Fight or Flight? Defending Against Sequential Attacks in the Game of Siege

Abstract

This paper examines theory and behavior in a two-player game of siege, sequential attack and defense. The attacker’s objective is to successfully win at least one battle while the defender’s objective is to win every battle. Theoretically, the defender either folds immediately or, if his valuation is sufficiently high and the number of battles is sufficiently small, then he has a constant incentive to fight in each battle. Attackers respond to defense with diminishing assaults over time. Consistent with theoretical predictions, our experimental results indicate that the probability of successful defense increases in the defenders valuation and it decreases in the overall number of battles in the contest. However, the defender engages in the contest significantly more often than predicted and the aggregate expenditures by both parties exceed predicted levels. Moreover, both defenders and attackers actually increase the intensity of the fight as they approach the end of the contest.

Keywords: Colonel Blotto, conflict resolution, weakest-link, game of siege, multi-period resource allocation, experiments

JEL Classification: C72, C91, D72, D74