<p>Septoria tritici blotch (STB), caused by Zymoseptoria tritici, is one of the most destructive foliar diseases limiting spring wheat production in the central highlands of Ethiopia. Despite its economic importance, the epidemiological drivers and predictive determinants of STB epidemics in this region remain insufficiently characterized. This study assessed the distribution, incidence, severity, and associated risk factors of STB across major wheat-growing areas of the central highlands during the 2021–2023 cropping seasons. Disease incidence and severity data were collected from 252 farmers’ fields, together with agronomic, environmental, and spatial variables. Bivariate association analysis, fixed-effects multivariate logistic regression, and machine-learning approaches were used to identify significant predictors and evaluate model performance. STB was widely distributed across surveyed areas, with mean incidence ranging from 49.3% to 100% and severity from 4.9% to 69.1%. Logistic regression analysis identified sowing date, previous crop, pycnidia density, wheat variety, landscape position, soil type, relative humidity, rainfall, number of rainy days, and altitude as significant predictors of disease intensity. Early sowing, cereal-to-cereal rotations, susceptible varieties, Vertisols, and flat landscapes were associated with significantly higher epidemic risk. Random Forest models outperformed generalized linear models in predicting both incidence and severity, achieving higher predictive accuracy for incidence (R² = 0.602, RMSE = 1.92) and severity (R² = 0.909, RMSE = 0.26), compared with GLM performance (R² = 0.51 and 0.52; RMSE = 0.17 and 0.13 for incidence and severity, respectively). These findings provide a robust framework for STB risk forecasting, disease surveillance, and targeted management strategies to support sustainable wheat production in Ethiopia.</p>