Gambler's Fallacy

The false belief that past random events affect future random events—that a coin that has landed heads 10 times in a row is 'due' for tails. Each flip is independent; the coin has no memory. The gambler's fallacy stems from the representativeness heuristic: people expect small samples to look like the population (50/50), so a streak of heads feels 'unrepresentative' and creates the expectation of correction. In business: a string of successful hires doesn't mean the next one is 'due to fail,' and a string of failed projects doesn't mean the next one is 'due to succeed.' Past independent events don't affect future independent events.

When to use it

When streaks seem to require 'correction'; when independent events are being treated as if they're linked; when risk assessment after a string of successes produces inappropriate complacency or after failures produces inappropriate pessimism; when distinguishing between independent and dependent sequences matters for decision-making.

How it can help

For any sequence of independent events: resist the urge to predict reversal after streaks. Each event has its own probability regardless of past outcomes. However: distinguish between truly independent events (coin flips, roulette) and events where past performance IS informative (skill-based activities, correlated markets). The test: does the outcome of one event physically or causally affect the next? If no: gambler's fallacy applies—past results don't predict future results. If yes: the events aren't independent and past results ARE informative. In hiring: each candidate IS independent; past hiring success doesn't predict future failure.