The Impact of Intermediate Information on Sabotage in Tournaments with Heterogeneous Contestants

Abstract

Applying detailed within-tournament information about intermediate scores and contestants' ability in rank-order professional soccer tournaments, this study empirically analyzes the impact of interim results on the sabotage activities of heterogeneous contestants. Intermediate information that suggests that a contest is decided early decreases total sabotage. Splitting contestants into favorites and underdogs reveals that both contestants sabotage the most if intermediate information about the score compensates for or decreases ex ante heterogeneity between the teams. However, engaging in sabotage does not pay off for the contestants.

Misconduct and Leader Behaviour in Contests New Evidence from European Football

“…For our empirical analysis, we combine the method of Deutscher and Schneemann (2015) with the minute-by-minute approach (Buraimo et al, 2010;Buraimo et al, 2012;del Corral et al, 2010). As the respective dependent variables are bivariate, we use probit models to estimate the probability of a yellow card.…”

“…As the respective dependent variables are bivariate, we use probit models to estimate the probability of a yellow card. Following Buraimo et al (2010) and Deutscher and Schneemann (2015), we cluster the data by match to account for dependencies of observations within matches. To capture fixed effects of seasons and the different leagues, two sets of dummy variables are included in all specifications.…”

“…Their results indicate that weak contestants engage more in sabotage, while contestants with greater abilities exert more compliant effort. Deutscher and Schneemann (2015) further refine the analysis of the effect of ability on sabotage by using information on ex-ante heterogeneity between competing football teams, based on betting odds, and within-game information, namely the goal difference. Like del Corral et al…”

“…A particular advantage of this measure is that it not only considers the respective teams' latest results, but also other relevant and recent information, such as injuries and fitness of (key) players, dismissals of coaches, etc. Similar to the case of the goal difference, the variable's square, difference in bookmaker probability squared, is included to account for potential non-linearities (Buraimo et al, 2012 Another factor that might affect the behaviour of contestants is the atmosphere in the stadium (Deutscher & Schneemann, 2015). Hence, we include the log of the number of spectators at the respective match into the model (attendance …”

“…Sports contests are also suitable to study sabotage, because destructive actions against the opponent team aiming to reduce the opponent's chances of winning can be observed quite directly and reliably. Thus, in previous sport studies on destructive contest behaviour, any actions that violate the rules of the respective game have been used to study sabotage (Deutscher & Schneemann, 2015).…”