Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Exercise Science

First Advisor

Dr. Mark Abel


BACKGROUND: Athlete evaluation in American football is dominated by the evaluation characteristics that underly key performance indicators (KPI’s) in the physical domain of performance. Team sport athlete performance is comprised of a combination of attributes in the physical, technical, tactical, and mental domains of performance. Using only KPI’s in the physical domain of performance to predict competition performance of American football athletes is challenging due to the differences in performance outcomes of the varying positions in American football and absence of the other performance dimensions that are not accounted for. Advances in athlete assessment have revolutionized the process of evaluating competition performance by assigning athletes standardized competition grades that enable analysis across position groups. However, the efficacy of these systems to predict competition performance and participation has yet to be empirically determined. Therefore, the purpose of this study was to determine if performance domains predicted competition performance and participation. METHODS: A multi-level multivariate model was utilized to investigate the consistency of competition performance outcome measures and the relationship between athlete characteristics and competition performance of American football players with Hierarchical Linear Modeling (HLM) software. Data from Division I collegiate American football players (n = 65) during two seasons, totaling 24 competition events, were used in this analysis. The performance outcome measures included modified season Pro Football Focus (PFF) grades, and the percentage of plays participated in by an athlete during a season. Evaluations of athlete characteristics in the physical, mental, technical, and tactical domains of performance were assessed on a 1-5 Likert type scale and averaged to provide a composite score in each domain of performance. RESULTS: Competition outcomes (x̄ ± SD) for the sample included: Play Participation = 55.2 ± 25.4%; Season PFF Grade 63.5 ± 4.87 (0-100). Performance domain composite scores (x̄ ± SD) for the sample included: Physical = 3.43 ± 0.58; Technical = 3.21 ± 0.65; Tactical = 3.37 ± 0.88; Mental = 3.21 ± 0.80. The results indicated that there was robust inconsistency in competition performance outcomes and the inconsistency was not related to athlete characteristics in the four performance domains (r = 0.077 to r = 0.030). However, the results indicated that the composite mental performance domain score was universally important to competition participation (p = 0.053) and competition performance (p = 0.040). Other results indicated that the composite tactical performance domain score (p = 0.035) was a significant predictor of athlete participation. Physical, mental, technical, and tactical composite scores explained 3% (= 0.025) of the variance in modified season PFF grades and 18% (= 0.18) of the variance in athlete participation. CONCLUSION: This study provides a framework for future analysis of KPI’s in American football. The methodology suggests that athlete characteristics in the mental performance domains of performance may be universally important to competition performance and participation. It also provides evidence to suggest the underlying mechanisms of competition performance are multifaceted, and coaches and performance specialists should consider evaluating athletes in more than just the physical domain of performance in the future. However, all domains of performance were not generally related to competition performance and participation.

Digital Object Identifier (DOI)