Experimental determination of log erosion barrier effectiveness is a difficult proposition. Spatial variability in precipitation can yield inconclusive results for small watershed studies, while small scale plots can miss the effects of cascading overtopping. Time scales available for experiments are also short, with substantial changes in hillslope properties taking place within a few years of the fire. In order to design experiments to obtain information efficiently, a clearer expectation of the hillslope scale response is needed. We used observations of the spatial continuity of water repellency and log barrier dimensions installed on a burned hillslope in Idaho as a basis for a simple model of barrier effect at a hillslope scale. Under some circumstances, the logs capture a significant proportion of the runoff, but under others, both lesser and greater in magnitude, the logs contributed little. As installed, the examined structures retained 0.2 mm of precipitation, creating a narrow window. We propose that a probabilistic description of barrier effectiveness may be useful for planning purposes.