Tailoring straw return strategies to specific regional conditions optimize soil health, crop yield, and carbon emission reduction. This study investigated yield-driven, spatial-scale, and long-term strategies for straw return of maize, rice, and wheat cultivation in China. Through a meta-analysis of 108 studies and the application of a machine learning model, the yield effects of straw return were thoroughly examined. Nitrogen fertilizer inputs were identified having the most significant impact on yield, and the yield benefits of straw return stabilized after approximately 13 years. A optimized strategy for straw return was developed for provinces in China, resulting in average yield increases of 13.9 % for maize, 12.2 % for wheat, and 14.0 % for rice. Implementing these locally adapted straw return strategies will contribute to the long-term development of sustainable agriculture in China.