Harvest data provide readily available and relatively inexpensive information about populations of game species. However, these data are not necessarily representative of standing populations and may have limited applicability in management. We applied a method of harvest data analysis based on the changing sex ratio of the harvest with age to American black bear (Ursus americanus) harvest data from 1985–2005 in Montana. We assessed the ability of this method to identify assumption violations and the extent of resulting bias. A change in the relative vulnerability of females at primiparity due to protection of mothers with cubs from harvest was observable as a drop in the proportion of females in the harvest at the age of maturity. A changing harvest rate produced changing harvest rate estimates, but the estimates lagged up to 10 years behind the actual rate. Other assumption violations, such as unequal non-harvest mortality between sexes and stochasticity in the harvest rate, are not apparent in the harvest data themselves. If total harvest is known and the harvest rate is estimable, it may be possible to use harvest to identify population declines. However, we found with simulations that, in many cases, 10–15 years of harvest data are needed to identify a statistically significant decline. If all assumptions are met, we estimated harvest rates in Montana as 4.6% for females and 10.4% for males; these are overestimates if males have higher non-harvest mortality than females. Montana's harvest data did not show an apparent decline in the relative vulnerability of females at maturity, despite nominal protection of mothers accompanied by cubs. Analyses of harvest data also contradicted the hypothesis, based on meta-analysis of demographic data, that black bears were declining in Montana.