Purposes The focus in the evaluation of total knee arthroplasty has shifted from objective measures of implant position and knee function, to patient-reported outcome measures (PROMs). The relation between these two measures was investigated and the possibility of prediction of the patient satisfaction level was evaluated by defining thresholds for improvement of (1) range of motion (ROM), (2) 6-min walk test (6MWT), (3) sit-to-stand test (STS) and (4) quadriceps force after TKA? Methods Fifty-seven patients were prospectively tested at preoperative and 6 months postoperative intervals. The ROM, 6MWT, STS-test and quadriceps force were evaluated. Two clusters were created based on the postoperative KOOS, OKS and the satisfaction subscore of the new KSS, cluster 1 consisted of patients with good to excellent PROMs, cluster 2 of patients with poorer PROMs. Patients in each cluster were more similar to each other than to those in the other cluster. Receiver operating characteristic (ROC)-curve analysis was used to identify thresholds for the functional outcomes that established cluster allocation. Multiple logistic regression was used to define a model to predict cluster allocation. Results Patients with high postoperative PROMs (cluster 1 allocation) showed higher postoperative functional outcomes (p < 0.05). Thresholds for the improvement of ROM (≥ 5°, OR 6.3, 95% CI 1.23-31.84), 6MWT (≥ 50 m, OR 8.2, 95% CI 1.61-42.18) STS (≥ 1.05 s, OR 3, 95% CI 0.56-16.07) and normalized Q4 force (≥ 1.5 N/BMI, OR 2.5, 95% CI 0.49-12.89) were found to be predictors of cluster allocation. A model to predict the cluster allocation contained gender, ROM improvement and 6MWT improvement (sensitivity 91.1%, specificity 75%). Conclusions Thresholds for improvement of functional parameters can predict the patient satisfaction cluster. Patients, who are male, improve on the 6-min walk test with 50 m or more and have an increased range of motion of 5° or more, compared to the preoperative situation, are 6-8 times more likely of being satisfied after TKA. These tests are easy to use in clinical practice and can predict the level of patient satisfaction after TKA. Level of evidence Level II, prognostic study.