"Is this game even winnable?" Every solitaire player has asked it, usually after twenty minutes of effort ends in a dead board. The honest answer is that it depends entirely on which game you're playing - And the differences between variants are enormous. Some solitaires are winnable more than 99.99% of the time; others are lost roughly nineteen deals out of twenty no matter how well you play. This guide explains what "winnable" actually means and walks through the odds for the major variants.
What "Winnable" Actually Means
There's a subtle but important distinction hiding in that word, and it explains why you'll see wildly different percentages quoted for the same game.
- Theoretically winnable means a deal can be won by a perfect player with complete information - Someone who can see every face-down card and never makes a mistake. This is what card solvers measure.
- Actual win rate is how often real people playing under normal rules - Hidden cards, no take-backs, human error - Actually win. It's always lower, often dramatically so.
Klondike is the classic illustration: the theoretical winnable rate is very high, but the real-world win rate is under half. The gap is the combined cost of hidden information and imperfect play. Keep the two ideas separate and the numbers below stop looking contradictory.
There's a third figure worth naming, too: the solver-verified rate you get when software plays the deal with full knowledge but under the game's normal move rules. It sits between the two - Higher than any human average, lower than the loose "could this ever be won" ceiling - And it's the number most modern winnable-only modes actually use. When you see three different percentages quoted for the same game, this is usually why: each is measuring a different kind of player.
The Odds, Variant by Variant
FreeCell - The near-certainty
FreeCell is the champion of winnability. Because every card is dealt face up, there's no hidden information to get unlucky with - Only a puzzle to solve. Of the 32,000 classic numbered deals, exactly one (deal #11982) is unwinnable, putting the winnable rate at roughly 99.999%. In practice, that means when you lose at FreeCell, it's almost always a solvable position you misplayed rather than a bad deal. Try it on our FreeCell table and the odds are overwhelmingly in your favor from move one.
Klondike - High in theory, humbling in practice
Klondike, the game most people just call "solitaire," is where the theory and reality split most dramatically. Computer analysis puts the theoretically winnable share of Turn 1 deals at roughly 79–82% - Some studies land a bit higher - But that assumes a perfect player who can see the face-down cards. Real skilled players, working with hidden information and no perfect foresight, win on the order of 43% of Turn 1 games; casual players win considerably fewer. Turn 3, which deals three stock cards at a time and only lets you play the top of each fan, is substantially harder still, with realistic win rates roughly in the 11–26% range.
Spider - It all depends on the suits
Spider's odds swing enormously with the difficulty level. One-suit Spider is a very high win-rate game - Practiced players win the large majority of deals, since suit never restricts a move. Two suits pulls that down well below the one-suit rate, and four-suit Spider is one of the hardest common solitaires, with skilled win rates often in the low double digits. Same game, same layout, wildly different odds - The only variable is how much the in-suit movement rule fights you. See how the levels feel on our Spider table.
Pyramid - A game of the deal
Pyramid, where you clear a triangle of cards by pairing them to sum to thirteen, is heavily luck-dependent. Under standard rules the odds of a win are low - On the rough order of 1 in 20 - Because so much depends on the buried cards falling in a workable order. Skill helps at the margins, but Pyramid is fundamentally a game where a large majority of deals simply can't be won.
Golf, TriPeaks, and the other quick games
Beyond the big four, a whole family of fast solitaires - Golf, TriPeaks, and their relatives - Sit at the luck-heavy end of the spectrum, closer to Pyramid than to FreeCell. In these games you clear cards by playing them onto a single waste in ascending or descending order, and much of the outcome is fixed by how the layout happens to line up with your stock. Skill matters at the margins (choosing which branch to open first, saving a wild card for a jam), but a large share of deals simply won't go, and long win streaks are rare. They're built for quick, breezy rounds rather than deep solving, and their odds reflect that design choice.
Where the classics fall
| Game | Roughly how winnable | What decides it |
|---|---|---|
| FreeCell | ~99.999% (near-certain) | Pure planning; no hidden cards |
| Klondike (Turn 1) | ~79–82% in theory, ~43% in practice | Hidden cards + your decisions |
| Klondike (Turn 3) | Lower; ~11–26% realistic | Restricted stock access |
| Spider (1 suit) | Very high | Suit never blocks a move |
| Spider (4 suit) | Low double digits (skilled) | In-suit movement rule |
| Pyramid | Low (~1 in 20) | Luck of the buried order |
Why the Numbers Vary So Much
Three factors explain almost all the spread between variants:
- How much is hidden. FreeCell hides nothing and is nearly always winnable; Klondike and Spider bury cards face down, injecting the luck that separates theory from practice.
- How free your moves are. One-suit Spider lets any run move; four-suit Spider chains you to in-suit runs. Turn 1 Klondike exposes every stock card; Turn 3 locks two-thirds of them behind others.
- How much the scoring or structure fights you. Games like Pyramid depend so heavily on the exact deal that skill can only nudge the odds.
What About "Winnable-Only" Mode?
Many modern solitaire apps offer a winnable-only or "solvable deals" mode, which uses a solver to deal you only shuffles that are theoretically winnable. It's a genuinely useful feature - It removes the frustration of grinding on an impossible Klondike deal, and it turns every loss into a lesson, because you know for certain a solution existed.
Two caveats are worth understanding. First, "winnable" here means winnable by a perfect player with full information - A solvable Klondike deal can still be extraordinarily hard for a human, and you can absolutely lose it. Second, for FreeCell the mode barely matters, since essentially every deal is winnable already. Winnable-only shines brightest in Klondike, where it lifts the ceiling from that ~80% theoretical rate to a guaranteed 100%.
So Should You Blame the Deal?
Sometimes, honestly, yes - And knowing the odds tells you when. Lose a FreeCell game and the deal is almost never at fault; the solution was there. Lose a four-suit Spider or a Pyramid deal and luck genuinely may have beaten you. Lose a Klondike game and the truth is usually in between: the deal might have been one of the roughly one-in-five that no one could win, or it might have been winnable and you missed the line.
That's the real value of understanding winnability - It tells you when to shrug off a loss and when to replay and learn. If you want deals where a solution is guaranteed, look for a winnable-only setting; if you want to know the loss wasn't your fault, remember the odds for the variant in your hand. Either way, a fresh game is one click away on our solitaire table.