What chasing losses means
Chasing losses means continuing to play, often with larger stakes, in order to win back money already lost. It is one of the most common warning signs of harmful play, and it is driven by a feeling that is easy to understand: the loss is uncomfortable, and winning it back seems like the way to make the discomfort go away.
The math is clear, though. A loss that has already happened has no effect on the odds of the next play, and raising stakes does not change what a game owes the player on average.
Past results do not change future odds
Take a fair coin. After five tails in a row, the chance that the next flip is tails is still 1 in 2. The chance of seeing five tails in a row at the start was 1 in 32, or about 3.1%, but once they have happened, that is history. The coin, a roulette wheel or a random number generator in a machine has no memory of what came before.
This holds when plays are independent events. In games where earlier cards or tiles change what remains, such as a deck dealt without replacement, the odds do shift, but that is because the remaining composition changes, not because of the player’s earlier bankroll losses. Nothing in a player’s own loss total alters the game.
The belief that a win is “due” after a run of losses is the gambler’s fallacy, and it is a close relative of the hot hand belief. Both read a pattern into results that are independent.
What raising the stake does
Suppose a game has an average result per unit staked. Call it the return. Whatever stake is placed, the expected result of the play is the stake times the return. In a game with negative expectation, a bigger stake means a bigger expected loss. Raising stakes after a loss therefore increases the average amount lost, since you are staking more money at the same unfavourable terms.
A doubling example, checked
A well-known way of chasing is to double the stake after each loss, so that one win recovers everything and leaves a profit of one unit. Start with a stake of 1 unit on even-money plays:
- Stakes go 1, 2, 4, 8, 16, 32 across six plays.
- If all six lose, the total lost is 1 + 2 + 4 + 8 + 16 + 32 = 63 units. A seventh stake would be 64.
- If a win comes at any point, the net result is +1 unit. For example, losing 1 + 2 + 4 = 7 and then winning 8 leaves +1.
Now suppose the pool is exactly 63 units, so the plan can go at most six plays.
A fair game (probability of winning each play 1/2). The chance of six losses in a row is (1/2)^6 = 1/64. So with probability 63/64 the series ends with +1, and with probability 1/64 it ends with −63.
Average result: 63/64 × 1 − 1/64 × 63 = 63/64 − 63/64 = 0.
The method yields small wins most of the time and a rare big loss, and the average is still zero. No edge has been created. It has only been reshaped, with the volatility concentrated into a rare event.
A game with a built-in disadvantage (probability of winning each play 0.45). The chance of six losses in a row is 0.55^6, which is about 0.0277, or 2.8%. So the series ends with +1 about 97.2% of the time.
Average result: 0.9723 × 1 − 0.0277 × 63 = 0.9723 − 1.7439 = about −0.77 units per series.
As a cross-check, each play has an expected loss of 10% of its stake (0.45 − 0.55 = −0.10). The expected total staked across the series is 1 + (2 × 0.55) + (4 × 0.55²) + (8 × 0.55³) + (16 × 0.55⁴) + (32 × 0.55⁵), which is about 7.72 units. Ten percent of that is 0.77, matching the figure above. Doubling raises the amount staked, so it raises the average loss.
Why the plan cannot be followed forever
The doubling example also shows two practical limits.
A pool is finite. The stakes grow quickly: six losses need 63 units, ten would need 1,023, and twenty losses would need over a million units. Few people can supply that.
Limits cap the stakes. Many venues and services set maximum stakes, so the sequence can be stopped before the recovery arrives.
In other words, the apparent safety comes from the long series of small wins. The rare series that fails costs far more than all those wins combined, and it is the one most likely to take someone beyond what they can afford. That connects to risk of ruin, where unlimited play against a built-in disadvantage ends in depletion.
How it distorts choices
Beyond the arithmetic, chasing changes behaviour in predictable ways:
- It turns entertainment spending into a quest to get back to even, which makes ordinary losing results feel unacceptable.
- It encourages larger stakes, longer sessions and riskier choices, because the goal is no longer enjoyment but recovery.
- It leads people to borrow, to top up, or to reach into money meant for other things.
- It is reinforced by the sunk cost feeling that the money already lost must be justified.
- Because wins sometimes do arrive, they strengthen the belief that the approach works. It reflects the chance and psychology of intermittent rewards.
What helps instead
Rather than trying to undo a loss, the useful step is to accept it as the price of the activity and stop. Setting a loss limit in advance, taking a break after reaching it, and treating the lost amount as spent are all practical ways of breaking the loop. A fixed limit set beforehand takes the decision out of the moment.
If you recognise this pattern, such as playing to recover money, hiding losses, or borrowing, it is a good time to speak to someone. Local problem gambling helplines and support services are usually free and confidential, and a doctor or counsellor can also help. Outcomes in games of chance are uncertain, and no sequence of stakes is guaranteed to recover a loss.