Two beliefs that point in opposite directions
After a streak, people tend to make a prediction. Some expect the streak to stop. Others expect it to carry on. The first expectation is known as the gamblers fallacy, the second as hot hand belief. They look like opposites, and in a way they are, but both come from the same habit: treating a short run of results as a signal about what happens next.
Whether either belief is a mistake depends on what is producing the results. That is the key to telling them apart.
The gamblers fallacy: expecting reversal
The gamblers fallacy is the belief that after a run of one outcome, the opposite becomes more likely. A roulette-style example, in general terms: after several reds in a row, a player feels black is “due”.
When events are independent, meaning each one has no memory of the last, this is simply wrong. A fair coin that has landed heads six times has the same one-in-two chance of heads on the seventh flip. Dice and well-made random number generators work the same way.
The belief feels reasonable because people expect small samples to look like the long-run average. Over many flips, heads and tails even out in proportion, so a run of heads feels like an imbalance that needs correcting. But the law of large numbers works by dilution, not by compensation. Future flips do not make up for past ones; they just make the early imbalance a smaller share of the total.
Streaks are also more common than intuition suggests. Any particular run of five heads starting at a chosen flip has a chance of only 1 in 32, but across a long sequence of flips there are many possible starting points, so a run of several identical results somewhere in the sequence is quite likely. Runs are a normal feature of randomness.
Hot hand belief: expecting continuation
Hot hand belief is the expectation that a person or object who has succeeded several times is more likely to succeed again. The name comes from basketball, where fans and players often feel that a shooter who has made a few shots in a row is “on fire”.
Unlike the coin case, this belief is not obviously wrong, because a human shooter is not a coin. Confidence, rhythm, health and defensive attention might all change from one shot to the next. If any of these effects exist, a streak could carry real information.
What the evidence says
A well-known study from the 1980s examined basketball shooting records and found little sign that a made shot made the next one more likely. It was widely taken as evidence that the hot hand was a cognitive illusion, with observers reading patterns into ordinary random variation.
Later work challenged parts of that conclusion. Researchers pointed out a subtle statistical issue in how streaks in short sequences are analysed, which can make real streakiness look weaker than it is. Re-analyses have suggested that a modest hot hand effect may exist in some settings, though how large it is, and in which settings, is still debated. A cautious summary is:
- In independent chance events, such as coin flips, dice and fair roulette-type games, neither belief is justified.
- In skill activities, a small streak effect may exist, but people usually overestimate it.
- Observers tend to see streaks as meaningful even when the effect is weak or absent.
The disagreement is a useful reminder that what looks like a pattern can need careful analysis, and that the accurate position is often “less than it feels, but not necessarily zero” for skill and “none” for pure chance.
Telling the two situations apart
A practical test is to ask what produces each result.
| Situation | Source of results | Streak tells you |
|---|---|---|
| Fair coin, dice, random draw | Independent chance | Nothing about the next outcome |
| Shuffled deck, cards not replaced | Dependent chance | Something, but through the cards left, not “luck” |
| Human skill task | Chance plus ability and state | Possibly a little, often less than it feels |
The second row is worth a note. If cards are dealt without replacement, a run of low cards does change the composition of what remains, so later outcomes are not independent. That is a real effect with a mathematical reason, not an instance of either belief, and it connects to the idea of drawing without replacement.
When systems blur the lines
Some games are built so that streaks matter. Video game developers sometimes adjust random results so that long runs of bad luck are interrupted, using hidden streak breakers. In those games, the gamblers fallacy can be partly correct, because the game really does remember. The reverse can also apply, where a game keeps a hidden score of recent outcomes. These are design choices of specific systems, not features of chance itself.
Why it matters for decisions
Both beliefs can push choices in costly directions. The gamblers fallacy can tempt someone to raise stakes after losses, expecting a reversal, which is a version of chasing losses. Hot hand belief can encourage escalating bets during a winning run, on the feeling that confidence or luck is on your side. In games with a house edge, neither feeling changes the expected result, and outcomes remain uncertain throughout.
A calm approach is to decide in advance how much time and money you are willing to spend, and to treat each independent outcome as a fresh event. If streaks, in either direction, are driving your choices, that is a reasonable cue to take a break. Anyone who finds this difficult to do can seek confidential help from a local support service.