This study investigated the effect cadence has on the estimation of critical power (CP) and the finite work capacity (Wʹ) during the 3-minute all-out cycling test. Ten participants completed 8 tests: 1) an incremental test to calculate gas exchange threshold (GET), maximal aerobic power (MAP) and peak oxygen uptake (V̇O2peak), 2–4) three time-to-exhaustion tests at 80, 100 and 105% MAP to calculate CP and Wʹ, 5–7) four 3-minute all-out tests to calculate end power (EP) and work done above EP (WEP) using cadences ranging from preferred −5 to preferred +10 rev·min−1 to set the fixed resistance. Significant differences were seen between CP and EP-preferred (267.5 ± 22.6 W vs. 296.6 ± 26.1 W, P < 0.001), CP and EP−5 (267.5 ± 22.6 W vs. 303.6 ± 24.0 W, P < 0.001) and between CP and EP+5 (267.5 ± 22.6 W vs. 290.0 ± 28.0 W, P = 0.002). No significant differences were seen between CP and EP+10 (267.5 ± 22.6 W vs. 278.1 ± 30.9 W, P = 0.331). Significant differences were seen between Wʹ and WEP at all tested fixed resistances. EP is reduced when cycling at higher than preferred cadences, providing better estimates of CP.