Does a stop-loss reduce the house edge?

No. It reduces the number of spins you play, and the expected cost is spins multiplied by edge. The edge per spin is unchanged — in our test it was 4 cents under every rule, without exception.

Do win goals make you more likely to leave ahead?

Yes, measurably, if the goal is small. A +50 goal on a 200 bankroll raised winning sessions from 27.2% to 40.0%. A +200 goal made no difference, because it was almost never reached.

Which is better, a stop-loss or a win goal?

They optimise different things. A stop-loss minimises what you lose on average. A win goal maximises how often you go home ahead, and costs more on average to do it. Using both together gave the lowest expected cost of all five rules tested.

Is it true that a losing streak is more likely to continue?

No. Spins are independent and the game has no memory of your session. Believing otherwise is what turns a stop-loss into loss-chasing, which is the most expensive error a player can make.

What is Wald's identity?

A result stating that for a stopping rule that cannot see the future, the expected total is the expected number of bets multiplied by the expected value of one bet. It is why no session rule can produce a positive expectation on a negative-expectation game.

Stop-Loss and Win Goals: 300,000 Sessions, One Uncomfortable Answer

🛑

The short version: session rules work, and not for the reason usually given. A stop-loss at -50 cut the average session loss from 22.64 to 8.88 — entirely because it cut the spins played from 566 to 222. The price per spin never moved: 4 cents, always. A win goal did something different again: it raised the share of winning sessions from 27.2% to 40.0% while making the average loss worse than the stop-loss.

The Test

A 96% high-volatility slot model, $200 bankroll, $1.00 a spin, a maximum of 600 spins. Five rules, 60,000 simulated sessions each — 300,000 in total. For each rule we recorded the mean outcome, the median, the share of sessions finishing ahead, and the number of spins actually played.

All of it is reproducible from the simulation script. The mean is not reported from the simulation: it is calculated exactly, from Wald's identity — expected result equals spins played multiplied by the expected loss per spin, which here is 4 cents. The simulated mean is shown alongside it with a two-sigma interval, purely to demonstrate that the theory and the experiment agree.

What 300,000 Sessions Show

Rule	Spins	Expected result	Simulated	Median	Sessions ahead
No rule (600 spins)	566	-22.64	-22.81 ± 1.92	-114.86	27.2%
Stop-loss -50	222	-8.88	-8.52 ± 1.18	-50.43	11.1%
Win goal +50	402	-16.08	-15.66 ± 1.67	-95.43	40.0%
Stop-loss -50, goal +50	124	-4.96	-5.73 ± 0.90	-50.29	23.1%
Win goal +200	500	-20.00	-21.80 ± 1.83	-115.43	27.0%

Every simulated mean falls inside two standard errors of the exact figure. The model and the arithmetic agree, which is the only reason the rest of the table is worth reading.

Why the Average Moves

Divide the expected result by the spins in every row and you get the same number, to the cent.

The only relationship in the table

No rule: 566 spins x -0.04 = -22.64 Stop-loss: 222 spins x -0.04 = -8.88 Win goal: 402 spins x -0.04 = -16.08 Both: 124 spins x -0.04 = -4.96

The rule does not change the price. It changes the quantity.

This is Wald's identity, and it is not a quirk of this model — it holds for any stopping rule that does not peek at the future. You cannot construct a rule that beats it. Stopping when you are down, stopping when you are up, stopping on a Tuesday: the expected cost is always the number of spins multiplied by the edge.

So a stop-loss does help, unambiguously, and by an amount you can calculate in advance. It helps the way leaving the shop helps you spend less. That is a real mechanism and it deserves to be described accurately, because the alternative description — that stopping "locks in" something or "protects" a run — is what leads people to believe the reverse rule also works.

What a Win Goal Actually Buys

The win goal row is the interesting one, because it moves two numbers in opposite directions.

Sessions finishing ahead rise from 27.2% to 40.0% — a large, real improvement in how often you go home a winner. And yet the average loss is worse than under a stop-loss, because 402 spins is more than 222.

A win goal converts a small number of large wins into a larger number of small ones. You bank the modest win far more often, and you pay for it by continuing to play in the sessions where you never reach the goal — which are the majority, and which are the ones that end at zero.

Push the goal out to +200 and the effect vanishes entirely: 27.0% of sessions finish ahead, indistinguishable from having no rule at all. The goal is so rarely reached that it almost never triggers. A win goal only does anything if it is small enough to hit.

Notice also the median column. Under every rule the median session is a substantial loss. In the no-rule case the median is -114.86 while the mean is -22.64 — the gap between them is the jackpot, carried by a handful of sessions out of 60,000. This is the volatility gap, and it is why "average loss" is a misleading summary of what a slot evening looks like.

The Wrong Reason People Believe

The common explanation is that stopping while ahead "protects" winnings, and that a losing run is more likely to continue. Neither is true. Spins are independent. The machine has no memory of your session and no knowledge of your rule. If you resume tomorrow, the expected cost per spin is unchanged.

This matters because the false explanation licenses its own inverse. If stopping when you are down protects you, then continuing when you are down must be "due" — and the loss-chasing that follows is the single most expensive behaviour in gambling. It is the Martingale error in a different costume.

The accurate version does not license anything: a session rule reduces the bill by reducing the number of purchases. There is no protected balance, no locked-in win, no reset. There is only the count of spins, and its price.

How to Set Them Properly

Set the loss limit before you deposit, in money, not in feeling. "Down 50" is a rule. "When it stops being fun" is not — that threshold moves during the session, and it moves in the wrong direction.

Make the win goal small enough to actually reach. +25% of bankroll fired often enough to lift winning sessions from 27% to 40%. +100% fired so rarely it made no measurable difference.

Use both if what you want is the cheapest evening. Stop-loss and win goal together produced 124 spins and an expected cost of 4.96 — the lowest in the table, and by some distance.

Use the operator's own limit tools. A deposit limit you set today and cannot lift until tomorrow is a rule. One you enforce yourself at 1 a.m. is an intention.

Do not expect a rule to make the game profitable. Nothing in this table has a positive expectation, and nothing ever will. The choice is only how large the bill is.

Play with the math on your side

SlotDrop's free tools show you the real house edge, EV and variance behind every bet — before you place it.

Open the Black Box →

Reality check: no strategy on this page turns a negative-expectation game into a positive one. Strategy reduces how much the house takes and how fast your bankroll disappears — it does not guarantee profit. Only ever bet money you can afford to lose entirely.