Copaifera oleoresin is an important traditional non-timber forest product in the Amazon; however, the Carajas National Forest has no plans to harvest this product to benefit extractive families. The objective of this study was to determine Copaifera oleoresin extraction potential in the Carajas National Forest, Eastern Amazon, identifying variables that could increase the chances of finding oleoresin. We mapped 129 trees and formulated a multiple logistic regression model (logit) with the collected variables: diameter at breast height (DBH), tree crown, tree health, presence of termites, and presence of damage or bifurcation. Among the mapped trees, we perforated 50 trees to harvest the oleoresin. Most of the oleoresin production occurred in trees with a DBH between 50 and 75 cm. The logit model predicted that an increase in DBH, dense crown, unhollowed trunk, and presence of damage or bifurcation increased the chances of oleoresin extraction, whereas termites did not. The model could contribute to the sustainable management of forest resources by avoiding random perforation of trees and establishing rational routes and cycles of extraction. Since the variables are easily measured, the model can be continuously used for future extractions.