Evaluation of power prediction equations: Peak vertical jumping power in women

Purpose: The purpose of this investigation was to: 1) compare actual peak power (PPactual) to estimated values (PPest) derived from three different prediction equations (Sayers and Harman), 2) determine the ability of the prediction formulas to monitor change following 6 wk of plyometric training, and 3) generate a new regression model. Methods: Twenty college females (age = 20.1 +/- 1.6 yr; body mass = 65.9 +/- 8.9 kg) were randomly assigned to a control or intervention group. Pre- and posttest countermovement jump (CMJ) height and PPactual were determined simultaneously on a force platform. Body mass and maximal CMJ height were used to predict peak power. Results: All three PPest were significantly correlated to PPactual and to each other on pre (r = 0.84-0.99) and post (r = 0.88-0.99) tests. PPactual (2425.4 +/- 372.4 W) was significantly less than PPest (Sayers: SJ = 2920.8 +/- 482.6 W; CMJ = 2925.1 +/- 473.0 W) but was not different from PPest (Harman: 2585.0 +/- 409.7 W). Posttests revealed similar differences between PPactual and PPest for the intervention group, however no significant differences were observed for the control group. Mean differences from pre and posttests did not differ within or between PPest. Regression analysis determined the formula: PPest = 65.1 X (jump height) + 25.8 X (body mass) - 1413.1 (R-2 = 0.92; SEE = 120.8), which slightly underestimated (0.77%) peak power compared with PPactual in our cross-validation sample (N = 7). Conclusions: Changes in peak power is accurate using any of the regression equations; however, the new prediction formula and that of Harman seem to more precisely estimate peak power. Strict jumping technique along with simultaneous measurement of power and jump height should be used as the standard for comparison.