ABSTRACT.- Eucalyptus species have a great capacity for regeneration after harvest, which allows a second rotation as a coppice crop. The decision whether to manage the next rotation as a coppice crop or to replant depends on the expected economic result of each alternative. The problem that foresters face is the difficulty of predicting the productivity in the next rotation, which will depend largely on the percentage of stumps that resprout. Therefore, the objectives of this work were:(i) to identify the preharvest variables that influence stump regrowth and (ii) to develop a model for predicting the probability of stump regrowth in commercial plantations of Eucalyptus globulus in Uruguay based on pre-harvest information. Thirty-three plots were established in commercial plantations, in which silvicultural management, growth and health status before harvest were recorded, as well as the number of stumps that sprouted after harvest. Significant differences were found in the percentage of resprout for the following variables:genetic material, type of harvest machine, proportion of trees with DBH > 14 cm, proportion of trees with bark cankers, proportion of trees with epicormic shoots, and proportion of trees with apical death. However, the logistic regression model adjusted to predict the probability of regrowth only included as explanatory variables the genetic material, the type of harvesting machine, the proportion of trees with DBH > 14 cm, and the proportion of trees with bark cankers. The use of this model will allow managers of E. globulus plantations to make more informed decisions for the next rotation. © The Author(s), under exclusive licence to Springer Nature B.V. 2023