Comparison of allometric scaling methods for normalizing strength, power, and speeed in American football players

Assessments of physical abilities are consistently and systematically done in American football in order to determine progress and effectiveness of conditioning programs. Typically, such assessment are recorded as absolute values without regard to the influence of body mass (BM) on the performance variables.

Test results were collected on NCAA football players (mean + SD: age = 20.1+1.3, mass=107.38+20.30 kg, height=186.76+8.6 cm). Players were categorized into seven weight groups. Data by weight groups were compared by absolute and by allometrically scaled values for 1-RM squat (N.=566), vertical jump (N.=581), and 40 yd (36.58 m) sprint (N.=560) over a seven-year period.

Results of ANOVAs yielded significant (P<0.05) and a near linear pattern of absolute strength by BM. Allometrically scaled 1-RM squat resulted in normalized data void of significant wt group differences while allometrically scaled power and speed did not normalize data to the point that the data could be compared regardless of BM.

Results suggest that it may be possible to determine if an athlete falls within an a normal range of the established standards derived from allometric scaling for the 1-RM squat. However, caution should be taken when applying allometric scaling for power and speed. It is suggested that additional research look into the possibility of adjusting the exponent (performance variable x BMx) to better reflect a normalized condition for more accurate comparisons of athletes.