Intra- and intergroup conflict are likely to have been recurrent features of human evolutionary history; however, little research has investigated the factors that affect men’s combat alliance decisions. The current study investigated whether features of previous one-on-one combat with an opponent affect men’s interest in allying with that opponent for future group combat. Fifty-eight undergraduate men recruited from a psychology department subject pool participated in a one-on-one laboratory fight simulation. We manipulated fight outcome, perceived fighter health asymmetry, and the presence (...) of a witness over six sets of five rounds of fighting. Following each set, we asked men how interested they would be in allying with their opponent for future group combat. We found that men were more interested in allying with their opponent for future group combat if their opponent won the fight or if a witness was present, but perceived fighter-health asymmetry did not affect men’s decision to ally with their opponent. Exploratory analyses revealed a two-way interaction between fight outcome and the presence of a witness, such that winners without a witness present expressed less interest in allying with their opponent for future group combat. Our findings suggest that men attend to the benefits of allying with a man who has demonstrated relatively superior fighting ability. Alliance with a previous opponent for group combat may vary with the relationship value of the opponent and the utility of demonstrating cooperativeness to third-party observers. These findings inform our understanding of coalition formation. (shrink)