The bench press is performed in parapowerlifting with the back, shoulders, buttocks, legs and heels extended over the bench, and the use of straps to secure the athlete to the bench is optional. Thus, the study evaluated muscle activation, surface electromyography (sEMG), maximum velocity (MaxV) and mean propulsive velocity (MPV), and power in paralympic powerlifting athletes under conditions tied or untied to the bench. Fifteen experienced Paralympic powerlifting male athletes (22.27 ± 10.30 years, 78.5 ± 21.6 kg) took part in the research. The sEMG measurement was performed in the sternal portion of the pectoralis major (PMES), anterior deltoid (AD), long head of the triceps brachii (TRI) and clavicular portion of the pectoralis major (PMCL). The MaxV, MPV and power were evaluated using an encoder. Loads of 40%, 60%, 80% and 100% 1RM were analyzed under untied and tied conditions. No differences were found in muscle activation between the tied and untied conditions; however, sEMG showed differences in the untied condition between AD and TRI (F (3112) = 4.484; p = 0.005) in the 100% 1RM load, between PMCL and AD (F (3112) = 3.743; p = 0.013) in 60% 1RM load and in the tied condition, between the PMES and the AD (F (3112) = 4.067; p = 0.009). There were differences in MaxV (F (3112) = 213.3; p < 0.001), and MPV (F (3112) = 248.2; p < 0.001), between all loads in the tied and untied condition. In power, the load of 100% 1RM differed from all other relative loads (F (3112) = 36.54; p < 0.001) in both conditions. The tied condition seems to favor muscle activation, sEMG, and velocity over the untied condition.