Pesticides are widely used in agriculture to protect crops from animal pests, diseases, and weeds, helping to maintain yields under diverse production conditions. However, their widespread and repeated use has led to environmental contamination, biodiversity loss, and growing concerns about human health. While Integrated Pest Management (IPM) and organic farming have sought to reduce pesticide dependency, both approaches still permit pesticide use, and their adoption remains limited due to technical and economic constraints. In this context, we explore the feasibility of a third way: pesticide-free agriculture based on agroecological crop protection (ACP) principles. Drawing from the Rés0Pest experimental network launched in France in 2012, we present ten years of results from nine sites covering a range of pedo-climatic conditions and socio-economic contexts. Rés0Pest implemented cropping systems that excluded all pesticide use, including seed treatments, while maintaining synthetic fertilizer inputs. The systems were co-designed through participatory methods, following a system experiment approach that evaluates the effects of a combination of cropping practices and their interactions on cropping system performance over the long term. Results showed that in pesticide-free systems, it is possible to achieve yields comparable to conventional and higher than organic systems and, in some cases, generate higher net farm income. Pest and pathogen crop damage did not significantly increase over time, although weed management remained a key challenge. These findings suggest that technically and economically viable pesticide-free arable systems are possible under certain conditions, and that new solutions are needed to support their adoption across a wider range of contexts. We discuss implications for research, farming, and policy, and emphasize the need for adaptive experimentation and systemic performance assessment to support agroecological transitions.