"Recent theoretical and experimental studies have identified several limitations of winner-take-all tournaments that might lead contest sponsors to seek different designs ( [Lazear, 1999] and [Lazear, 2000]). Relative to piece rate wages, winner-take-all incentives may lead to greater variance in effort by players ( [Bull et al., 1987] and [Nalbantian & Schotter, 2006]Eriksson et al., 2009) or sabotage among them (Münster, 2007; [Chen, 2003] and [Harbring & Irlenbusch, 2008]), and the outcomes are also affected by heterogeneity among players ( [Schotter & Weigelt, 1992] and [Harbring et al., 2007]) as well as risk-sharing incentives (Krishna and Morgan, 1998). These considerations may discourage players from entry and distort performance, and thus reduce the total effort elicited in winner-take-all tournaments. Winner-take-all tournaments can also lead to a more unequal distribution of income (Frank and Cook, 1996). Moldovanu and Sela (2001) show that an alternative tournament design with multiple prizes elicits higher aggregate performance when the cost of effort is convex. One of their predictions is tested in a maze-solving contest by Freeman and Gelber (2010), who find that the multiple-prize structure does result in higher aggregate performance than the winner-take-all payment.4
This paper studies a new type of tournament: a proportional-prize contest, in which the prize is divided among participants in proportion to their achievement.5 This type of contest imitates some forms of competition among firms, for example, whose effort may be rewarded through a share of industry profit. Shared prizes can also be awarded in lobbying contests, such as the allocation of import quota licenses among competing importers (Krueger, 1974). Proportional contests may also be used within firms to reward workers, or as a type of procurement contract to elicit effort among suppliers. For example, poultry meat processors in the United States use proportional-payment competitions among their suppliers to spur cost reductions; Zheng and Vukina (2007) study the case of one firm that switched to such contracts in 1984, and estimate the resulting increase in performance compared to the rank-order contests used previously."
I'd really like to get my league started up with some serious competitors, especially after seeing this.