In step‐pool stream channels, flow resistance is created primarily by bed sediments, spill over step‐pool bed forms, and large woody debris (LWD). In order to measure resistance partitioning between grains, steps, and LWD in step‐pool channels we completed laboratory flume runs in which total resistance was measured with and without grains and steps, with various LWD configurations, and at multiple slopes and discharges. Tests of additive approaches to resistance partitioning found that partitioning estimates are highly sensitive to the order in which components are calculated and that such approaches inflate the values of difficult‐to‐measure components that are calculated by subtraction from measured components. This effect is especially significant where interactions between roughness features create synergistic increases in resistance such that total resistance measured for combinations of resistance components greatly exceeds the sum of those components measured separately. LWD contributes large proportions of total resistance by creating form drag on individual pieces and by increasing the spill resistance effect of steps. The combined effect of LWD and spill over steps was found to dominate total resistance, whereas grain roughness on step treads was a small component of total resistance. The relative contributions of grain, spill, and woody debris resistance were strongly influenced by discharge and to a lesser extent by LWD density. Grain resistance values based on published formulas and debris resistance values calculated using a cylinder drag approach typically underestimated analogous flume‐derived values, further illustrating sources of error in partitioning methods and the importance of accounting for interaction effects between resistance components.