In gambling and economics, the favorite-longshot bias is an observed phenomenon where on average, betters tend to overvalue "longshots" and relatively undervalue favorites. That is, in a horse race where one horse is given odds of 2-to-1, and another 100-to-1, the true odds might for example be 1.5-to-1 and 300-to-1 respectively. Betting on the "longshot" is therefore a much worse proposition than betting on the favorite. In the long run, losing 5% by betting on the favorite, but losing 40% on longshots is not uncommon.
The phenomenon was first discovered by Griffith. Various theories exist to explain why people willingly bet on such losing propositions, such as risk-loving behavior, risk-averse behavior"We discount the chances of any party at 100/1 or bigger. The reverse of tweak 1 applies here. Almost all of these probably have effectively zero chance. Why don’t we just make them a bigger price? We don’t think we’ll take much extra money, certainly not enough to compensate us for the day we get it wrong." Matthew Shadwick, Ladbrokes, 2010-02-25. See Ladbrokes Election Forecast Feb 25th 2010 or simply inaccurate estimation as presented by Sobel and Raines.Russell S. Sobel & S. Travis Raines, 2003. "An examination of the empirical derivatives of the favourite-longshot bias in racetrack betting," Applied Economics, Taylor and Francis Journals, vol. 35(4), pages 371-385, January
Methods such as the goto_conversion,Convert Betting Odds to Probabilities More Accurately and Efficiently than Shin and Power Methods PowerAdjusting Bookmaker’s Odds to Allow for Overround and ShinOn determining probability forecasts from betting odds can be used to measure the bias by converting betting odds to true probabilities.
See also Rank-dependent expected utility
References