Risk-Return Optimisations Can Lead to Favourite-Longshot Bias in Prediction Markets Operated by Bookmakers
17 Pages Posted: 30 Sep 2017
Date Written: September 28, 2017
Abstract
In this paper we propose a new explanation for Favourite-Longshot Bias (FLB). FLB refers to a well known phenomenon, pervasive in many sport betting markets. As measured by historically realised gains, bets on favourites appear to be more favourable than longshot bets. We show that FLB can be caused by intentional adjustments of odds, resulting from the process of risk-return optimisations by bookmakers’, aimed at minimising risk and maximising profits. Research literature offered so far a number of plausible explanations for this effect. They can be divided into two groups. The first group, applicable to parimutuel betting markets, is focused on risk preferences and behaviour of punters. The second group, applicable to markets dominated by bookmakers, considers the market micro-structure effects like power of bookmakers and limited arbitrage. In this paper we propose an alternative explanation that is applicable to betting markets operated by bookmakers. Our model is based on two basic assumptions: (1) Bookmaker’s essential task is to maximise gains and to minimise risks; and (2) Punters’ demands for particular bets are increasing functions of their Von Neuman - Morgenstern expected utilities. Using these assumptions we formulate, discuss and solve the risk-return optimisation problem. The analytical solution to this problem is a set of optimal odds, with embedded adjustments that are shown to lead to FLB. The predictions of our model agree qualitatively well with empirical observations.
Keywords: Favourite-Longshot Bias, Football Betting, Odds, Prediction Markets, Risk Management, Risk-Return Optimisation, Market Microstructure, Expected Utility
JEL Classification: G14, C57, C70, C81, C55, C58, C49, G02, L11, Z20
Suggested Citation: Suggested Citation