We used the Gait Deviation Index (GDI) as method to compare preoperative to postoperative gait changes after uncemented 50 mm (median) large-head and 28/32 mm total hip arthroplasty (THA). We also identified predictors of improvements in GDI. Gait analysis and patient-reported (WOMAC) data were recorded in 35 patients before, 2 and 6-months after treatment. Twenty age-matched adults provided normative gait data. Contrary to our hypothesis, patients who received large-head THA had less improvement in GDI compared with patients who received 28/32 mm THA. The preoperative GDI score was identified as a predictor of postoperative GDI improvement, while WOMAC, age, gender and walking speed were not. This study provides useful information for clinicians and rehabilitation specialists about gait improvement that can be expected after THA.