Leading-effect vs. risk-taking in dynamic tournaments : evidence from a real-life randomized experiment

Two "order effects" may emerge in dynamic tournaments with information feedback. First, participants adjust effort across stages, which could advantage the leading participant who faces a larger "effective prize" after an initial victory (leading-effect). Second, participants lagging behind may increase risk at the final stage as they have "nothing to lose" (risk-taking). We use a randomized natural experiment in professional two-game soccer tournaments where the treatment (order of a stage-specific advantage) and team characteristics, e.g. ability, are independent. We develop an identification strategy to test for leading-effects controlling for risk-taking. We find no evidence of leading-effects and negligible risk-taking effects.