OBJECTIVES: Non-pharmaceutical interventions have been crucial to reduce transmission of the SARS-CoV-2 virus in many countries including the United Kingdom. A key research priority has been to better understand psychological and social determinants of health behaviours. We aimed to quantify the impact of luck perception on contact and preventive behaviours among adults in the UK, adjusting for key confounders. STUDY DESIGN: A cross-sectional study. METHODS: Data were collected between July 28 and August 31, 2020. Luck perception, which refers to a belief whether individual's SARS-CoV-2 infection status is determined by fate or chance, was measured using Chance score, drawing on Health Locus of Control Theory. Self-reporting online questionnaires were administered to obtain participants' contact patterns and frequencies of avoiding crowds, hand washing and wearing a mask. Associations between luck perception and protective behaviours and contact patterns were quantified using regression models. RESULTS: Data from 233 survey respondents were analysed. Chance score was negatively associated with all protective behaviours; avoiding crowds (adjusted odds ratio (aOR) 0.46, 95% confidence interval (CI) 0.25-0.86, p = 0.02), washing hands (aOR 0.35, 95%CI 0.17-0.70, p = 0.003), and wearing masks (aOR 0.58, 95%CI 0.34-0.99, p = 0.046). For non-physical contacts (with or without distancing), a significant interaction was identified between Chance score and ethnicity. Chance score increased the number of non-physical contacts among white British, an opposite trend was observed for non-white participants. CONCLUSIONS: Luck perception during the pandemic may affect individuals' health protection behaviours and contact patterns. Further mechanistic understandings of human behaviours against infectious diseases are indispensable for effective response to future pandemics.