We aim at characterizing which kind of functions could be explained (rationalized) as the best reply of payoff-maximizing agents in contests for a fixed prize. We show that the rationalizability strongly differs between Decisive Contests, where the prize is allocated with certainty, and Possibly...