Brazilian sugarcane yield is below its physiological potential, which has compromised the crop’s profitability. This, together with the expansion of the crop to marginal areas with limiting climatic conditions, requires studies to quantify crop yield gaps (YG) and to identify their main causes (i.e. droughts and/or crop management). One way to determine YG is through crop simulation models, which vary in complexity, mainly in terms of input data requirements. This study evaluated whether a simple agrometeorological crop yield model could be suitable for estimating sugarcane YG at a national level, in order to consider and suggest practices to mitigate yield losses. The model was calibrated and evaluated for different conditions across the country. The calibrated model was used to estimate plant and ratoon sugarcane potential (Yp) and best farmer (Ybf) yields for 259 locations representing all regions of the country where sugarcane is grown. Weather data from 1984 to 2013 and general local soil information were used as inputs. The Yp and Ybf simulations were performed for 30 growing cycles, with the final yields being weighted by the proportion of plant (20%) and ratoon (80%) canes in each area. These data were compared with actual average yields (Yavg), obtained from official surveys. Sugarcane yields varied considerably across the country: Yp range was 68.5–232.7 t ha^–1, Ybf 61.7–123.3 t ha^–1, and Yavg 11.2–101.1 t ha^–1. These yields resulted in an average total YG of 133.2 t ha^–1. The main source of YG was water deficit, accounting for 75.6% of total losses, while crop management was responsible for 24.4%. Considering the main sources of YG for sugarcane in Brazil, the use of drought-tolerant cultivars, irrigation, and deep soil preparation seems the best strategy to mitigate the risks, improving yields. Based on these results, the simple agrometeorological crop yield model proved suitable to estimate sugarcane YG at national level.