The gambler's fallacy: why past results don't change future odds
A roulette wheel lands on red six times in a row. You're convinced black is "due." You bet bigger on black — and red hits again. That feeling of being overdue isn't intuition. It's the gambler's fallacy, one of the most common and costly thinking errors in gambling. This guide explains what it is, why your brain falls for it, and how to stop it costing you money.
What is the gambler's fallacy?
The gambler's fallacy is the mistaken belief that if something has happened more frequently than normal during a given period, it's less likely to happen in the future — or vice versa. It's the feeling that a coin that has landed heads five times in a row is somehow "due" to land tails.
The reality: each flip is an independent event. The coin has no memory. The probability is 50/50 every single time, regardless of what happened before. The same applies to roulette spins, dice rolls, slot machine outcomes, and lottery draws. The previous result has zero influence on the next one.
Why your brain gets this wrong
The fallacy stems from a misunderstanding of the law of averages. People correctly intuit that over thousands of coin flips, heads and tails will roughly even out. But they incorrectly assume this evening-out must happen in the short term — that the coin needs to "correct" itself after a streak. It doesn't.
The evening-out happens because future results dilute past streaks, not because the universe compensates for them. After 5 heads in a row, the next 1,000 flips will be roughly 50/50 — and that 5-flip imbalance becomes statistically insignificant. The coin didn't "fix" anything. The sample just got bigger.
Psychologists call this the representativeness heuristic — we expect small samples to look like the overall pattern. Five heads in a row doesn't "look random," so we assume tails must be coming. But randomness is far streakier than people expect. A sequence like HHHHHT feels non-random, but it's exactly as likely as HTHTHT.
The Monte Carlo fallacy — the most famous example
On 18 August 1913, at the Monte Carlo Casino, the roulette ball landed on black 26 times in a row. As the streak grew, gamblers piled money onto red, convinced it was overdue. They lost millions. The ball didn't care about the streak. Each spin had the same probability — and the house cleaned up.
This is why the gambler's fallacy is also called the Monte Carlo fallacy. The event perfectly illustrates how the bias works at scale: the longer the streak, the more certain people became that it would end, and the more they bet. The casino's edge didn't change. The gamblers' behaviour did.
The probability of black hitting 26 times in a row on a European roulette wheel is approximately 1 in 136 million. Extremely unlikely — but not impossible. And crucially, after 25 blacks in a row, the probability of the 26th spin being black was still exactly 48.6%. The streak changed nothing about the next spin.
Independent events — what the maths actually says
The table below makes the core point. No matter how many times red has come up in a row, the probability on the next spin is unchanged.
| Previous Results | Probability of Red | Probability of Black |
|---|---|---|
| 1 red in a row | 48.6% | 48.6% |
| 5 reds in a row | 48.6% | 48.6% |
| 10 reds in a row | 48.6% | 48.6% |
| 26 reds in a row | 48.6% | 48.6% |
| 100 reds in a row | 48.6% | 48.6% |
(European roulette: 48.6% red, 48.6% black, 2.7% green/zero. Each row is identical — that's the point.)
The gambler's fallacy in everyday gambling
Roulette
The classic case. "Red has come up 5 times — black is due." No, it isn't. Each spin is independent. The wheel has no memory and no obligation to balance itself out. Casinos often display the last 10–20 results on screens next to the table — not to help you, but because they know the gambler's fallacy will make you bet more.
Slots
"This machine hasn't paid out in hours — it must be about to hit." Modern slots use random number generators (RNGs) that determine each spin independently. The RNG cycles through millions of number combinations per second, and the outcome is determined the instant you press spin. A machine that hasn't paid out is no more likely to pay out on the next spin than one that just hit a jackpot. The concept of a "hot" or "cold" machine is the gambler's fallacy applied to software.
This is true for online slots as well. Every spin is generated independently by the RNG, which is regularly tested and certified by independent auditing firms. There is no hidden cycle, no "due" payout, and no way to predict when a win will come.
Lottery
"These numbers haven't been drawn in months — they're overdue." Every draw is independent. The balls have no memory. Numbers that haven't appeared recently are exactly as likely as any other combination. Lottery syndicates that choose "overdue" numbers are applying the gambler's fallacy to a system with odds of roughly 1 in 45 million.
Sports betting
"This team has lost 5 in a row — they're due a win." This one is more nuanced than roulette. Sports outcomes aren't perfectly independent — form, injuries, morale, and schedule all matter. A team that has lost five in a row might genuinely be more likely to win the next game if key players return from injury or if the opposition is weaker.
But the fallacy still applies when people assume a losing streak must reverse purely because it's "been going on too long." Regression to the mean is a real statistical phenomenon — extreme results do tend to move back toward the average over time. But it doesn't mean the very next game is more likely to be a win. The timing of any turnaround is unpredictable, and the betting market has already priced the streak into the odds. If a team's poor run is public knowledge, the odds already reflect it — there's no hidden value in betting on a reversal simply because you feel one is due.
Gambler's fallacy vs sunk cost fallacy
These two biases often appear together but are fundamentally different:
The gambler's fallacy says: "I've lost 5 times in a row, so I'm due a win." It's driven by the false belief that past random results affect future odds.
The sunk cost fallacy says: "I've already lost £100, so I need to keep playing to win it back." It's driven by the emotional weight of money already spent.
A gambler chasing losses typically experiences both simultaneously. The sunk cost fallacy provides the motivation — the need to recover what's been lost. The gambler's fallacy provides the false justification — the belief that a win is imminent. Together, they create a powerful loop that's hard to break in the moment.
Our article on the sunk cost fallacy explores the other half of this pair in detail.
The hot hand fallacy — the gambler's fallacy in reverse
The hot hand fallacy is the opposite belief: that someone on a winning streak is more likely to keep winning. In gambling, it sounds like: "I've won four bets in a row — I'm on fire, I should bet bigger."
If the outcomes are independent — as in roulette, slots, or lottery — a winning streak has no predictive power over the next result. The hot hand is just the gambler's fallacy wearing a different hat: instead of expecting a streak to end, you're expecting it to continue. Both assume past results influence future outcomes. Neither is true for games of pure chance.
In sport, the picture is different. There's some evidence that basketball shooting streaks may have a small real component — a player's confidence and rhythm can genuinely affect performance. But even in sports, people dramatically overestimate the predictive power of streaks. And in any game where the outcome is determined by a random number generator, the hot hand is pure illusion.
How to avoid the gambler's fallacy
Remember the independence rule
Before every bet, ask yourself: "Does the outcome of this bet depend on what happened last time?" For roulette, slots, dice, and lottery, the answer is always no. If the events are independent, past results are irrelevant. Full stop. No exceptions, no matter how long the streak.
Watch for the word "due"
If you catch yourself thinking "it's due," "it has to change," or "it can't keep going," that's the gambler's fallacy talking. Those feelings are real and powerful — but the maths doesn't care about feelings. The probability is the same on every spin, every draw, every roll.
Understand what randomness actually looks like
Humans expect randomness to look evenly distributed. It doesn't. Real randomness produces streaks, clusters, and patterns that look meaningful but aren't. If you flip a coin 20 times, getting 7 heads in a row at some point is not unusual — it's expected. If you looked at the results of a roulette table over an evening, you'd see runs of 4, 5, or 6 of the same colour regularly. That's normal.
Training yourself to expect streaks in random data helps neutralise the urge to bet against them. The streaks aren't signals. They're noise — and they're exactly what randomness is supposed to look like.
Use tools, not instincts
Our odds calculator shows you the actual probability of any bet. That number doesn't change based on previous results. If you're making a decision about the next bet, check the maths — not the streak.
Frequently asked questions
It's the false belief that past random events affect future ones. If a roulette wheel has landed on red 10 times in a row, the gambler's fallacy is believing that black is now more likely. It isn't — each spin is independent and the probability hasn't changed.
Written by
Ciaran McEneaney
Ciaran is a gambling industry writer based in Ireland with over a decade of experience covering the regulated betting sector. He specialises in gambling regulation, industry statistics, player protection, and responsible gambling policy. At WiseStaker, Ciaran covers UK and international gambling data, support resources, and the psychology behind gambling behaviour.
Follow on X →