(05) ROOKIE ANTHROPOMETRIC AND PERFORMANCE DIFFERENCES BY DRAFT ROUND USING THE 2022 NATIONAL FOOTBALL LEAGUE (NFL) COMBINE

The National Football League (NFL) conducts an annual combine to assess anthropometric and athletic ability of hundreds of collegiate athletes in a variety of tests in preparation for the draft. The growth in sport analytics can potentially help teams more accurately predict which athletes to draft to aid team performance for the following season. Purpose: The purpose of this study was to determine potential differences in average normalized anthropometric and performance results of the athletes invited to the 2022 NFL Combine compared to subsequent draft status. Methods: Data from 3 anthropometric (Body Mass Index; Hand Size; Arm Length) and 5 performance tests (40 Yard Sprint; Vertical Jump; Broad Jump; 3-Cone Drill; 20 Yard Shuttle) of 315 collegiate athletes were analyzed from open-source databases. Data from completed tests were normalized (Z-scores) to create average anthropometric (AvgAZ), performance (AvgPZ), and total (AvgTZ) based Z-Scores. Players were also assigned a number (1-8) based on their draft round, with undrafted players assigned a value of 8. A One-way ANOVA (α = 0.05) was conducted between AvgAZ, AvgPZ, and AvgTZ and subsequent Draft Round in the 2022 NFL Draft. Additionally, players were separated into positions (Offensive Line [OL]; Defensive Back [DB]; Defensive Line [DL]; Running Back [RB]; Linebacker [LB]; Quarterback [QB]; Tight End [TE]; Wide Receiver [WR]) and analyzed for potential differences within each position group via One-way ANOVAs. Post-hoc analyses on significant omnibus findings were analyzed with an LSD correction factor. Results: A significant omnibus result for all 315 athletes was noted for AvgAZ [F(7, 307) = 2.37, p = 0.03, η² = 0.06] and AvgTZ [F(7, 307) = 2.67, p = 0.03, η² = 0.06]. Post-hoc analyses for AvgAZ noted that athletes drafted in Round 1 had significantly greater scores compared to Round 2 (p = 0.04; Mean Difference [95%CI]: 0.42au [0.01, 0.82]) and Undrafted players (p < 0.01; 0.48au [0.15, 0.81]). Undrafted athletes were also significantly worse than Round 3 (p = 0.02; -0.37au [-0.07, -0.67]) and Round 5 (p < 0.01; -0.46au [-0.13, -0.79]) players. For AvgTZ, post-hoc analyses revealed that Undrafted players were significantly worse than Round 1 (p < 0.01; -0.32au [-0.12, -0.52]), Round 2 (p < 0.01; -0.29au [-0.09, -0.49]), Round 3 (p = 0.01; -0.24au [-0.06, -0.43]), Round 4 (p = 0.02; -0.25au [-0.05, -0.44]), and Round 5 (p < 0.01; -0.27au [-0.08, -0.48]) athletes. No significant difference was noted for AvgPZ [F(7, 307) = 1.79, p = 0.09, η² = 0.05].  When separated by position group, significant differences were noted in AvgAZ (QB), AvgPZ (DB, DL, LB), and AvgTZ (RB, LB). No differences were noted for OL, TE, or WR. Conclusions: The findings suggest that higher drafted players are larger anthropometrically and have a better overall NFL Combine performance compared to undrafted players. However, these differences are relatively small when measurements are normalized. Additionally, these are expanded to other rounds in some cases when normalized to specific positions. Practical Applications: These findings further support the use of normalized Z-scores to supplement team and scout assessments to determine an athlete's draft status. However, due to the varied relationships when separated by position group, it is theorized that the NFL and teams should reconsider what is measured at the NFL Combine to better the evaluation process of comparing athletes for draft selection.

Acknowledgements: None