Snowpack dynamics have a major influence on wildlife movement ecology and predator–prey interactions. Specific snow properties such as density, hardness, and depth determine how much an animal sinks into the snowpack, which in turn drives both the energetic cost of locomotion and predation risk. Here, we quantified the relationships between five field‐measured snow variables and snow track sink depths for widely distributed predators (bobcats Lynx rufus, cougars Puma concolor, coyotes Canis latrans, wolves C. lupus) and sympatric ungulate prey (caribou Rangifer tarandus, white‐tailed deer Odocoileus virginianus, mule deer O. hemionus, and moose Alces alces) in interior Alaska and northern Washington, USA. We first used generalized additive models to identify which snow metrics best predicted sink depths for each species and across all species. Next, we used breakpoint regression to identify thresholds of support for the best‐performing predictor of sink depth for each species (i.e. values wherein tracks do not sink appreciably deeper into the snow). Finally, we identified ‘danger zones,' wherein snow impedes the mobility of ungulates more than carnivores, by comparing sink depths relative to hind leg lengths among predator–prey pairs. Near‐surface (0–20 cm) snow density was the strongest predictor of sink depth across species. Thresholds of support occurred at near‐surface snow densities between 220–310 kg m–3 for predators and 300–410 kg m–3 for prey, and danger zones peaked at intermediate snow densities (200–300 kg m–3) for eight of the ten predator–prey pairs. These results can be used to link predator–prey relationships with spatially explicit snow modeling outputs and projected future changes in snow density. As climate change rapidly reshapes snowpack dynamics, these danger zones provide a useful framework to anticipate likely winners and losers of future winter conditions.