Relationship Between the National Football League (NFL) Combine Measurables and Playing Time in the 2024 NFL Rookie Class

ABSTRACT

Purpose: This study investigated the relationship between anthropometric and performance measures collected at the 2024 National Football League (NFL) Combine and playing time (PT) during the 2024 NFL regular season. Methods: Data from four anthropometric (Body Mass Index; Hand Size; Arm Length; Wingspan) and seven performance tests (40-yard Dash; 10-yard Split; Vertical Jump; Broad Jump; 3-Cone Drill; 20-Yard Shuttle; 225lb Bench Press) of 315 players were standardized into average Anthropometric Z-Scores (AZ), Performance Z-Scores (PZ) and Total Z-Scores (TZ) for analyses. PT was calculated as a player’s total number of regular season snaps during their 2024 rookie season. Pearson correlations were used to investigate the relationships (α = 0.05) between AZ, PZ, and TZ to PT. Players were also analyzed for potential relationships within each position group. Results: A significant, weak, positive correlation existed between PZ and PT (r = 0.19, p < 0.01) and TZ and PT (r = 0.20, p < 0.01) for all players. No relationship existed for AZ and PT (r = 0.02; p = 0.73). Additionally, significant relationships existed among: Offensive Line  – PZ and PT (r = 0.33, p = 0.01) and TZ and PT (r = 0.35, p < 0.01); Wide Receiver – PZ and PT (r = 0.39, p = 0.03) and TZ and PT (r = 0.46, p < 0.01); Linebacker – TZ and PT (r = 0.39, p = 0.05). Conclusions: NFL Combine performance metrics may provide insight on PT, but anthropometric measurables were not related to PT. The lack of relationship within position groups indicates the NFL Combine may not be valuable in evaluating a rookie’s success on the field. Applications in Sport: Professionals who work with prospects may choose to train Combine specific techniques to maximize a prospect’s chances of playing in the NFL. However, individualized training that focuses on position specific demands or weaknesses that are not directly measured by NFL Combine tests may be more useful in increasing PT. The NFL Combine may be a useful supplement to all factors that go into an NFL team’s decision to draft a player.

Key Words: performance testing, predictive analytics, scouting, correlational analysis, American football

INTRODUCTION

The National Football League (NFL) hosts an annual Scouting Combine in Indianapolis, Indiana of elite college football players. Only about 3% of college football players are invited to the NFL Combine and therefore represent those with the highest chance of being drafted into the NFL (4). The purpose of the NFL Combine is to allow coaches, scouts, and other team personnel representing the 32 NFL teams the opportunity to assess hundreds of players from all divisions of collegiate football.

Football has position-specific skills that are needed to excel at the highest level. However, there are similarities between each position. All positions need vertical and horizontal power, agility, and strength. During this weeklong event, players participate in a multitude of tests. These tests include anthropometric measurements (Height; Weight; Wingspan; Arm Length; Hand Size) and performance tests (40-yard dash; 10-yard split; Vertical Jump; Broad Jump; 3 Cone Drill; 20 Yard Shuttle; 225lb Bench Press). All the events in the NFL Combine have been shown to have face validity (4). NFL player personnel departments use the NFL Combine data as part of their criteria to determine whether to select a player in the upcoming NFL Draft.

While the NFL Combine tests are designed to determine that aptitude to play at the next level, research is conflicted on the ultimate usefulness of the NFL Combine in determining player performance and playing time (PT). Kuzmits and Adams (4) found no consistent significant relationship between NFL Combine tests and player performance during the years of 1999 to 2004. Research also noted that the NFL Combine from 2013 to 2015 lacked the ability to predict game performance when specifically analyzing first year game performance (3). Teramoto, Cross, and Willick (12) looked at whether the NFL Combine could predict future performance of Running Backs (RB) and Wide Receivers (WR). The results of this study were that the time on 10-yard split was the most important predictor of yards per attempt for RB while vertical jump was significantly associated with receiving yards per reception for WR. However, the measures cannot explain a large part of the variance in the future performance of RBs and WRs. Vincent et al. (15) looked at NFL Combine participants from 2005 to 2010 who then played in the NFL. Significant relationships were found between at least one NFL Combine measure and on-field success. Even though significant relationships were found the authors stated that the NFL Combine tests are only modest predictors of future performance. More recently, investigation of six physical skill tests at the NFL Combine to try and predict draft placement in the 2022 NFL Draft and showed no significant difference between drafted and nondrafted players in any of the six physical tests analyzed (14).

LaPlaca and McCullick (5) built on previous research looked at player performance from the years 2006 to 2018 and compared it to the NFL Combine from 2006 to 2016. They found that every position group, both offensive and defensive, had at least one NFL Combine test that was significantly correlated with player performance. The study made sure to disclose that even though they found significant correlations, the large sample size made it easier to find weaker correlations. A limitation that was discussed was that while the authors did use objective performance statistics such as Touchdowns scored, they also used a grading system through Pro Football Focus to determine player performance. This grading system was not purely objective because the grades are determined by multiple reviewers through the observation of game film. Therefore, the overall performance of each player was not entirely objective. Additionally, a robust study by Frank and colleagues (2) analyzed 20 years (2000-2018) of NFL Combine data and noted that for offensive positions, single measures often best predicted success, while various combinations of NFL Combine performance traits predicted success among defensive players. This study also suggested that NFL Combine data is best used in conjunction with scouting and personnel departments to supplement NFL draft decision making. Similarly, research was conducted looking at the impact of the NFL Combine on five-year performance data from the 2013-2017 NFL seasons and concluded that the NFL Combine lacked predictive ability during that timeframe (1). While historical research does exist in this field, each year provides another opportunity to determine the NFL Combine’s effectiveness in predicting success. Additionally, limited research exists discussing the relationship between NFL Combine Measurables and PT for first-year players. The primary purpose of this study was to determine if the anthropometric and performance measures of the athletes invited to the 2024 NFL Combine were related to PT during the 2024 NFL regular season.

METHODS

Participants

Participants for the data analysis in this study were college football players that participated in the 2024 NFL Combine (N = 315). Participants were also grouped by position for use of positional comparisons (Offensive Line [OL] (N = 70); Defensive Back [DB] (N = 67); Defensive Line [DL] (N = 50); Running Back [RB] (N = 29); Linebacker [LB] (N = 30); Quarterback [QB] (N = 14); Tight End [TE] (N = 16); Wide Receiver [WR] (N = 39)). All player positions were input based off their official designation at the time of the NFL Combine. Due to limited sample size (N = 6) and variations in specializations, NFL Combine athletes who were labeled Specialist (Kicker, Punter, Long Snapper) were excluded from analyses.

Procedures

Four anthropometric (Body Mass Index [BMI]; Hand Size; Arm Length; Wingspan) and seven performance measures (40-yard Dash; 10-yard Split; Vertical Jump; Broad Jump; 3-Cone Drill; 20-Yard Shuttle; 225lb Bench Press) were analyzed. BMI was calculated by the researchers using Height and Weight measurements taken at the NFL Combine. Full descriptions of the performance tests have been detailed previously by McShay (7).

The data from the NFL Combine was obtained from NFL.com/combine/tracker (8). Each participant’s scores were retrieved for every test that was completed. Standardization of data, via Z-scores, were created for every anthropometric and performance measure. The measures from the NFL Combine were standardized into averages for each player, taking each player’s combined Z-Score score and dividing by the number of NFL Combine events they participated in to account for players who did not complete every NFL Combine event. Standardized averages were created for Anthropometric Z-scores (AZ), consisting of the four anthropometric measures, Performance Z-scores (PZ), consisting of the seven performance measures, and Total Z-scores (TZ), consisting of all 11 NFL Combine measures, for analyses This method of standardization of NFL Combine data into Z-scores for analysis has previously been supported (1).

Once all NFL Combine data was standardized, researchers used Pro-football-reference.com (9) to retrieve offensive, defensive, and special teams snaps for each player during the 2024 NFL regular season. Each player’s total snap count was then combined to provide a single value to determine PT, which was used for analysis. Because of this study only requiring secondary analysis of data which is publicly available on web-based domains, which do not disclose individual’s health information, Institutional Review Board approval was not required, though the study was approved by the research institution.

Data Analyses

Pearson product moment correlations, using Statistical Product and Service Solutions (SPSS, v29.0, IBM Corporation, Armonk, NY), were used to determine the relationship (α = 0.05) between AZ, PZ, TZ to PT. Additionally, players were separated by position and Pearson product moment correlations (α = 0.05) were used to determine potential relationships within each group between AZ, PZ, and TZ, to PT. All data are presented as means ± standard deviation with 95% confidence intervals (95%CI) unless otherwise stated.

RESULTS

Descriptive Statistics

Of the 321 athletes whose data were collected, 315 were used for analysis. A total of six athletes were excluded from analysis due to their position of Specialist (punter, kicker, long snapper) because only anthropometric data was collected on this group. Of the 315 athletes used for analysis, 312 (99%) completed all anthropometric measurements. There was more variability in the performance testing, with 25 (8%) completing all seven performance events, and 263 (83.5%) completing at least one performance event. When broken down by event, 220 (69.8%) completed the 40yd (4.73 ± 0.31s) with a 10yd split (1.63 ± 0.11s), 227 (72.1%) completed the VJ (34.0 ± 4.3in), 220 (69.8%) completed the BJ (117.9 ± 9.0in), 78 (24.8%) completed the 3C (7.30 ± 0.40s), 89 (28.3%) completed the PRO (4.44 ± 0.28s), and 100 (31.8%) completed the BP (21.9 ± 5.6reps). When examining snaps played over the 2024 regular season, 239 (75.9%) players went on to play at least one snap, with 224 (71.1%) averaging more than one snap per game over the course of the season.

Anthropometric Correlation Analysis

The results of the correlation analysis for AZ and PT are presented in Figure 1. Pearson product moment correlation coefficients were calculated for the relationship between average AZ and PT for all players and separated by position group. No significant overall relationship existed for AZ and PT (n = 312; r = 0.02; p = 0.73).

Additionally, no significant relationships existed among position groups: OL (n = 70; r = 0.13; p = 0.29); RB (n = 29; r = 0.21; p = 0.29); WR (n = 37; r = 0.24; p = 0.16); TE (n = 16; r = 0.39; p = 0.14); QB (n = 13; r = 0.02; p = 0.95); DL (n = 50; r = -0.10; p = 0.52); LB (n = 30; r = 0.19; p = 0.33); DB (n = 67; r = -0.02; p = 0.89).

  Performance Correlation Analysis

The results of the correlation analysis for PZ and PT are presented in Figure 2. Pearson product moment correlation coefficients were calculated for the relationship between average PZ and PT for all players and separated by position group. A significant, weak, positive correlation existed between PZ and PT (n = 263; r = 0.19, 95%CI [0.07, 0.31]; p < 0.01). The positive direction of this relationship indicates that players who performed better at the NFL Combine played more snaps during the 2024 NFL regular season.

When separated by position groups, significant, positive relationships existed for the following groups: OL (n = 61; r = 0.33, 95%CI [0.09, 0.54]; p = 0.01); WR (n = 34; r = 0.39, 95%CI [0.06, 0.65]; p = 0.03). The positive direction of these relationships indicates that OL and WR who performed better at the NFL Combine accumulated more snaps during the 2024 NFL Regular season. No significant correlations were noted for: RB (n = 25; r = 0.31; p = 0.14); TE (n = 12; r = 0.07; p = 0.15); QB (n = 7; r = -0.39; p = 0.40); DL (n = 43; r = 0.30; p = 0.06); LB (n = 26; r = 0.31; p = 0.13); DB (n = 55; r = -0.02; p = 0.89).

Total Correlation Analysis

The results of the correlation analysis for TZ and PT are presented in Figure 3. Pearson product moment correlation coefficients were calculated for the relationship between average TZ and PT for all players and separated by position group. A significant, weak, positive correlation existed between TZ and PT (r = 0.20, 95%CI [0.08, 0.31]; p < 0.01) for all players. The positive direction of this relationship indicates that players who had higher average TZ scores played more snaps in the 2024 NFL regular season.

When separated by position groups, significant, positive relationships existed for the following groups: OL (n = 61; r = 0.35, 95%CI [0.11, 0.56]; p < 0.01); WR (n = 34; r = 0.46, 95%CI [0.15, 0.69]; p < 0.01); LB (n = 26; r = 0.39, 95%CI [0.01, 0.68]; p = 0.05). The positive direction of these relationships indicates that players in these position groups who had higher average AZ scores played more snaps in the 2024 NFL Regular season. No significant correlations were noted for: RB (n = 25; r = 0.31; p = 0.14); TE (n = 12; r = 0.07; p = 0.85); QB (n = 7; r = -0.24; p = 0.61); DL (n = 43; r = 0.30; p = 0.06); DB (n = 55; r = 0.06; p = 0.70).

Discussion

The main finding of this study is that PZ and TZ may have a weak relationship to PT in a player’s first year in the NFL. There was no relationship between a player’s AZ and subsequent PT across all athletes nor when separated by position group. The study did find a significant weak positive correlation between average PZ and PT for all players. However, when separated by position groups significant, positive relationships existed for OL and WR. Finally, there was a significant weak positive correlation between TZ and PT for all players. When separated by position groups, significant, positive relationships existed for OL, WR, and LB.

There could be many reasons why these relationships exist for WR, LB, and OL. Previous movement analysis research for NFL players by position found that WR had highest in-game velocity and highest total running volume by an offensive position (6). Therefore, the 40-yard dash and 10-yard split may carry more importance among WR. The same study showed that LB had the most high-velocity efforts and high-velocity distance in game compared to all other positions. LB also showed the largest variability across player-games which is likely due to the roles that LB perform which include rushing the QB, play in space and cover offensive players, or primarily to tackle an opponent. Additionally, OL noted a positive relationship in the current study, with better NFL Combine performances leading to more PT.  While previous research (11) has noted that OL have worse NFL Combine values compared to other positions, the nature of the OL position may lend itself to more direct relationships from NFL Combine performance, since these athletes require multidirectional power over limited space. The positional findings in the current study do support previous research that noted relationships between NFL Combine performance metrics and PT among WR (40-yard Dash, Vertical Jump), LB (40-yard Dash, 20-yd Shuttle) and OL (20-yard Shuttle, Vertical Jump) (1, 2).

The NFL is not the only sport that uses a combine to test and evaluate future players’ abilities. Teramoto et al. (13), investigated the National Basketball Association (NBA) scouting Combine to determine whether the NBA Combine could predict PT. The study showed that the NBA Combine metrics had minimal correlation with long-term performance. In the NBA, it was found that certain anthropometrics had slightly better predictive power than athletic tests, which contrasts with what researchers found about the 2024 NFL Combine. Both in the NFL and NBA Combine researchers have proposed that performance in college or in game is the biggest predictor of draft position and future performance (11, 13).

There are limitations associated with this study. As reported in the results only 25 (8%) of all prospects completed all seven performance events. Increasingly, players are opting out of some or all the NFL Combine process, due to injury concern, agent decision, recovering from an injury during the season, or to focus on performing well at individual workouts, where more variables can be controlled by that athlete. In the season being analyzed in this study, five of the first six picks in the NFL Draft did not participate in the NFL Combine process, which could impact these findings. A larger, more complete sample from all NFL Combine athletes would comprise a better representation of their athletic performance. Finally, players that played zero snaps their first year due to injury were included in analysis, due to limitations among researchers to determine the extent of every injury or whether a player was not on the field due to injury or coaching decisions. A player that may have had strong AZ, PZ, and TZ scores, but did not play during their rookie season because of injury, which would have impacted the relationship between those variables and PT.

CONCLUSIONS

Many studies have been conducted over the last 20 years to determine if and how NFL Combine measurables can predict performance in the NFL (1-6, 10, 12, 14, 15). These studies have found mostly found minimal relationships overall, though stronger relationships among certain position groups. Despite the general scientific consensus that the NFL Combine is not a strong predictor of future NFL success, a multitude of NFL Combine “prep courses” exist, with athletes paying for training specifically to improve in NFL Combine measurables. There has been scientific skepticism about these courses and their impact on performance at the NFL Combine and its translation to improved draft status or playing time. While these courses claim that they will improve an athlete’s chance of getting drafted, there is currently no scientific evidence to these claims (1, 4, 10). Training programs that focus on a prospect’s position specific demands or individual weaknesses that are not directly measured by NFL Combine tests may be more useful in increasing PT for that athlete. The results of the current study support the previous work in the literature, but do note that some position groups (OL, WR, LB) may benefit by improving NFL Combine-specific performance in the lead up to the NFL Combine and Draft.

APPLICATIONS IN SPORT

The results from the current study suggest PT among NFL rookies during the 2024 regular season could not be strongly predicted with data collected during the NFL Combine. However, due to the relationships that were found, specifically withing certain position groups, it may be important for athletes in those positions to train specifically for those performance tests to have a better chance at playing in their first year. The data can be important for NFL player personnel departments who may use data collected during the NFL Combine to influence drafting decisions. Due to the significant, but variable, nature of the relationships found in the current study, it appears that the NFL Combine may be a useful supplement to scouting, film analysis, interviews, and other factors that go into an NFL team’s decision to draft a player. However, it is apparent that there is more to determining PT during a rookie season than just superlative measurables collected during the NFL Combine.