), 3364-3368.

6. Bradley, P. S., Carling, C., Archer, D., Roberts, J., Dodds, A., Di Mascio, M., Paul, D., Gomez Diaz, A., Peart D., & Krustrup, P. (2011). The effect of playing formation on high-intensity running and technical profiles in English FA Premier League soccer matches. Journal of sports sciences, 29(8), 821-830.

7. Carling, C., Espié, V., Le Gall, F., Bloomfield, J., & Jullien, H. (2010). Work-rate of substitutes in elite soccer: A preliminary study. Journal of science and medicine in sport, 13(2), 253-255.

8. Cerezuela-Espejo, V., Courel-Ibáñez, J., Morán-Navarro, R., Martínez-Cava, A., & Pallarés, J. G. (2018). The relationship between lactate and ventilatory thresholds in runners: validity and reliability of exercise test performance parameters. Frontiers in physiology, 9, 1320.

9. Choice, E. E., Tufano, J. J., Jagger, K. L., & Cochrane-Snyman, K. C. (2022). Match-play external load and internal load in NCAA division II women’s soccer. The Journal of Strength & Conditioning Research, 10-1519.

10. Corrales, I. (2020). The Physical Demands of Women’s Collegiate Soccer Matches Assessed Using GPS Devices [master’s thesis, California State University, Fullerton], https://scholarworks.calstate.edu/.

11. Gentles, J. A., Coniglio, C. L., Besemer, M. M., Morgan, J. M., & Mahnken, M. T. (2018). The demands of a women’s college soccer season. Sports, 6(1), 16.

12. Ishida, A., Travis, S. K., Draper, G., White, J. B., & Stone, M. H. (2022). Player position affects relationship between internal and external training loads during Division I collegiate female soccer season. The Journal of Strength & Conditioning Research, 36(2), 513-517.

13. Jagim, A. R., Murphy, J., Schaefer, A. Q., Askow, A. T., Luedke, J. A., Erickson, J. L., & Jones, M. T. (2020). Match demands of women’s collegiate soccer. Sports, 8(6), 87.

14. Lago-Peñas, C., Rey, E., Lago-Ballesteros, J., Casais, L., & Dominguez, E. (2009). Analysis of work-rate in soccer according to playing positions. International Journal of Performance Analysis in Sport, 9(2), 218-227.

15. Laursen, P., & Buchheit, M. (2019). Science and application of high-intensity interval training. Human kinetics.

16. Martínez-Lagunas, V., Niessen, M., & Hartmann, U. (2014). Women’s football: Player characteristics and demands of the game. Journal of Sport and Health Science, 3(4), 258-272.

17. Mohr, M., Krustrup, P., Andersson, H., Kirkendal, D., & Bangsbo, J. (2008). Match activities of elite women soccer players at different performance levels. The Journal of Strength & Conditioning Research, 22(2), 341-349..

18. Parpa, K., & Michaelides, M. A. (2025). Ventilatory thresholds in professional female soccer players. International Journal of Sports Medicine, 46(02), 97-103.

19. Ramos, G. P., Nakamura, F. Y., Penna, E. M., Wilke, C. F., Pereira, L. A., Loturco, I., Capelli, L., Mahseredjian F., Silami-Garcia E., & Coimbra, C. C. (2019). Activity profiles in U17, U20, and senior women’s Brazilian national soccer teams during international competitions: are there meaningful differences?. The Journal of Strength & Conditioning Research, 33(12), 3414-3422.

20. Sausaman, R. W., Sams, M. L., Mizuguchi, S., DeWeese, B. H., & Stone, M. H. (2019). The physical demands of NCAA division I women’s college soccer. Journal of Functional Morphology and Kinesiology, 4(4), 73.

21. Vescovi, J. D. (2014). Motion characteristics of youth women soccer matches: female athletes in motion (FAiM) study. International journal of sports medicine, 35(02), 110-117.

22. Vescovi, J. D. (2015). Physical demands of regular season and playoff matches in professional women’s soccer: a pilot from the Female Athletes in Motion (FAiM) study. In International research in science and soccer II (pp. 95-106). Routledge.

23. Vescovi, J. D., & Falenchuk, O. (2019). Contextual factors on physical demands in professional women’s soccer: female athletes in motion study. European journal of sport science, 19(2), 141-146.

Troy Palmer2025-09-30T16:08:05-05:00March 4th, 2026|Research, Sport Education, Sport Training, Sports Coaching, Women and Sports|Comments Off