The Martingale System: Why Doubling After a Loss Always Fails
Martingale is the most famous betting system in the world, and the most misunderstood. The rule is simple: after every loss you double your stake, and after every win you reset to your base unit. Because each win recovers all previous losses plus one base unit of profit, the system appears to guarantee a win — until you do the math on what a losing streak costs.
This page works through exactly how Martingale behaves, why the cost of recovery grows exponentially, how much bankroll it really demands, and why a finite bankroll plus table limits make eventual ruin a mathematical certainty rather than bad luck.
How the Martingale system works
You pick a base stake — say $10 — and bet it on a near-even-money market. If you win, you bank one unit of profit and start again at $10. If you lose, you double the next stake to $20. Keep doubling after each loss; the first win in the streak recovers every dollar lost plus your original $10 profit.
The logic is airtight for a single sequence: 2 × the last stake always exceeds the sum of every stake before it, so one win clears the slate. The flaw is not in the recovery — it is in how fast the required stake explodes when the win does not come.
- Choose a base unit and an even-money bet.
- After a loss, double the next stake.
- After a win, reset to the base unit.
- A single win recovers the whole streak plus one base unit.
The math: exponential stakes
After n consecutive losses your next stake is base × 2ⁿ, and the total you have already risked across the streak is base × (2ⁿ − 1). Both grow exponentially. From a $10 base, the seventh bet in a streak is $640, and you have already staked $630 — risking $1,270 in total to win the original $10.
The table shows a single losing run from a $10 base. Notice the "to win" column never changes: every escalating bet is chasing the same $10 of profit.
| Loss # | Stake | Cumulative risked | Still chasing |
|---|---|---|---|
| 1 | $10 | $10 | $10 |
| 2 | $20 | $30 | $10 |
| 3 | $40 | $70 | $10 |
| 4 | $80 | $150 | $10 |
| 5 | $160 | $310 | $10 |
| 6 | $320 | $630 | $10 |
| 7 | $640 | $1,270 | $10 |
| 8 | $1,280 | $2,550 | $10 |
Bankroll requirement and risk of ruin
To survive n losses in a row you need a bankroll of base × (2ⁿ − 1) just to place the next bet. A $1,000 bankroll at a $10 base covers six straight losses but not the seventh — the doubling outruns almost any realistic bankroll within a handful of steps.
On a true coin-flip, seven losses in a row has roughly a 1-in-128 chance on any given sequence (0.5⁷ ≈ 0.8%). That sounds safe until you realise you face that risk on every sequence you run. Over a few hundred sequences the streak is not unlikely — it is expected. Real markets make it worse: vig pushes your win probability below 50%, and house or table limits cap how far you can double, removing the recovery bet entirely.
| Streak length | Bankroll needed ($10 base) | Chance per sequence |
|---|---|---|
| 4 losses | $150 | 1 in 16 (6.3%) |
| 6 losses | $630 | 1 in 64 (1.6%) |
| 8 losses | $2,550 | 1 in 256 (0.4%) |
| 10 losses | $10,230 | 1 in 1,024 (0.1%) |
Why it still loses
Martingale does not change the expected value of any bet, so the sum of your bets stays negative. What it does is reshape the outcomes into many small wins and one devastating loss. The small wins arrive often enough to feel like a system; the rare blow-up wipes out hundreds of them at once and then some.
It is a classic case of trading a high probability of a small gain for a low probability of a catastrophic loss — a trade that is, on average, a money-loser. No amount of discipline fixes that, because the problem is the math, not the execution.
Pros
- Simple to understand and run with no calculation mid-session.
- Produces frequent small wins, so most sessions end in the green.
- Recovers an entire losing streak with a single winning bet.
Cons
- Stakes grow exponentially — a modest streak demands a huge bankroll.
- Table and account limits can block the recovery bet exactly when you need it.
- One long losing run erases hundreds of small wins and can end your bankroll.
- Does nothing to change the negative expected value of the underlying bets.
Frequently asked questions
- Does the Martingale system work in the long run?
- No. It wins small amounts often but exposes you to a rare, exponentially large loss that, on average, outweighs all the small wins. Because doubling does not change the expected value of any bet, the sequence stays negative; a finite bankroll and table limits make eventual ruin a mathematical certainty.
- How big a bankroll does Martingale need?
- To cover n consecutive losses you need base × (2ⁿ − 1). A $10 base needs $630 to survive six losses and $10,230 to survive ten. Because the requirement doubles with each step, no realistic bankroll can outrun a long enough streak.
- Why do casinos and sportsbooks allow Martingale?
- Because it cannot beat them. Table limits and account caps stop the doubling before a bettor can recover a deep streak, and the house edge means the expected result is a loss regardless of staking. The system is, if anything, good for the book.
- Is there a safer version of Martingale?
- Anti-Martingale (the Paroli system) doubles after wins instead of losses, capping your downside but never producing the catastrophic loss — though it also never beats the edge. For pure risk control, flat or fractional-Kelly staking is far safer than any doubling scheme.
Helpful calculators
Glossary terms
Other betting systems
Educational content for informational purposes only. Not betting or financial advice. Please gamble responsibly. 21+ where applicable.