We estimated the population growth rates of Bull Trout Salvelinus confluentus and used population growth models to evaluate observation error and estimate Bull Trout persistence probabilities by using 25 data sets (averaging 19 years of record) that indexed abundance across Idaho. These data sets were derived from a variety of fish sampling techniques, including weirs, screw traps, redd counts, daytime snorkeling, electrofishing, and angler creel. Bull Trout populations in Idaho were relatively stable prior to 1994, but after 1994, substantially more population growth rates trended statistically upward (n = 14) than downward (n = 3). Average intrinsic rates of population change were 0.01 (SE = 0.03) prior to 1994 and 0.07 (SE = 0.02) after 1994; across all years of data, the rate of change averaged 0.07 (SE = 0.02). Fifty‐nine percent of the data sets had zero to minimal observation error according to Gompertz state‐space model estimates; observation error was least common in data from screw traps and redd counts and most common in snorkel data. Gompertz‐type density‐dependent models were most often the best‐fitting models for Bull Trout population growth. Moreover, few of the most reliable modeling results (i.e., those from data sets estimated to have zero to minimal observation error) contained a period effect or time (i.e., year) effect, suggesting that carrying capacity generally did not differ between the pre‐1994 and post‐1994 periods and generally was not trending positively or negatively through time. Parametric bootstraps predicted that the mean probability of falling below the quasi‐extinction level of 20 adults in the next 30 years was 9.8% (median = 4.7%) for data sets estimated to have zero to minimal observation error. The weight of evidence from our modeling results suggests that for most Bull Trout populations in Idaho, abundance is stable or increasing and the risk of extirpation in the foreseeable future is low.Received March 12, 2013; accepted November 18, 2013Published online February 10, 2014