The sustainability of livestock systems can be assessed through their productive and environmental performance, the ecosystem services they provide or their resilience to hazards. We used a modelling approach to assess how key economic performance indicators respond to technical and market hazards in five different meat-sheep farms, across France and Ireland. Hazards were related to seven technical or economic variables: ewe fertility, prolificacy, lamb mortality, prices of light and heavy lambs, concentrate use and energy use. We used a mechanistic model to simulate farm functioning and assess farm performance over 3000 iterations, based on simultaneous random draws with hazards to the previously mentioned seven variables. We quantified this way (i) the compensatory effects of different types of technical and economic mechanisms that lead to more stable economic performance and (ii) the probability of economic collapse of meat-sheep farms through a diachronic analysis. We showed that variations in technical variables have larger effects on income variability than variations in economic variables. We also showed that the most resilient systems, i.e. those with the lowest coefficient of variation of net income, are those that combine a low level of inputs with at least two lambing periods per year. Short duration of pregnancy in ewes makes multiperiod lambing possible, which can buffer the variability of technical variables and enhance the adaptive capability of the system by offering the possibility to move empty ewes to a new batch for re-mating. We thus analysed for the first time farm economic resilience to combined technical and economic hazards, and highlighted the buffer and adaptive mechanisms of resilience, with a mechanistic model.