Predicting postoperative outcomes following total knee arthroplasty (TKA) is important for patient selection. This study focuses on patient-reported outcomes and satisfaction in relation to preoperative radiographic arthritis severity. A retrospective chart review of 420 TKAs was performed. Patient satisfaction was determined using a structured telephone survey with questions focused on degree of satisfaction whether they would have surgery again and their ability to kneel. The radiographic severity of the arthritis of the tibiofemoral joint was graded by a musculoskeletal radiologist using the Kellgren–Lawrence grading scale. The patellofemoral compartment was graded using the scale described by Jones et al. Those grouped as severe arthritis had an overall satisfaction rate of 96% (76% fully satisfied and 20% partially satisfied) compared with 82% of the time (64% fully satisfied and 18% partially satisfied) if their arthritis was mild. Postoperatively 51% of TKA patients were able to kneel. Univariate logistic regression showed an association between higher rates of satisfaction and male gender (p = 0.053), severity of preoperative radiographic arthritis (p = 0.034) those who would have surgery again (p ≤ 0.0001) and those able to kneel (p = 0.005). Patients should be informed preoperatively that if their arthritis is only mild radiographically, their outcomes are less predictable. There should also be a discussion surrounding kneeling and activities patients do, which may require kneeling, as many are unable to kneel postoperatively. The Level of Evidence for the study is III.