Background and purpose — The concept of fast-track surgery has led to a decline in length of stay after total hip arthroplasty (THA) and total knee arthroplasty (TKA) to about 2–4 days. However, it has been questioned whether this is only achievable in selected patients—or in all patients. We therefore investigated the role of preoperative pain and functional characteristics in discharge readiness and actual LOS in fast-track THA and TKA. Methods — Before surgery, hip pain (THA) or knee pain (TKA), lower-extremity muscle power, functional performance, and physical activity were assessed in a sample of 150 patients and used as independent variables to predict the outcome (dependent variable)—readiness for hospital discharge —for each type of surgery. Discharge readiness was assessed twice daily by blinded assessors. Results — Median discharge readiness and actual length of stay until discharge were both 2 days. Univariate linear regression followed by multiple linear regression revealed that age was the only independent predictor of discharge readiness in THA and TKA, but the standardized coefficients were small (≤ 0.03). Interpretation — These results support the idea that fast-track THA and TKA with a length of stay of about 2–4 days can be achieved for most patients independently of preoperative functional characteristics.