Are All Solitaire Games Winnable? Win Rates by Game

TL;DR: No. Not every solitaire deal is winnable, and the odds swing wildly by game — FreeCell is solvable roughly 99% of the time, Klondike about 82%, and Accordion under 1%.

I keep a sticky note on my monitor tracking my win rates by hand. Which tells you everything you need to know about me. The single question it answers most often is the one you came here for. So let me give it to you straight.

Are all solitaire games winnable?#

That is worth sitting with for a second. When you lose a hand of Accordion, you almost certainly never had a chance. About 99 deals out of 100 have no solution at all. When you lose a hand of FreeCell, that one is on you.

I built a full breakdown of all 26 solitaire games ranked by win rate if you want the complete ladder. This post explains the why behind those numbers.

What "winnable" actually means: solvable vs your win rate#

Here is the distinction that fixes most of the confusion: a deal being "solvable" means a winning line exists. Your win rate is how often you actually find it. These are two different numbers, and people conflate them constantly.

Solvability is a property of the deal itself. A computer can prove, with certainty, whether a path to victory exists from a given starting layout. That is the ~82% figure you see quoted for Klondike. It is the share of deals where some sequence of moves wins.

Your personal win rate is lower, sometimes much lower. Most Klondike players win around 30–40% of hands even though 82% are theoretically winnable. Spotting the right early move is hard. I say that as someone who has spent years defending the honor of patience games — the gap between "possible" and "what I pulled off" is humbling.

There is a second wrinkle. The 82% figure for Klondike is itself an estimate, not a proven fact. Researchers have simulated millions of deals to land near ~79–82%. But the exact solvable rate for Klondike is famously unproven. So treat it as "estimated ~82%" rather than gospel.

So when you read "Klondike is 82% winnable," that is the solvable ceiling, not a promise. And a hedged one at that. The cards can be beatable and you can still lose.

Win rate by solitaire game#

The table below shows the solvable ceiling and approximate win rate for eight popular variants. Read it as "with optimal or near-optimal play." Your real-world numbers will sit lower, especially on the harder games.

Win rates are approximate and assume strong play; Penguin, the most winnable at ~99.9%, and Accordion — the hardest at ~1% — bracket the whole range.

So it is genuinely not just you. If you have been grinding Pyramid, winnable only ~5% of the time, and feeling cursed, the math is on your side. It says you should lose most hands. That is the design, not your skill.

Know someone who rage-quits Klondike? Send them this.

Why are some solitaire deals impossible?#

Some deals are unwinnable because the cards lock each other in a way no legal move can undo. A card you need is buried under cards that can only move if you first move the buried card. It is a deadlock, baked in at the deal.

The trigger varies by game. In Klondike, face-down cards hide information — you commit to exposing one without knowing what is underneath. Sometimes that commitment seals your fate before you see it coming. In Accordion, there are no free cells at all. Every placement is permanent, and one bad merge ends the game.

The cleanest proof that impossible deals exist comes from FreeCell. Of the original 32,000 numbered Microsoft FreeCell deals, the solver community established that all are solvable except exactly one — deal #11982. One deal, provably, has no solution. The rest do.

That single counterexample is why we say FreeCell is approximately 99%+ solvable rather than perfect.

I find that one deal weirdly comforting. It is the cleanest demonstration of the whole idea: a game that is almost perfectly solvable still hides exactly one trap. No skill on earth gets you out of it. If the friendliest game in the table has an impossible deal, of course the harsher ones are riddled with them.

Is Spider Solitaire always winnable?#

No. Spider Solitaire is not always solvable — the four-suit version wins only about 8% of the time even with strong play. It is one of the harder mainstream variants, despite its reputation as a casual pick.

The trouble is that Spider asks you to build four full suit-sequences across ten columns at once, with cards layered over each other. One forced move can bury a card you needed for the rest of the hand.

Easier Spider modes (one-suit, two-suit) lift your odds considerably. But the full four-suit deal is unforgiving — and plenty of those deals have no winning line at all.

If someone tells you every Spider game is winnable "if you just play better," they are wrong about the four-suit game. Skill raises your rate; it does not reach 100%.

Most vs least winnable solitaire games#

Penguin and FreeCell sit at the winnable end; Accordion, Golf, and Pyramid sit at the brutal end. The spread between them is enormous — from near-certain to near-hopeless.

On the kind end, Penguin clears nearly every deal because its starting layout is built around a single base card. FreeCell gives you full information plus four staging cells. These reward planning over luck.

On the cruel end, Golf (~3%) lets you move only one rank up or down from the waste, so the deal usually strands you. Canfield, a brutal ~8%, was literally designed for casinos to win. And Accordion — the hardest at ~1% — gives you nowhere to set a card down. The pile count explodes and almost never converges.

The lesson: choose your variant to match the experience you want. Want to win? Stay near the top of the table. Want to be humbled? You know where to go.

Is it me, or is it the deal?#

Honestly? Often it is the deal — and that is the relief most players need to hear. If your game of choice has a low solvable ceiling, you were fighting the cards before you made a single move.

I have a complicated relationship with this fact. Part of me wants every loss to be a lesson I can fix. But the data is clear: on a ~5% game like Pyramid, roughly 19 of every 20 deals were never yours to win.

Beating yourself up over those is like blaming yourself for the weather.

Here is the useful split. On high-ceiling games — FreeCell, Klondike, Penguin — a loss usually is a missed line, and reviewing it makes you better. On low-ceiling games, a loss is mostly the shuffle.

Knowing which game you are playing tells you whether to study the loss or shrug it off.

This reframed how I track my own numbers. My Pyramid column on that sticky note used to depress me. Then I lined it up against the ~5% solvable ceiling — and realized my results were roughly where the math says they should land, which is fine, which is completely fine. The deck was setting the pace, not my judgment.

So no, it is not just you. It is partly you, partly the deck, and the ratio depends entirely on what you are playing.

Bottom line: when to give up on a deal#

Walk away from a deal once you have exhausted your reversible moves and the remaining cards form a deadlock you cannot break. On low-win-rate games, that moment comes early and often. Quitting is not failure — it is reading the board correctly.

My rule of thumb after years of this is simple. On a sub-10% game, if you have made every safe move and nothing new opens up, the deal is probably dead. Restart and move on.

On a 99% game like FreeCell, assume the win exists and keep hunting. There is almost always a line you have not spotted.

Match your effort to the odds, and you will enjoy this hobby a great deal more. Want to pick your next game by exact difficulty? All 26 solitaire games ranked by win rate lays out the full ladder. It runs from the near-guaranteed wins down to the ones that exist mainly to keep you honest.