On August 18, 1913, at the Monte Carlo Casino, a roulette table ran players out of millions of dollars.[1] What had started out as an ordinary night quickly turned into a betting frenzy as casino goers noticed that one of the roulette wheels kept landing on black.[2] As the wheel kept spinning, the players threw money at the pot, positive that after so many black spins, any second now, the wheel would have to land on red and win them big.[3] But it was not so. 26 roulette spins in a row landed on black, and the poor roulette hopefuls left the Casino a few million dollars shorter than they came in with.[4]
That night at the Monte Carlo is an example of the cognitive bias known as the gambler’s fallacy. The gambler’s fallacy, also aptly known as the Monte Carlo Fallacy, refers to the well-documented tendency of people to incorrectly estimate the likelihood of streaks occurring by chance.[5] When the Monte Carlo gamblers noticed that the roulette wheel had been landing on black for such a long time, they incorrectly assumed that a red spin had to be coming up shortly. They assumed this even though each additional spin still only had a 50-50 chance of landing on red.[6] Similar studies have documented that people often believe that a sequence of coin flips that alternates, such as “HTHTH,” is more likely to occur than a sequence of flips that does not alternate, such as “HHHHT.” This belief again persists even though each individual coin flip occurs with equal probability.[7]
The issue with this misconception of chance is that the gambler’s fallacy doesn’t limit itself to the casino. Any decision maker suffering from the gambler’s fallacy may believe that sequences of the same type are unlikely to occur by chance.[8] This means that a decision maker, such as a judge hearing a sequence of cases, may unconsciously assume that a case will be of a certain quality, or will have a certain outcome, based on the previous cases they’ve heard.[9] In 2016, an Oxford study evaluated this issue by examining the presence of the gambler’s fallacy in immigration law asylum cases.[10] The study analyzed 357 asylum judges, in 45 different courthouses, across 150,357 cases, to determine whether the gambler’s fallacy was present in the courtroom.[11] The study found that not only was the gambler’s fallacy present, but that asylum judges were up to 3.3% more likely to reject asylum if they had approved the previous application.[12] This statistic persisted even if the two back-to-back applications were substantively identical.[13] Additionally, the study found that this negative correlation recurred between judges of all experience types, and even across hearings taking place over several, consecutive days.[14]
Although 3.3% may not seem like a lot, if you were one of the thousands of applicants who were rejected asylum merely because the prior applicant was granted asylum—then the gambler’s cognitive bias was life changing!
Furthermore, when the decision is not as life-altering as asylum, the author of the Oxford study found that the gambler’s fallacy had an even greater impact on decision making. The researchers also studied loan offices in India.[15] As detailed in the study, loan officers were eight percentage points more likely to deny a loan merely because they approved the previous loan they reviewed.[16] Similarly, in trial, where each individual ruling can alter the course of the case, those eight percentage points quickly add up.
A noted expert in teaching trial training techniques (and a retired federal judge to boot) used to say about objections to a witness testimony—in essence—”[d]on’t worry if your objection is overruled, your next objection will be sustained. Judges do this unknowingly.” At the time, I thought that he said this to encourage attorneys to not take it too personally when being overruled, and to keep our heads up for the next objection. But now, based on the gambler’s fallacy’s implications on judge’s’ rulings, I am no longer so sure.
The Oxford study demonstrates that the gambler’s fallacy exists in the courtroom. And, while the study limited its courtroom analysis to asylum judges, it’s not difficult to extrapolate that any legal professional could find themselves ensnared in this bias. People are fallible to bias, and judges, attorneys, and mediators are people. As noted in my prior article, there are over 100 cognitive biases, and the gambler’s fallacy is just one of them.
Litigation itself is a gamble; a gamble of time, money, and energy for everyone involved. But luckily, your case doesn’t have to settle for the roulette table. Resolutn removes the risk of the gambler’s fallacy and prioritizes true neutrality with its double-blind negotiation system. Even better, Resolutn only charges a fee if your case reaches a settlement, meaning you don’t have to gamble your money away for a shot at a positive litigation outcome. Don’t gamble the outcome of your case; choose true neutrality, choose Resolutn.
[1] Esther Inglis-Arkell, The Night the Gambler’s Fallacy Lost People Millions, GIZMODO (Jan. 8, 2014) https://gizmodo.com/the-night-the-gamblers-fallacy-lost-people-millions-1496890660; BBC, Why We Gamble Like Monkeys, BBC Future (Jan. 27, 2015) https://www.bbc.com/future/article/20150127-why-we-gamble-like-monkeys.
[2] Id.
[3] Id.
[4] Id.
[5] Daniel L. Chen, et.al., Decision Making Under the Gambler’s Fallacy: Evidence from Asylum Judges, Loan Officers, and Baseball Umpires, 131Oxford University Press: The Quarterly Journal of Economics 1181, (2016) (located at: https://doi.org/10.1093/qje/qjw017) (citing Amos Tversky & Daniel Kahneman, Judgment Under Uncertainty: Heuristics and Biases, 185 Science 1124 (1974)).
[6] Inglis-Arkell, supra note 1; BBC, supra note 1.
[7] Chen, et. al, supra note 5; Tversky & Kahneman, supra note 5.
[8] Chen, et. al, supra note 5.
[9] Id.
[10] Id.
[11] Id. at 1197.
[12] Id. at 1201.
[13] Chen, et. al, supra note 5 at 1201; Steve Maas, How “Gambler’s Fallacy” Affects Your Decision-Making, PBS (July 6, 2016 1:50 PM) https://www.pbs.org/newshour/economy/how-gamblers-fallacy-affects-your-decision-making.
[14] Chen, et. al, supra note 5 at 1201.
[15] Chen, et. al, supra note 5 at 1205.
[16] Chen, et. al, supra note 5 at 1209.