The study developed and validated a predictive model to predict the risk of pathological upgrade of colorectal polyps based on endoscopic forceps biopsy in 593 patients from Anhui Medical University (August 2022–October 2025). Patients were divided into a training cohort (n=415) and a test cohort (n=178). Using LASSO regression and multivariate logistic regression, independent risk factors were identified: polyp location in the anus, maximum tumor diameter (MTD) ≥ 30 mm, villous structure, polyp erosion and redness of its surface. The model was visualized with a nomogram and its performance evaluated by ROC curve (AUC 0.890 in the training and 0.922 in the test cohort), calibration plot, Hosmer-Lemeshow test and decision curve (DCA). The results showed a high agreement of the predictions with the observed data and good clinical practicality. The model enables endoscopists to accurately assess risk prior to treatment and supports personalized decisions.