The purpose of this study was to analyze the influence of exercise type, set configuration, and relative intensity load on relationship between 1 repetition maximum (1RM) and maximum number of repetitions (MNR). Thirteen male subjects, experienced in resistance training, were tested in bench press and biceps curl for 1RM, MNR at 90% of 1RM with cluster set configuration (rest of 30s between repetitions) and MNR at 70% of 1RM with traditional set configuration (no rest between repetitions). A lineal encoder was used for measuring displacement of load. Analysis of variance analysis revealed a significant effect of load (p<0.01) and a tendency in exercise factor (p=0.096), whereas the interaction effect was not significant. MNR at 70% of 1RM was lower for biceps curl (16.31+/-2.59 vs. 8.77+/-3 in bench press and biceps curl, respectively; p<0.05) and at 90% of 1RM (21.85+/-11.06 vs. 18.54+/-12.84 in bench press and biceps curl, respectively; p>0.05). Correlation between 1RM and MNR was significant for medium-intensity in biceps curl (r=-0.574; p<0.05) and between MNR and 1RM/body mass (r=-0.574; p<0.05). Neither 1RM nor 1RM/body mass correlated with velocity along set, so velocity seems to be similar at a same relative intensity for subjects with differences in maximum strength levels. From our results, we suggest the employment of MNR rather than % of 1RM for training monitoring. Furthermore, we suggest the introduction of cluster set configuration for upper-body assessment of MNR and for upper-body muscular endurance training at high-intensity loads, as it seems an efficient approach in looking for sessions with greater training volumes. This could be an interesting approach for such sports as wrestling or weightlifting.