Losing Streak Calculator

Variance. A 53% bettor wins 53% of bets long-run, but the order is random. Over 500 bets you'll see runs of 5, 7, 10 losses in a row purely from how the wins and losses are sequenced. The math doesn't predict any particular streak — it predicts that streaks of various lengths will occur with calculable frequency.

The intuition that "the model is broken" after a 10-loss streak is usually wrong. The right question isn't "is this streak normal?" — it's "is the streak frequency consistent with my expected win rate?" Cold streaks are how positive-EV bettors lose money on the short timescale.

P(next N bets all lose) is the probability that — starting from right now — your next N bets are all losers. For a 50% bettor and 7 bets, that's 0.5⁷ = 0.78%.

P(streak appears in sequence) is the probability that AT LEAST ONE losing streak of length N occurs somewhere in your total sequence of bets. This is always higher, often dramatically so. Over 200 bets, the probability of seeing a 7-loss streak somewhere is around 60% — even though the single-shot probability is under 1%.

Most bettors confuse these. The second is what matters for emotional planning over a season.

Approximate formula: log(N × q) / log(1/q), where N = total bets, q = loss probability.

For a 50/50 bettor over 200 bets: expected longest losing streak ≈ 6-7. Over 500 bets: ≈ 8. Over 1,000 bets: ≈ 9-10. The growth is logarithmic — even a 10x increase in volume only adds 3-4 to the expected longest streak.

This number is what to mentally prepare for, not the streak you're "due" for after a hot run. Variance doesn't owe you anything.