Mines odds and probability: the math behind every tile
Mines odds explained: the chance each tile is safe, the formula for safe picks in a row, a table for 1, 3, 5 and 10 mines, and how the house edge is set.
Neste artigo 6 seções
Mines looks like a game of nerve, but underneath it’s a counting problem. There are 25 tiles, a fixed number of mines, and you reveal tiles without putting them back. That’s enough to work out the exact chance of surviving any number of clicks, the fair payout for it, and how much the game keeps for itself.
This guide does the math step by step with worked examples. If you’re new to the game, start with how to play Mines.
What are the odds the next tile is safe?
The chance the next tile is safe equals the number of safe tiles still covered divided by the number of tiles still covered. With m mines and k safe tiles already found, that’s (25 − m − k) ÷ (25 − k).
Take 3 mines. The first tile is safe 22 times in 25, or 88%. If it was safe, 24 tiles remain and 21 of them are safe, so the second click is safe 21 in 24 (87.5%). The third is 20 in 23 (about 87.0%), and so on. Each click is a little riskier than the last, because the mines stay put while the safe tiles get used up.
What is the probability of several safe picks in a row?
The chance of revealing k safe tiles in a row with m mines on the board is C(25 − m, k) ÷ C(25, k), where C(n, k) is the number of ways to choose k items from n. It’s the chance that all k tiles you pick come from the safe ones.
C(n, k) is the binomial coefficient, “n choose k”, which NIST’s Digital Library of Mathematical Functions1 defines as the number of ways of choosing k objects from n distinct objects without regard to order. Drawing tiles from the grid is sampling without replacement from a population of two kinds of item (safe and mined), which is exactly the situation the hypergeometric distribution2 describes in Michigan State University’s statistics course notes. Revealing k tiles with none of them a mine is the simplest case of that distribution.
Worked example: 3 mines, 5 safe picks. There are 22 safe tiles. C(22, 5) = 26,334 ways to pick 5 safe tiles, out of C(25, 5) = 53,130 ways to pick any 5 tiles. So the chance is 26,334 ÷ 53,130 = 57/115, about 49.6%. You can get the same answer by multiplying the step-by-step odds:
22/25 × 21/24 × 20/23 × 19/22 × 18/21 = 57/115 ≈ 0.4957
In other words, even with only 3 mines on the board, getting five tiles in without hitting one is roughly a coin flip.
What are the odds for 1, 3, 5 and 10 mines?
The table gives the chance of surviving each number of reveals, computed with the formula above. Read across a row to see how much faster the odds fall with more mines.
| Safe reveals in a row | 1 mine | 3 mines | 5 mines | 10 mines |
|---|---|---|---|---|
| 1 | 96.0% | 88.0% | 80.0% | 60.0% |
| 2 | 92.0% | 77.0% | 63.3% | 35.0% |
| 3 | 88.0% | 67.0% | 49.6% | 19.8% |
| 4 | 84.0% | 57.8% | 38.3% | 10.8% |
| 5 | 80.0% | 49.6% | 29.2% | 5.7% |
Two things stand out. With 1 mine the odds fall by exactly 4 points per click: surviving k clicks just means the single mine is among the 25 − k tiles you didn’t pick, a chance of (25 − k) ÷ 25. And 3 mines with 5 picks has the same odds as 5 mines with 3 picks (49.6%): the formula is symmetric, so swapping the number of mines and the number of picks gives the same answer.
At the extremes, 1 mine with all 24 safe tiles found and 24 mines with the single safe tile found are both 1 in 25. With 12 mines, finding all 13 safe tiles is 1 in C(25, 12) = 5,200,300.
What is a fair multiplier in Mines?
A fair multiplier is 1 divided by the probability of getting that far. If something happens 49.6% of the time, a fair game would pay about 2.02 times the stake for it, so that on average you’d get your stake back exactly.
Here is the 3-mine walk-through with the fair multiplier at each step:
| Safe reveals | Odds of that click | Chance of getting this far | Fair multiplier |
|---|---|---|---|
| 1 | 22/25 | 88.0% | 1.14× |
| 2 | 21/24 | 77.0% | 1.30× |
| 3 | 20/23 | 67.0% | 1.49× |
| 4 | 19/22 | 57.8% | 1.73× |
| 5 | 18/21 | 49.6% | 2.02× |
How does the house edge change the multiplier?
The game pays the fair multiplier minus its margin: shown multiplier = (1 − house edge) × fair multiplier. With a 1% edge, the 3-mine, 5-pick payout drops from about 2.02× to about 2.00×; with a 3% edge, to about 1.96×.
That makes the average return on every round 1 − house edge, whatever you choose. The chance of getting paid is p, the payout is (1 − edge) ÷ p, and p × (1 − edge) ÷ p = 1 − edge. A game with a 1% edge returns 99% of what’s staked on average, which is what “return to player” means. The UK Gambling Commission’s guide to return to player3 stresses that this is “an average achieved over a significant number of game plays and not each time” you play.
Does the number of mines change your expected loss?
No. If the same edge is applied throughout the payout table, every mine count and every cash-out point returns the same 1 − edge on average. What changes is the spread of results, which statisticians call variance.
With 1 mine and one click you get a small payout 96% of the time. With 10 mines and five clicks you get a large payout only 5.7% of the time and lose the stake the other 94.3%. Both lose the same fraction of the stake on average. Small rounding in real payout tables can shift this slightly from one step to the next.
Your total expected loss grows with the total amount you stake, so the number of rounds matters far more than how you play each one. Our Mines strategy guide looks at what that means in practice, and provably fair explains how some games let you check the mines really were placed at random.
Fontes
- NIST Digital Library of Mathematical Functions §26.3: Binomial coefficients dlmf.nist.gov
- Michigan State University STT 231: Hypergeometric distribution stt.msu.edu
- UK Gambling Commission: Return to player, how much gaming machines pay out gamblingcommission.gov.uk
- UK Gambling Commission: RTS 3, rules, game descriptions and the likelihood of winning gamblingcommission.gov.uk
Facts last checked November 2026. Spotted something out of date? Tell us at editor@mines.now.