Working Paper: CEPR ID: DP3266
Authors: Isabelle Brocas; Juan D. Carrillo
Abstract: This note argues that a rigorous application of simple game theory may provide unambiguous yet non-trivial theoretical insights about the behaviour of players in simple games. This contrasts with a commonly held view that many predictions in applied game theory are either obvious or inconclusive. To illustrate our point, we analyse the merits of two controversial changes in soccer rules, namely the ?three-point victory? and the ?golden goal?. Starting from standard premises, we present some original conclusions that are neither trivial nor the result of a twisted argument. We feel that soccer is a particularly good example for our exercise due to the simplicity of its main rules, but also to the proliferation of ad-hoc reasoning among soccer fans.
Keywords: game theory; soccer rules
JEL Codes: C72
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| three-point victory (3PV) rule (P30) | teams adopt more offensive strategies towards the end of the match when tied (C73) |
| three-point victory (3PV) rule (P30) | teams play more defensively at the beginning to avoid conceding an early goal (C73) |
| three-point victory (3PV) rule (P30) | teams play more defensively under 3PV than under the traditional two-point victory (2PV) rule (P30) |
| golden goal (GG) rule (C73) | alters the incentives for offensive play (L21) |
| golden goal (GG) rule (C73) | prevents the team that concedes a goal from recovering (C73) |
| combination of the 3PV and GG rules (C52) | enhances the incentives for teams to play offensively during extra time (Z23) |
| combination of the 3PV and GG rules (C52) | leads to higher average incentives for offensive play compared to either rule applied in isolation (C73) |