Correlative analyses predict that anthropogenic climate warming will cause widespread extinction but the nature and generality of the underlying mechanisms is unclear. Warming-induced activity restriction has been proposed as a general explanatory mechanism for recent population extinctions in lizards, and has been used to forecast future extinction. Here, I test this hypothesis using globally applied biophysical calculations of the effects of warming and shade reduction on potential activity time and whole-life-cycle energy budgets. These 'thermodynamic niche' analyses show that activity restriction from climate warming is unlikely to provide a general explanation of recent extinctions, and that loss of shade is viable alternative explanation. Climate warming could cause population declines, even under increased activity potential, through joint impacts on fecundity and mortality rates. However, such responses depend strongly on behaviour, habitat (shade, food) and life history, all of which should be explicitly incorporated in mechanistic forecasts of extinction risk under climate change.