OK Win Mines Odds: Chance of Each Safe Pick
Each safe pick on a Mines-style grid is less likely than the one before. This tool gives the chance of every pick, the multiplier it needs and what the game keeps. It opens on 25 tiles, 3 mines and a cash-out after 2 safe picks.
At the time of writing (8 October 2026), OK Win’s category row had a Mini games label that we did not open, and we read no tile called Mines. That proves nothing either way. This tool prices a grid game wherever you meet one. Many searches for Mines are for predictors, and the last section deals with them.
Start from the address
- A page titled OK.Win, and a grid of tiles with a mine count on it. Copy the grid size, the number of mines and the multiplier shown after each safe pick. The button’s address forwards to one of several domains, so the page can sit at a different address each time. Read the address bar as well.
- A “Mines predictor” site, bot or file. It claims to know where the mines are, which is a claim and not the game. Score its address in the link checker, then read the Mines predictor page.
- A grid in a video or a chat. It is someone else’s display. Price it here only as a sum on its numbers.
The grid in numbers
The first pick on 25 tiles with 3 mines is safe 22 times in 25. The second is safe 21 times in 24. Multiply the steps and you get the chance of getting that far. The edge is the tool’s example 3%, not a figure we read.
| Safe picks | Chance of getting that far | Break-even multiplier | Pays with a 3% edge | Back per ₹100 |
|---|---|---|---|---|
| 1 | 88.0% | 1.14× | 1.1× | ₹97 |
| 2 | 77.0% | 1.3× | 1.26× | ₹97 |
| 3 | 67.0% | 1.49× | 1.45× | ₹97 |
| 4 | 57.8% | 1.73× | 1.68× | ₹97 |
| 5 | 49.6% | 2.02× | 1.96× | ₹97 |
| 6 | 42.1% | 2.37× | 2.3× | ₹97 |
| 8 | 29.6% | 3.38× | 3.28× | ₹97 |
| 10 | 19.8% | 5.05× | 4.9× | ₹97 |
The last column does not move. If a game pays as the table assumes, the point where you cash out changes how often you win, and what the game keeps stays the same.
The opening round
The calculator opens on a cash-out after 2 safe picks, ₹40 a round, 100 rounds. A round gets through 77.0% of the time and pays 1.26×, so a win nets about ₹10 and a mine costs ₹40. You stake ₹4,000 and the game keeps about ₹120. A session ends ahead 28 times in 100, and the longest run of mine rounds to expect is 3. By pick 5, a mine is more likely than not to have ended the round.
More mines, same cost
Keep the cash-out at 2 picks and change only the mine count. Each row costs about ₹120 over 100 rounds at ₹40.
| Mines | First pick hits a mine | Round gets through 2 picks | Pays | Ends ahead | Longest losing run |
|---|---|---|---|---|---|
| 1 | 4% | 92.0% | 1.05× | 18 in 100 | 1 |
| 3 | 12% | 77.0% | 1.26× | 28 in 100 | 3 |
| 5 | 20% | 63.3% | 1.53× | 33 in 100 | 4 |
| 10 | 40% | 35.0% | 2.77× | 37 in 100 | 9 |
| 15 | 60% | 15.0% | 6.47× | 43 in 100 | 18 |
The chance of finishing ahead rises with the mines while the cost holds still. It is a sign of a bumpier ride, not a lower price.
Check the multiplier on your screen
Games do not always state what they keep. Type the multiplier your screen shows after a number of safe picks in the optional box, and the tool works out the edge it implies. On this grid, a shown 1.3× after 2 picks implies about 0%, while 1.1× implies 15.3%. A negative figure would mean the screen pays more than break-even, so recheck what you copied.
Why a predictor cannot help
If the layout is drawn at random, any two tiles you open are safe 77.0% of the time on this grid, whichever two they are. A predictor that names “safe” tiles gets there as often as a coin would, so its record needs testing before it is believed. The prediction simulator and the hack guide show how. The games guide covers the other tiles on the lobby.
What it cannot tell you
We are not OK Win. We cannot see inside it or read how any grid game places its mines or sets its multipliers. The practice grid uses your browser’s random numbers and pretend money. See the tools index for the rest.
Checking a grid game on the platform’s own page? Read the address bar first and copy the mine count and multipliers before you stake.
Opens the platform’s own page in a new tab. Referral link. 18+ only.
Common questions
How many mines make the very first tile a risk?
The first pick hits a mine with chance mines ÷ tiles. On 25 tiles that is 4% with 1 mine, 12% with 3, 20% with 5, 40% with 10 and 60% with 15. The risk climbs with each tile you open, because the mines stay the same while the unopened tiles shrink.
Does cashing out later cost less than cashing out early?
On the tool’s assumption, no. Wherever you stop, about ₹97 of every ₹100 staked comes back on a 3% edge. Stopping later changes how often a round pays: 88.0% of rounds get through 1 pick, and 19.8% get through 10. A game with different rules would need its own numbers.
After how many safe picks is a mine more likely than not?
It depends on the mine count. On 25 tiles it is pick 13 with 1 mine, pick 5 with 3 mines, pick 3 with 5 mines and pick 2 with 10 mines. The tool prints the figure for your grid. It is the point where the chance of still being safe drops under one half.
My screen shows 1.2× after two safe picks. What does that mean for the game’s share?
On 25 tiles with 3 mines, two safe picks happen 77% of the time, so the break-even multiplier is 1.3×. A screen that shows 1.2× keeps about 7.6% of each stake on average, and 1.1× would keep 15.3%. Type your multiplier in the optional box and the tool works out the share for your own grid.
A predictor says it clears five tiles. How often would a blind guess do that?
On 25 tiles with 3 mines, five safe picks in a row happen 49.6% of the time with no knowledge at all. A record of a few clean rounds therefore shows little. The hack guide shows how to test a claimed record against luck.
Does the pattern of tiles I tap change my chance?
Not if the layout is drawn at random, which is the tool’s assumption. Every set of tiles is then equally likely to be clear. The tool has a test that tries four tapping habits on 2,000 random layouts so you can watch them come out alike. We could not read how any real game places its mines.