Promoting tolerance towards wildlife that pose a threat to humans or their livelihoods is critical to achieving species conservation goals. While conservation discourses often revolve around retaliation, tolerance and intolerance can vary from passive acceptance to active support, and according to level and scale of loss. We developed an experimental tool with a gamified survey-based field experiment to test multiple intended behaviours along a gradient of livestock loss to predation by jaguars (Panthera onca) in Calakmul, Mexico. Using a within-subject design, participants played a card-choice game against livestock loss scenarios. The choices offered were “No action [taken]”, “Protect [domestic animals]”, “Kill [the jaguar]”. First, our experimental design was validated by increased frequencies of “Kill” responses in reaction to hypothetical jaguar incursion with increasing damage. The greatest difference occurred between zero and minimal damage (one chicken), showing that whether damage occurs at all is at least as important as its extent. Secondly, average frequency of intentional retaliation to jaguar incursion was higher for participants who had already killed a jaguar. Finally, age, education, and ethnicity all correlated with intentional behaviours to implement preventive measure or kill jaguars. Past experience of attack had no effect. Our results showed that survey/game-based experiments can assess tolerance to potentially hazardous wildlife interactions, while allowing more nuanced understandings of people's tolerance on a gradient of loss. This allows us to move away from a dualistic model of “conflict” versus “coexistence” towards a more complex understanding incorporating individual attitudes and experience, cultural significance, and socio-economic context.