Objective This study aimed to observe the analgesic effect of the cocktail formulation with diprospan during total hip arthroplasty (THA). Methods From September 2018 to April 2019, 120 patients undergoing primary unilateral THA were included in this prospective, randomized, observer‐blinded study. Patients were randomized into three groups, according to the different local infiltration analgesia (LIA) strategies: LIA with ropivacaine (the ropivacaine group, n = 40), LIA with a new cocktail containing ropivacaine, diprospan, and morphine (the cocktail group, n = 40), and the control group ( n = 40). The primary outcomes included postoperative pain scores. The resting visual analogue scale (VAS) scores were measured at 2, 6, and 12 h after the surgery (a.m. and p.m.) on postoperative day (POD) 1, POD2, and the day of discharge. Movement VAS scores were assessed at 6 h, 12 h after the operation (a.m. and p.m.) on POD1, POD2, and the day of discharge. The secondary outcomes included opioid consumption, postoperative hospital stay, range of motion of the hip at discharge, patient satisfaction, and the results of the follow‐up. Results After the screening, 120 patients were randomized into three groups (40 patients in each group). All of the patients completed the trial. The resting VAS scores in the ropivacaine group and cocktail group at 2 h were lower than those in the control group ( P < 0.001 and P < 0.001, respectively, F = 17.054), and the same trend was also postoperatively found at 6 h ( p = 0.005 and P = 0.002, F = 6.212). Twelve hours after the operation, the pain score in the cocktail group was lower than that in the other two groups, but only the difference between the cocktail group and the control group was statistically significant ( P = 0.018, F = 3.144). From the morning of the first postoperative day to the a.m. on POD 2, the VAS scores in the cocktail group were significantly lower than those in the ropivacaine group and the control group. Furthermore, the movement VAS scores in the ropivacaine group and the cocktail group were better than those in the control group at 6 and 12 h post‐operation ( P < 0.05). The per capita opioid consumption in the cocktail group was less than that in the ropivacaine group and the control group within 24 h post‐operation. There were no significant differences in the comparison of additional indicators among the three groups. Conclusion The new cocktail with diprospan had a better result and longer duration time for early postoperative pain control in primary THA via the posterolateral approac...