This study examined the effect of different training regimes on moment and power generation during maximal knee extensions at low to very high extension velocities (0-1000 degrees.s-1 individual range). A group of 24 soccer players performed 12 weeks of progressively adjusted strength training of the knee extensors at either high resistance (HR, n = 7), low resistance (LR, n = 6), loaded kicking movements (FU, n = 6), while one group served as controls (n = 5). Moment and power generation of the knee extensors were determined before and after the training period with a nonisokinetic measuring method recently described. Following HR training, knee extension moment increased 9%-10% at knee angular velocities 0 (isometric) and 30 degrees.s-1 (P < 0.05), peak moment increased 20% at 240-300 degrees.s-1 (P < 0.05), while power generation increased 5%-29% at 240-480 degrees.s-1 (P < 0.01). In addition, in the HR group maximal recorded power increased 45% (P < 0.01). After FU training a 7%-13% increase in moment and power was observed at 30-180 degrees.s-1 (P < 0.05). Following LR training, peak moment increased 9% at 120 degrees.s-1 (P < 0.05). Improvements in knee extension moment and power were generally related to the angular velocities employed during training. However, as evaluated using the present measuring method, moment and power increased not only at very low but also at high knee angular velocities following the high-resistance strength training.