Approximately 20% of patients do not perceive functional improvement after a primary total knee arthroplasty (TKA). This study aims to assess which patient-related and clinical determinants at baseline and six months postoperative can predict lack of self-reported functional improvement at 12 months after primary TKA. In a retrospective cohort study of 569 patients who received a primary TKA between 2015 and 2018, self-reported functional improvement, measured as ≥7 points increase in Oxford Knee Score (OKS) from baseline to 12 months postoperative, was assessed. Patient characteristics and patient-reported variables at baseline and six months postoperative were entered in a logistic regression model with manual backward elimination. Incidence of functional improvement in this study was 73%. Preoperative variables were no strong predictors of the outcome. An increase in pain between baseline and six months postoperative was a risk factor for not functionally improving (odds ratio (OR) 1.13 (95% confidence interval (CI) 1.03–1.23)). An improvement in knee pain and function was a protective factor for lacking functional improvement (OR 0.78 (95% CI 0.74–0.82)). The prediction model explained 44% of variance and showed good calibration and discrimination. Sensitivity and specificity were 82% and 76%, respectively. Using pre- and postoperative variables, a prediction model for self-reported functional improvement one year after TKA was developed. This prediction tool was easy to use at six months postoperative and allowed identification of patients at high risk for not functionally improving one year after TKA. This could facilitate early interventions directed at functional improvement after TKA.