Mines Game Odds and When to Cash Out (2026)

Mines Game Odds and When to Cash Out (2026)

Every Mines guide I have read answers the wrong question. They all argue about when to cash out, as though there were a right moment hiding in there somewhere, and none of them ever computes what stopping at each point is worth.

So I computed it. Three mines, a 99% paytable, stopping after one pick through eight picks, and the expected return came out the same to four decimal places every single time: 0.9900.

That result kills an entire genre of advice, and it points at the thing that actually costs you money in this game. Here is the full maths.

Quick answer: Mines odds and the cash-out rule

The numbers

What this does not mean

It does not mean cash-out timing is irrelevant to your experience. Stopping early wins often and small, holding on wins rarely and big, and the difference in how a session feels is enormous.

It also assumes the operator prices the paytable cleanly. Not all of them do, and where they do not, some stopping points genuinely are worse than others. Checking that is the one piece of real analysis available to a Mines player, and it is covered below.

Disclosure

We earn a commission when a reader opens an account through some of the links on this site. Nothing below was supplied or reviewed by an operator. Every probability is computed from the hypergeometric formula for drawing safe tiles without replacement, and the figures reproduce the published values on Wizard of Odds exactly where they overlap, which is the check that the method is right.

How Mines actually works

The grid and the seed

A 5x5 grid holds 25 tiles. You choose how many mines to bury, anywhere from 1 to 24, and the game hides them. You then click tiles one at a time. Each safe tile raises your multiplier. Hitting a mine ends the round and takes the stake.

The layout is decided before your first click. A server seed is generated, hashed, and the hash is shown to you in advance, which is what stops the game from moving a mine under your cursor. After the round the seed is revealed and you can confirm it matches.

The formula

The chance of surviving P picks with M mines on the grid is:

In plain terms: count how many ways to choose P tiles from the safe ones, divide by the ways to choose P tiles from all of them. The fair multiplier is 1 divided by that number, and whatever the game pays you below the fair multiplier is the operator's margin.

Survival odds for every pick

Three mines, the common default

Half of all rounds are over by the fifth pick. A ten pick clear, which feels like a good but ordinary round, happens once in five.

Five and ten mines

At ten mines, surviving eight picks happens roughly once in 168 rounds and pays about 166x on a 99% table. That is a lottery ticket dressed as a skill game.

Why more mines is not "riskier" in the way people mean

Raising the mine count lowers your survival chance and raises the payout by exactly the offsetting amount. The house edge does not move. What moves is the shape of your results, from frequent small wins toward rare large ones.

Choosing 10 mines over 3 is the same decision as choosing a high volatility slot over a low volatility one. Our volatility calculator sizes the bankroll each shape needs, and the same logic is laid out in slot volatility explained.

When to cash out: the answer nobody gives

Expected value at every stopping point

Here is the table the strategy guides never print. Three mines, 99% paytable, every stopping point from one pick to eight.

The expected return column is flat. The standard deviation column quadruples.

So what are you actually choosing

You are choosing how much your balance jumps around, and nothing else. The chance of getting there falls at precisely the rate the payout rises, which is what "cleanly priced" means and what makes the product constant.

The practical read

Stopping early is the right choice if your bankroll is small relative to your stake, because you need to survive long enough for a 1% edge to be the only thing working against you. Holding longer is the right choice if you are buying a shot at a big number and have accepted the cost.

Neither is more profitable. Anyone selling a specific pick count as optimal is selling a preference as a discovery. Our session simulator shows what each preference does to a real bankroll over a few hundred rounds, which is a more useful question than the one the guides argue about.

Picking a stopping rule you can actually hold

The rule matters less than sticking to it. A player who intends to stop at three picks and pushes to six after a good run has not chosen a strategy, they have chosen whichever variance the last round put them in the mood for.

Two rules survive contact with a real session. Fix the pick count before the round starts and let auto-cashout enforce it, or fix a session loss limit and let the pick count vary. Mixing both, deciding fresh each round, is what turns a flat expected value into a bankroll that empties faster than the maths predicts, because the rounds where you push are also the rounds where the stake has crept up.

Where to play Mines

Everything that moves money in Mines is set before the first click: how many mines sit on the grid, and what multipliers that particular table pays. Picking the table is the strategy people keep hunting for in click patterns.

Outside the US, Mines at 1win is the table we point readers to. The site takes players from India, Nigeria, Bangladesh, Pakistan, Kenya and dozens of other markets. From the same instant-game shelf it runs Plinko and the crash titles JetX and Lucky Jet; the full catalogue with mechanics sits in our 1win games overview. US players are not accepted there, so the licensed option named earlier applies instead.

What to check on any specific Mines table before your first bet:

• Whether the multiplier table is published for every mine count, not just one configuration
• How far the offered multiplier sits below the fair one you can compute with the formula above
• Whether the round hash is issued before the first click

The fair multiplier takes seconds to work out, and it is the only check that reliably saves money over time.

Each click is slightly worse than the one before

The marginal odds

Players tend to feel safer as a round goes on, having "got past" the dangerous part. The arithmetic runs the other way.

Each safe click removes a safe tile while leaving all three mines on the board, so the proportion of danger rises every time. The eighth click is meaningfully more dangerous than the first.

This is the opposite of the instinct that the round is "warming up", and it is the same family of error as thinking a cold streak is due to end. Our page on the gambler's fallacy covers why that feeling is so persistent.

Check what your casino actually pays

Fair multiplier against what is on screen

The single piece of genuine analysis available here takes about a minute. Work out the fair multiplier for a configuration, then read what the game offers for it.

That last pair is worth dwelling on. A single pick with 24 mines has a 1 in 25 chance, so the fair payout is exactly 25x. The tables published by Wizard of Odds show 23.75x for that bet, which works out at a 5% margin, and my own calculation reproduces their figure exactly. A game paying 24.75x for the same bet is charging 1%.

The uncomfortable part

Same game, same grid, same 1 in 25 chance, and one version takes five times as much of your money as the other. Nothing on screen announces which one you are looking at.

Run the numbers with our house edge calculator or check the survival chance directly in the win probability calculator if you would rather not compute binomial coefficients by hand.

Mines RTP is not one number

What we can verify

The BGaming figures come from our database of 3,764 tracked game entries with a provider source and verification date on each record. The spread across the category is enormous, and the one-mine case is a genuine trap: at some pick counts that setting returns less than half your stake in expectation.

That is the practical warning. One mine feels like the safe option, and on some paytables it is the worst bet on the board. Our RTP database and highest RTP games list track published figures where operators disclose them.

The one-mine trap in detail

One mine looks like the conservative setting. Twenty-four safe tiles, one bomb, and a 96% chance that any given click is fine. Players who want a low-risk grind pick it for exactly that reason.

The problem is that the payouts at one mine are tiny, and tiny payouts are where rounding and lazy paytable design do the most damage. A fair one-pick payout at one mine is 1.0417x. If the operator rounds that down to 1.03x, it has taken 1.1%. If it publishes 1.02x, it has taken 2.1%. The player cannot feel the difference between those numbers, and at 400 rounds an hour the difference is the whole session.

At the far end of that setting the pricing breaks down completely. The Wizard of Odds analysis finds one-mine returns ranging from 41% to 96% depending on how many picks you take, which means some stopping points on that setting are worse than a slot machine.

The general rule is that low-payout configurations hide margin more effectively than high-payout ones, because a few hundredths off a 1.04x multiplier is invisible while a few hundredths off a 25x multiplier is not.

Patterns, tricks and "safe tiles"

Why no pattern can exist

The search results for this game are full of tile patterns, corner strategies and "secret pro tips". They share one flaw: there is no information in the game for a pattern to be built on.

In Minesweeper, the classic desktop game, revealed squares show numbers telling you how many mines touch them. That is a genuine logic puzzle. Casino Mines shows you nothing except safe or dead, so the state of the board after four safe clicks tells you only that four safe tiles are gone.

Every unrevealed tile is equally likely to be a mine. Corners, edges, diagonals and the "cross pattern" are all the same bet.

What to verify instead

Verification is the thing that is worth your attention. Take the pre-round hash, take the revealed seed, hash it and compare. Our provably fair checker handles the hashing, and if the values match, the layout was fixed before your first click.

An operator that cannot show a pre-round hash has no answer to the "are the mines moving" question, and that question is the only rigging concern in this game that would actually matter.

What Mines costs per hour

Auto-pick makes the edge compound faster

Most Mines implementations offer an auto mode: pick a tile count, pick a stopping rule, and let it run. The expected value per round does not change, and the rounds per hour roughly triple.

A player clicking manually at three picks per round might complete 200 rounds an hour. The same rule on auto clears 600 without any sense of having bet more, because the decision was made once rather than six hundred times. Expected loss triples along with it.

If you use auto mode, set a round limit rather than a time limit. Time is the variable that auto mode quietly detaches from cost.

The hourly figures

A cautious one-pick-and-cash-out style produces very fast rounds, which means high turnover, which means the edge gets applied more times per hour. The safest looking way to play is also the way that pushes the most money through the game.

That is the real cost structure, and it applies the same way to crash games and to Plinko, both of which resolve in seconds. Size the session with our bankroll calculator before the speed decides it for you.

How Mines sits against the rest of the instant category

Mines and crash share a structure: an entry price, then a sequence of decisions that redistribute variance without touching value. Plinko and slots do not even offer that, because there is nothing to decide once the ball or the reel is released.

The practical consequence is the same in all four. The choice that moves money is which game and which paytable, made before you start, and not any decision available to you afterwards.

The verdict

Mines is a clean game with one honest decision in it, and it is not the decision everyone argues about. Cash-out timing sets your variance, not your return, and on a properly priced table every stopping point is worth the same.

The decision that does matter is which table you are playing on. Fair multipliers are computable in seconds, published paytables range from 1% to 5% margin on identical bets, and one popular setting can return as little as 41%. That gap dwarfs anything a stopping rule could recover.

Compute the fair multiplier for the configuration you like, compare it to what your game offers, and play the one that pays closest to fair. Readers in the US looking for a licensed operator can start with Bovada, and the paytable question is worth asking wherever you land.