As governments have implemented interventions addressing COVID-19, mortality statistics have varied over time and among countries. But what impact, if any, does this information have on attitudes relevant for politics and public health? In a pre-registered experiment (N = 2,915), we showed UK respondents an image displaying the UK’s cumulative COVID-19 deaths—which remain the highest in Europe—either in isolation or alongside nine other European countries having fewer deaths. Mortality figures in either format did not significantly impact respondents’ concern about the pandemic, support for restrictions, or anticipated behaviors. However, respondents seeing UK-only data, particularly those holding lower trust in government (which was an exploratory test conducted after pre-registration), evaluated the UK’s response more positively compared to the control group. This was the opposite direction from what we had predicted at the pre-registration stage. By contrast, those seeing the UK’s mortalities in context—particularly respondents holding higher trust in government (again, an exploratory test)—expressed more negative evaluations compared to those who saw the UK-only data, which was in line with our pre-registered hypotheses. We attribute these divergent patterns to whether cumulative mortalities or rates of change were more apparent: respondents associated the declining rate rather than high gross figures with government “winning the battle” when relative benchmarks were not present, but punished the incumbent government when its standing among peers as the “worst of the bunch” in terms of cumulative deaths was made visibly salient. Our results highlight the political stakes of information and its communication during crises.