Background By using deceptive experimental designs, several investigations have observed that athletes may increase their performance when told they were given caffeine, when in fact they received a placebo (i.e., the placebo effect of caffeine). However, most of these investigations used participants not familiarized to caffeine intake, while habituation to caffeine may affect the placebo effect of caffeine. Thus, the aim of the present study was to analyze the placebo effect of caffeine on maximal strength and strength-endurance performance during the bench press exercise (BP) in healthy recreationally trained women who consume caffeine on a daily basis. Methods Thirteen resistance-trained women (BP 1RM = 40.0 ± 9.7 kg) habituated to caffeine (4.1 ± 1.7 mg/kg/day) completed a deceptive randomized experimental design with two experimental trials. On one occasion, participants were told that they would receive 6 mg/kg of caffeine but received a placebo (PLAC) and in other occasion they did not receive any substance and participants were told that this was a control situation (CONT). In each experimental trial, participants underwent a 1RM BP test and a strength-endurance test consisting of performing the maximal number of repetitions at 50% of their 1RM. Results In comparison to CONT, PLAC did not enhance 1RM (40.0 ± 10.5 vs 41.0 ± 9.5, respectively; p = 0.10), nor the number of repetitions (32.2 ± 5.1 vs 31.8 ± 4.5; p = 0.66) or mean power (130 ± 34 vs 121 ± 26; p = 0.08) in the strength-endurance test. Conclusion The expectancy induced by telling participants they were given caffeine did not modify any performance variable measured in this investigation. Thus, the use of the placebo effect of caffeine seemed an ineffective strategy to enhance muscle strength and strength-endurance during the BP exercise in women habituated to caffeine.