Mines Demo Guide: Exact Odds and Multipliers for Every Pick, and How to Use Free Play
Quick answer
A mines demo lets you play the 5×5 Mines game with fake money. The odds are exact: with m mines, the chance of k safe picks in a row is C(25−m, k) ÷ C(25, k). With 3 mines, 5 safe picks happen 49.57% of the time. A demo is ideal for learning these numbers but can't predict real results.
| Grid | 5×5 = 25 tiles, you choose 1 to 24 mines |
|---|---|
| Chance of k safe picks | C(25−m, k) ÷ C(25, k) |
| 3 mines, 5 picks | 49.57% chance; 1.9974x at a 1% edge |
| 1 mine, first pick | 96% chance; 1.0312x at a 1% edge |
| 10 mines, 5 picks | 5.65% chance; 17.5154x at a 1% edge |
| Demo mode | Fake money; same rules, no real wins or losses |
A mines demo is the free-play version of Mines, the crypto casino game where you uncover tiles on a 5×5 grid and try not to hit a hidden mine. You play with fake money, so nothing is at stake, which makes it the best place to learn how the odds work. And the odds are exact: with 3 mines, the chance of picking 5 safe tiles in a row is 49.57%, and every multiplier follows from that kind of calculation.
Below are the full odds and multiplier tables for 1, 3, 5 and 10 mines, a worked example of the formula, why demo mode is genuinely useful, and how provably fair Mines lets you check real games.
How the Mines casino game works
- Choose the number of mines, from 1 to 24. The rest of the 25 tiles are safe.
- Place a bet (in a demo, with play credits).
- Reveal tiles one at a time. Each safe tile raises your multiplier.
- Cash out whenever you like, or keep going. Hit a mine and the round ends with the stake lost.
Mine positions are set before you click. In provably fair versions they're generated from a server seed that's committed (as a hash) before the round, your client seed and a nonce. One major operator's game documentation describes its Mines game on exactly this 5×5 grid.
The odds and the math
Picking tiles is drawing without replacement: once you've revealed a safe tile, there's one fewer safe tile and one fewer tile overall. That's the hypergeometric setting, and for "all picks safe" it reduces to a ratio of binomial coefficients:
P(k safe picks with m mines) = C(25 − m, k) ÷ C(25, k)C(n, k) is the number of ways to choose k tiles from n. The top counts all-safe selections; the bottom counts all selections.
Worked example: 3 mines, 5 picks.
- Safe tiles: 25 − 3 = 22. Ways to choose 5 of them: C(22, 5) = 26,334.
- Ways to choose any 5 tiles: C(25, 5) = 53,130.
- P = 26,334 ÷ 53,130 = 49.565%.
You get the same answer pick by pick: 22/25 × 21/24 × 20/23 × 19/22 × 18/21 = 0.4957.
The multiplier is the inverse of that chance, trimmed by the house edge:
fair multiplier = 1 ÷ P; with a 1% edge, multiplier = 0.99 ÷ P3 mines, 5 picks: 1 ÷ 0.49565 = 2.0175x fair, or 0.99 ÷ 0.49565 = 1.9974x with a 1% edge.
House edges vary by game. We use 1% as a worked example throughout; check the multiplier your game actually shows.
Mines odds tables: 1, 3, 5 and 10 mines
| Safe picks in a row | 1 mine | 3 mines | 5 mines | 10 mines |
|---|---|---|---|---|
| 1 pick | 96.00% | 88.00% | 80.00% | 60.00% |
| 2 picks | 92.00% | 77.00% | 63.33% | 35.00% |
| 3 picks | 88.00% | 66.96% | 49.57% | 19.78% |
| 4 picks | 84.00% | 57.83% | 38.30% | 10.79% |
| 5 picks | 80.00% | 49.57% | 29.18% | 5.65% |
| 6 picks | 76.00% | 42.13% | 21.89% | 2.83% |
| 8 picks | 68.00% | 29.57% | 11.65% | 0.59% |
| 10 picks | 60.00% | 19.78% | 5.65% | 0.09% |
| 12 picks | 52.00% | 12.43% | 2.42% | 0.00875% |
| 15 picks | 40.00% | 5.22% | 0.47% | 0.00003% |
| Safe picks in a row | 1 mine | 3 mines | 5 mines | 10 mines |
|---|---|---|---|---|
| 1 pick | 1.0312x | 1.1250x | 1.2375x | 1.6500x |
| 2 picks | 1.0761x | 1.2857x | 1.5632x | 2.8286x |
| 3 picks | 1.1250x | 1.4786x | 1.9974x | 5.0044x |
| 4 picks | 1.1786x | 1.7120x | 2.5848x | 9.1747x |
| 5 picks | 1.2375x | 1.9974x | 3.3926x | 17.5154x |
| 6 picks | 1.3026x | 2.3498x | 4.5235x | 35.0308x |
| 8 picks | 1.4559x | 3.3485x | 8.5001x | 166.3962x |
| 10 picks | 1.6500x | 5.0044x | 17.5154x | 1,078x |
| 12 picks | 1.9038x | 7.9615x | 40.8692x | 11,315x |
| 15 picks | 2.4750x | 18.9750x | 208.7250x | 3,236,072x |
How fast the odds fall with 3 mines:
And how many picks you can make before the chance of a clean run drops below 50%:
| Mines | First pick safe | First-pick multiplier (1% edge) | Most picks with ≥50% chance of all safe |
|---|---|---|---|
| 1 mine | 96.00% | 1.0312x | 12 |
| 2 mines | 92.00% | 1.0761x | 7 |
| 3 mines | 88.00% | 1.1250x | 4 |
| 5 mines | 80.00% | 1.2375x | 2 |
| 10 mines | 60.00% | 1.6500x | 1 |
| 24 mines | 4.00% | 24.7500x | 0 |
Two things jump out. With 1 mine you can make 12 safe picks and still have a better-than-even chance (52%), but the reward is only 1.90x. With 10 mines, even the first pick fails 40% of the time. Higher multipliers are always paid for with lower probabilities, at the same edge.
Why a mines demo is useful
Playing Mines with fake money has real benefits:
- You learn the curve. Watching the multiplier climb while the odds in the tables above fall makes the trade-off concrete.
- You can check the game's edge. Multiply any multiplier the demo shows by the matching chance from our table. For 3 mines and 5 picks, a game paying 1.9974x gives 1.9974 × 0.49565 = 99%, a 1% edge. If the product is lower, the edge is higher.
- You find your settings without cost. Try different mine counts and cash-out points to see which volatility suits you.
- You test your discipline. Practise cashing out at a planned point instead of clicking "one more".
- You see variance in action. Play 20 demo rounds with 3 mines aiming for 5 picks. Our table says you'd clear about half of them over the long run, but over just 20 rounds you may clear far more or far fewer. Seeing that swing with fake money makes it less of a shock with real money.
What a demo can't do is predict anything. Each round's mines are placed independently. A clean run in demo mode says nothing about the next real round, and believing otherwise is the gambler's fallacy. We also can't verify how any particular demo is coded, so treat "mines gambling fake money" as practice for the interface and the math, not as evidence about a real-money game.
Provably fair Mines basics
In provably fair Mines, before you play:
- The operator generates a secret server seed and shows you its hash.
- You set (or accept) a client seed.
- Each round uses a nonce, a counter that goes up by one per bet.
The mine positions are derived from those three values. After you rotate to a new server seed, the old one is revealed, and you can recompute the hash and the mine layout for every past round. One major operator's implementation notes explain how the hash output is turned into numbers and then into game results. If the revealed seed doesn't match the hash you were shown, the results weren't fair.
Mines strategy: what works and what doesn't
What doesn't work:
- "Safe" tiles or patterns. Corners, edges and repeated layouts have no advantage; positions are random each round.
- Predictor apps and bots. They can't reverse the hash, so anyone selling one is selling something that can't work.
- Raising stakes after a loss. Same edge, bigger bets, faster losses.
What helps:
- Pick a cash-out target before the round using the tables above.
- Match mine count to your risk appetite. Few mines for frequent small wins, many for rare big ones.
- Keep bets small relative to your bankroll.
Playing Mines with USDT
Mines is native to crypto casinos, and many players use USDT so their bankroll stays steady in dollar terms. Keep network fees small relative to your deposit; our USDT payment guides compare networks. When you move from demo to real play, verify a few rounds with your own client seed.
If you like games where every probability is visible, our crypto dice odds guide lets you set the win chance directly (and the dice odds calculator does the arithmetic), while our crash gambling guide shows the same 99%-return logic applied to a rising multiplier. For Plinko's version of the math, see Plinko landing odds. All Mines guides live on our Mines hub.
Common mistakes
- Treating a demo winning streak as a strategy.
- Choosing many mines for the big multiplier without checking how rarely it hits.
- Forgetting to set a cash-out point and giving back a winning run.
- Skipping provably fair verification on real-money rounds.
Frequently asked questions
What is a mines demo?
A mines demo is a free-play version of the Mines casino game that uses fake money. You choose how many mines hide on a 5×5 grid, reveal tiles one by one, and cash out before you hit a mine. It's useful for learning how the multiplier grows and how quickly your odds fall, without risking real funds.
What are the odds in Mines?
With m mines on 25 tiles, the chance that your first k picks are all safe is C(25−m, k) ÷ C(25, k). For 1 mine, the first pick is safe 96% of the time. For 3 mines, 5 safe picks happen 49.57% of the time. For 10 mines, the same 5 picks succeed only 5.65% of the time.
Does Mines demo mode have the same odds as real money?
The rules and the math for a fair board are the same, so the probabilities in our tables apply to both. We can't verify how any specific demo is coded, though. For real-money play, use provably fair verification to check that the mine positions came from the committed server seed.
Is there a Mines strategy that beats the house?
No. Every pick has its probability built into the multiplier, so the house edge applies whether you take one tile or ten, use few mines or many. Patterns, 'safe' corners and predictor apps don't work, because the mine positions are fixed by a hash before you click anything.
How many mines should I pick in Mines?
It's a choice of volatility, not of edge. Few mines give high hit rates and small multipliers; many mines give rare, large multipliers. With 1 mine, ten safe picks happen 60% of the time for 1.65x; with 10 mines, three safe picks happen 19.78% of the time for about 5x.
Sources
- Binomial Coefficient — Wolfram MathWorld. Number of ways to choose k from n
- Hypergeometric Distribution — Wolfram MathWorld. Drawing without replacement formula (keno, mines)
- Provably Fair: Game Events — Stake.com. Limbo uses houseEdge 0.99 (1%), so P(result >= x) = 0.99/x; Plinko path per row; Mines on 5x5 grid
- Provably Fair: Implementation — Stake.com. Server seed hashed before play, client seed, nonce, cursor; HMAC_SHA256 output split into floats
- Doctrine of the maturity of the chances — Encyclopaedia Britannica. Gambler's fallacy: falsely assumes plays are dependent