Ivory Towers: Trusting to Gluck

TODAY we celebrate a paper that offers practical advice of unusual relevance: when to bet on the favourite in a horse race.

'Gluck's Second Law: An empirical investigation of horserace betting in early and late races' (Psychological Reports, 1993, vol 72), by Johnnie Johnson and Alistair Bruce, is based on a survey of 1,212 bets placed in betting offices throughout the UK. All bets were 'win-singles', wagers on a particular horse to come first in a race, placed in the 15 minutes prior to the start.

The research was prompted by earlier work on 'Gluck's Laws of Betting Behaviour':

1) Weekday bettors make better bets;

2) Bettors tend to avoid favourites in the last race as they attempt to recoup their losses.

Since the amount of money wagered on a horse affects its odds, Gluck's Second Law, if true, would mean that the odds offered about favourites were better value in the last race. Earlier surveys had confirmed Gluck's Laws, but the present research casts some doubt.

The results showed that gamblers tended to stake larger amounts in the later races of the day (an average bet of pounds 6.51 for the last three races compared with pounds 4.34 on the first two), and they tended to make better bets (winning in 18.5 per cent of last-race bets compared with 10.2 per cent on the first three), but Gluck's Second Law was not confirmed. In fact the results show 'a pattern of behaviour which clearly refutes the premise that last race bettors tend to favour longer- priced selections'. Indeed the mean odds for last-race bets were 6.3 to 1, while those of first-race bets were 12.5 to 1.

The favourite explanation is that gamblers, knowing Gluck's Second Law, now try to take advantage of it by betting on favourites in the last race. But don't bet your shirt on it.