Sugarcane straw is a crop residue that has a dual purpose and can be maintained on the fields or used to produce bioenergy. The straw retention in the field provides multiple ecosystem services, and the complex interactions between straw and yield responses are hard to predict by local studies. The aim of this study was to investigate the straw removal effects on sugarcane yield in southcentral Brazil. To achieve the objective, a set of 21 field studies was conducted in contrasting edaphoclimatic conditions. In addition, data of seven studies from the literature were included to build a more robust dataset. Since straw removal treatments were not homogeneous in all experiments, they were grouped as follows: NR, no removal (baseline treatment); LR, low removal; MR, moderate removal; and TR, total removal. To facilitate the data analysis, the experiments were gathered in four macroregions: southern Goiás, western São Paulo, central-eastern São Paulo, and northeastern São Paulo. The site location was the most effective individual factor to explain the straw removal effects on sugarcane yields. Compared with NR treatment, the average yield losses induced by LR, MR, and TR were 2, 10, and 13 Mg ha −1 in southern Goiás and 2, 4, and 6 Mg ha −1 in western São Paulo states, respectively. In other regions, no clear pattern was observed, and only site-specific effects were observed. Straw removal affected sugarcane yields in all soil types, but higher responses were observed under best conditions for sugarcane growth (i.e., clayey soils in southern Goiás). Our findings indicated harvesting season has a relevant role on determining yield responses to straw removal and younger ratoons are more responsive to straw removal. Conclusions from this study suggest straw removal recommendations cannot be designed based on isolated factors but require holistic and integrated knowledge to ensure that the straw amount left on field is enough to sustain multiple soil ecosystem services and crop yields.