- With a 50% hit rate, over 500 bets a streak of about 8 straight losses is typical (exact calculation).
- A streak of 10 losses with a 1% stake costs 10% of your bankroll; with a 5% stake, 50%.
- You cannot avoid streaks: the only thing you control is how much you risk on each bet.
How many losses in a row are normal?
More than you imagine.
If you bet often, you will have ugly streaks.
It is not exceptional bad luck.
It is math.
In this article we calculate it with code, with no made-up numbers.
First, the idea.
Every bet is a loaded coin: it wins with a certain probability.
If you flip a coin many times, sooner or later several heads come up in a row.
The same happens with losses.
Every example in this article is labeled as an example.
The hit-rate figures (40%, 50% and 55%) are assumptions to illustrate, not anyone's results.
This is educational content, for adults 18 and over only.
How we calculated it (so you can redo it)
We used two independent methods.
For streaks, an exact dynamic-programming calculation, which gives the true probability and not an estimate.
For the bankroll, a Monte Carlo simulation.
The simulation used 20,000 runs with a fixed seed.
Each run is a full sequence of 100, 500 or 1,000 bets.
In each sequence we measured the worst drop from a peak.
In the simulation we used fair odds, with no bookmaker margin.
For example, 2.00 for a 50% hit rate and 2.50 for 40%.
That way we measure variance only, without the effect of an edge or a disadvantage.
With a 50% hit rate: probability of a streak of n losses
Let us start with an even example.
Imagine you win half of your bets.
The table shows the probability that, at some point, a streak of at least n straight losses appears.
| Streak of at least | 100 bets | 500 bets | 1,000 bets |
|---|---|---|---|
| 5 losses | 81.0% | 100% | 100% |
| 8 losses | 17.0% | 62.5% | 86.1% |
| 10 losses | 4.4% | 21.5% | 38.5% |
| 12 losses | 1.1% | 5.8% | 11.4% |
| 15 losses | 0.1% | 0.7% | 1.5% |
Look at the 8-loss row.
Over 100 bets it happens about one time in six.
Over 1,000 bets it happens in more than eight out of ten sequences.
The trend is what matters.
The more bets you place, the more certain a long streak becomes.
Time is on the side of the bad streak.
What if you win less or more? The same math with 40% and 55%
The hit rate changes the picture a lot.
A bettor who goes for high odds wins less often and suffers longer streaks.
One who plays low odds wins more often and the streaks are shorter.
| With a 40% hit rate | 100 bets | 500 bets | 1,000 bets |
|---|---|---|---|
| 8 losses or more | 49.0% | 97.1% | 99.9% |
| 10 losses or more | 20.5% | 70.5% | 91.5% |
| 12 losses or more | 7.7% | 35.1% | 58.2% |
| 15 losses or more | 1.6% | 8.8% | 17.0% |
| With a 55% hit rate | 100 bets | 500 bets | 1,000 bets |
|---|---|---|---|
| 8 losses or more | 8.4% | 36.9% | 60.4% |
| 10 losses or more | 1.7% | 8.8% | 17.0% |
| 12 losses or more | 0.3% | 1.8% | 3.7% |
| 15 losses or more | 0.0% | 0.2% | 0.3% |
See how the table shifts.
With a 40% hit rate, a streak of 10 is almost a certainty over 1,000 bets.
With 55% it is possible, but it happens in fewer than one in five sequences.
What is the typical longest streak?
Another way to look at it is the average.
It is the longest streak that, on average, appears in a complete sequence.
It also comes from the exact calculation.
| Hit rate (example) | 100 bets | 500 bets | 1,000 bets |
|---|---|---|---|
| 40% | 7.9 losses | 11.0 losses | 12.4 losses |
| 50% | 6.0 losses | 8.3 losses | 9.3 losses |
| 55% | 5.3 losses | 7.3 losses | 8.1 losses |
So the normal range is a streak of 6 to 12 losses.
It depends on your hit rate and how many bets you place.
If you hit a streak of 8, you are not looking at a rare case.
How much of your bankroll a streak eats
So far we have talked about bets.
But what hurts is the money.
With a flat stake, the math is simple: each loss subtracts the same amount.
A streak of n losses costs n times the stake.
That table needs no simulation, it is a multiplication.
| Streak (example) | 1% stake | 2% stake | 5% stake |
|---|---|---|---|
| 5 losses | 5% | 10% | 25% |
| 8 losses | 8% | 16% | 40% |
| 10 losses | 10% | 20% | 50% |
| 12 losses | 12% | 24% | 60% |
| 15 losses | 15% | 30% | 75% |
The percentages are of the starting bankroll.
In practice, if your stake is a percentage of the current bankroll, each loss subtracts a little less.
But the logic is the same: high stake, big drop.
Worst expected drawdown: a 20,000-run simulation
A streak is not the whole story.
There are also stretches where you win and lose mixed together and still go down.
That is why we measured the worst drop from a peak in each sequence.
The table shows the median and the 95th percentile, in stake units.
The median is the typical case.
The 95th percentile is a bad scenario: only 1 in 20 sequences is worse.
| Hit rate (fair odds) | 100 bets | 500 bets | 1,000 bets |
|---|---|---|---|
| 40% (odds 2.50) | 13.0 · p95 26.0 | 30.5 · p95 60.0 | 43.5 · p95 85.5 |
| 50% (odds 2.00) | 11.0 · p95 21.0 | 24.0 · p95 49.0 | 35.0 · p95 70.0 |
| 55% (odds 1.82) | 9.5 · p95 19.6 | 22.5 · p95 44.7 | 32.3 · p95 63.9 |
Read the 50% row at 500 bets.
The typical drop is 24 stake units.
In a bad scenario, 49 units.
Now convert units into a percentage of your starting bankroll.
This is where the real difference between a prudent stake and an aggressive one shows up.
| Typical drop (50%) | 1% stake | 2% stake | 5% stake |
|---|---|---|---|
| Over 100 bets | 11% | 22% | 55% |
| Over 500 bets | 24% | 48% | 120% |
| Over 1,000 bets | 35% | 70% | 175% |
A value above 100% means that, with a flat stake in units of the starting bankroll, the bankroll ran out.
With a stake of 5% of the starting bankroll, the typical drop over 500 bets already exceeds all your capital.
That is the underlying reason not to bet big.
What to do to survive a streak
You cannot choose whether a bad streak comes.
You can choose how much it costs you.
These are the decisions you do control.
- Small, flat stake: between 1% and 2% of your bankroll per bet, as we show in the article on flat vs variable staking.
- A separate bankroll: money you can lose without touching the rent, as explained in what a bankroll is.
- Record everything: a tracking spreadsheet shows you the streak in numbers, not in feelings.
- Daily and weekly loss limits: set them before you bet, not during the streak.
- A mandatory pause: after a set number of straight losses, stop for the day.
To calculate the size of each bet from your estimated edge, try the Kelly calculator.
Use a fraction of the result, because full Kelly has very deep drawdowns.
Why the Martingale makes everything worse
The Martingale proposes doubling your stake after every loss.
It sounds logical: the next win recovers everything.
The math of streaks shows the problem.
With a 50% hit rate, a streak of 8 losses has a 17.0% probability over 100 bets.
If you doubled your stake each time, your ninth bet would be 256 times the initial stake.
You would already have lost 255 units.
You do not need a strange streak.
You need a normal one.
The Martingale turns a nuisance into a catastrophe.
The Martingale does not change the probability of winning the next bet.
It only changes how much you risk.
Read the full analysis in our article on the Martingale in sports betting.
How this connects to expected value
A bet with a real edge still suffers streaks.
Expected value speaks about the long-run average, not the next bet.
That is why you need a large sample to see the result.
If you want to practice the math, use the expected value calculator.
And to understand why a small sample misleads, read how many bets you need to know if you are winning.
At Gambeta, picks are recorded before the match, with their odds.
You can see real streaks in the public record and read the daily context in the Betting Journal.
What to read next
You now know streaks are normal.
The next step is knowing when a streak is no longer just chance.
That depends on the size of the sample.
Continue with how many bets you need to know if you are winning.
If you want the full framework, the AI for betting guide brings everything together in one place.
And remember: bet only what you can afford to lose, and if gambling stops being entertainment, see the responsible gambling guide.
Frequently asked questions
How many lost bets in a row are normal?
It depends on how many bets you place and on your hit rate.
With a 50% hit rate and 500 bets, the usual longest streak is about 8 losses.
With a 40% hit rate, over 500 bets the usual figure is 11.
All of these are calculations from this article.
Does a losing streak mean my method is bad?
Not necessarily.
A long streak shows up even when the method is good, because chance produces it on its own.
To judge a method you need a large sample, as we explain in the article on how many bets you need.
How much bankroll can I lose in a bad streak?
With a flat stake, a streak of n losses costs you n times your stake.
If you bet 1% and lose 10 in a row, you lose 10% of your bankroll.
If you bet 5%, you lose 50%.
What stake size helps you survive streaks?
A low one.
In our simulations, with a 1% stake the typical worst stretch over 500 bets eats roughly a quarter of the bankroll.
With 5%, that drop exceeds the entire bankroll.
Does raising your stake after a loss help you recover?
No.
The Martingale and other recovery systems turn a normal streak into a huge loss.
They do not change the probability of the next bet, and they multiply what you risk.
Are winning streaks normal too?
Yes, for the same reason.
If the chance of winning is 50%, the longest winning streak behaves just like the longest losing one.
Be careful: they also mislead you, because they make you believe you have an infallible method.