The New Yorker reports that two University College London researchers have empirically validated the existence of hot hand effects — shifting probabilities made in consecutive bets during a winning streak — in online betting:
Juemin Xu and Nigel Harvey, the study’s authors, took a sampling of 569,915 bets taken on an online sports-gambling site and tracked how previous wins and losses affected the probability of wins in the future. Over all, the winning percentage of the bets was somewhere around forty eight per cent. Xu and Harvey isolated the winners and tracked how they fared in their subsequent bets. In bet two, winners won at a rate of forty-nine per cent. From there, the numbers go haywire. A player who had won two bets in a row won his third bet at a rate of fifty-seven per cent. His fourth bet won sixty-seven percent of the time, his fifth bet seventy-two. The best gamblers in Las Vegas expect to win fifty-five per cent of their bets every year. Seventy-two per cent verges on omniscience. The hot hand, it appears, is real.
Losers, unsurprisingly, continued to lose. Of the 190,359 bettors who lost their initial bet, fifty-three per cent lost their next, and those who had enough money left for a third round lost sixty per cent of the time. When unfortunate bettors got to five straight losses, their chance of winning dropped to twenty-three per cent. The losing streaks should be familiar to problem gamblers and can be explained by another well-worn theory called the gambler’s fallacy. If you’ve ever called heads on a coin flip, seen the coin land tails up, and then called heads again because “heads is due,” you’ve been caught up in the gambler’s fallacy. [emphasis added]
The original study can be found here.
The study’s findings have implications for the subjective judgment of independent events; the difference between skill and chance; the halo and winner effects due to hot hand streaks; and the survivorship bias of traders and funds who survive over a long period of time – until asset classes have alpha decay, and financial markets undergo market microstructure or regime changes.
The New Yorker‘s commentary notes but does not explore the statistical edge that the best Las Vegas gamblers develop: the study suggests that such an edge may be situational or fleeting.
This behavioural finance research is extremely useful to understand how cognitive biases can affect decision-making under uncertainty.