Consensus is lacking regarding the influence of vehicle speed and traffic volume on deer–vehicle collision (DVC) rates. Yet, annual average daily traffic fl ow (AADT) and posted speed limit (PSL) typically are used to measure these variables. To resolve this confl ict, we studied the effects of traffic volume and vehicle speed on DVCs in Utah. Our results showed no relationship between AADT or PSL and DVC occurrence. There are at least 3 explanations for our results: (1) no causal relationship exists; (2) AADT and PSL, as measured, actually explain little of the variation; and (3) data quality problems exist. We discuss the likelihood for each explanation. We argue that even though traffic speed and volume have been used to predict DVC occurrence and may be useful explanatory variables, the metrics AADT and PSL are poor surrogate variables. Thus, uses of these variables to predict risk will likely provide unreliable results.