Counts of nesting birds are often used to monitor the abundance of breeding pairs at colonies. Mean incubation counts (MICs) are counts of nests with eggs at intervals that correspond to the mean incubation period of a species. The sum of all counts during the nesting season (MIC_total) and the highest single count during the season (MIC_max) are metrics that can be generated from this method. However, the utility of these metrics as measures of the number of breeding pairs has not been well tested. We used two approaches to evaluate the bias and precision of MIC metrics for quantifying annual variation in the number of breeding Red-tailed Tropicbirds (Phaethon rubricauda) nesting on two islands in the Papahnaumokukea Marine National Monument in the northwest Hawaiian Islands. First, we used data from nest plots with individually marked birds to generate simulated MIC metrics that we compared to the known number of nesting individuals. The MIC_total overestimated the number of pairs by about 5%, whereas the MIC_max underestimated the number of pairs by about 60%. However, both metrics exhibited similar precision. Second, we used a 12-yr time series of island-wide MICs to compare estimates of temporal trend and annual variation using the MIC_max and MIC_total. The 95% confidence intervals for the trend estimates were overlapping and the residual standard errors for the two metrics were similar. Our results suggest that both metrics offered similar precision for indices of breeding pairs of Red-tailed Tropicbirds, but that MIC_total was more accurate. ?? 2009 Association of Field Ornithologists.