We empirically explore the relevance of risk taking behavior in tournaments. We make use of data from the NBA season 2007/2008 and measure risk taking by the fraction of three-point shots. Current heterogeneity of teams is taken into account by intermediate results. It turns out that indeed teams who are behind increase the risk in terms of more three-point attempts. We additionally analyze the consequences of this change in behavior. Enhanced risk taking is inefficient for the vast majority of cases and only beneficial in terms of a higher winning probability if a team is behind with a rather large amount of points. We discuss possible explanations for these decision errors.