Yukon Solitaire: no stock, no luck of the draw, just the table

By Admin, free-solitaire.nz · Figures measured 18 September 2026

Yukon looks like Klondike at first glance, with seven columns and a row of foundations. Then you notice there is no stock pile. All 52 cards are dealt onto the table at the start, and every card you need is already somewhere in front of you. Yukon also has one rule that changes everything: you can move any face-up card, together with everything sitting on top of it, even if that pile is a jumble.

That makes Yukon a game of digging rather than drawing. We ran our own simulation study to see how winnable it really is, and the answer is more encouraging than its reputation suggests. A solver that can see every card won at least 67% of 500 deals and proved only 3.4% impossible. The rest were left undecided within its search limit.

Deal a game, it is free with nothing to download or sign up for, and read on for the rules, the numbers and how to use that big-move rule well.

How Yukon works

The deal

• Seven columns. The first holds a single face-up card. The other six hold 6, 7, 8, 9, 10 and 11 cards.
• The bottom five cards of each of those six columns are face up. The rest are face down.
• No stock and no waste pile. Every card is on the table from the start.
• Four foundations, one per suit, built up from Ace to King.

How cards move

• You can pick up any face-up card along with all the cards on top of it. They do not need to be in order.
• The card you pick up must land on a card one rank higher and the opposite colour: a black 9 on a red 10, for example.
• Only a King (with anything on it) can go into an empty column.
• When a face-down card is uncovered, it turns face up.
• Get all 52 cards onto the foundations to win.

What our simulation found

We measured Yukon two ways, because each tells you something different.

Between two-thirds and nearly all deals can be won

Our solver knew where every card was and searched up to 50,000 positions per deal. It won 67.0% of deals outright and proved 3.4% impossible. The other 29.6% it could not settle within that limit, and some of those will be winnable. So the true share of winnable Yukon deals lies somewhere between about 67% and 97%. What we can say for certain is that impossible deals are rare. When you lose at Yukon, the deal is very probably not to blame.

Winnable is not the same as easy

Our second player is a bot that sees only face-up cards and never takes a move back. It won just 4.5% of 10,000 games and got an average of 7.4 cards home. That is a big gap from the solver, and most of it is the bot's weakness: Yukon's free moves create so many choices that a short look-ahead cannot tell which ones matter. We treat the 4.5% as a measure of how hard Yukon is without planning or Undo, not as a typical human result.

Buried Aces did not matter as much as you would think

On average, 1.63 of the four Aces start face down. You might expect those deals to be much harder, but our bot won 4.4% of games with two or more Aces buried and 4.5% with none or one. The difference is within the margin of error. In Yukon you can lift a buried card out in one move, pile and all, so a hidden Ace is less of a trap than it is in Klondike.

Playing on this site

Everything is free, with nothing to install and no account or email. It runs in any modern browser on a computer, tablet or phone.

• Unlimited Undo and Restart, ideal for a game where so many moves are possible.
• Hint when you cannot see a way forward.
• Large, clear cards and a zoom control, handy when columns grow long.
• Auto Play to finish a won game, and drag and drop or tap to move whole piles.
• Wins, time and moves tracked for the current game, fullscreen, table colours, and sound you can switch off.

Strategy for Yukon

1. Use the big move to dig, not to tidy

Being able to move any face-up card with its pile is powerful, but it is easy to waste on moves that shuffle cards around without uncovering anything. Prefer moves that turn over a face-down card.

2. Work on the long columns

The right-hand columns hold up to six face-down cards. Those are where the trouble hides. Clearing them early gives you information and, eventually, an empty column for a King.

3. Get Kings out of the way early

Only a King can fill an empty column, and a King buried near the top of a long column blocks everything beneath it. When a column empties, think about which King most needs to move, not just which one is easiest to reach.

4. Send low cards home early

In Yukon, Aces, 2s and 3s rarely help you build. Moving them to the foundations as soon as they are free usually costs nothing and clears clutter.

5. Plan around the colour rule

Because piles move as a lump, the card at the bottom of the pile you are moving decides where it can go. Before you commit, check whether that card has an opposite-colour home one rank higher. If it has two, pick the one that uncovers more.

6. Explore with Undo

Yukon usually offers a lot of legal moves at once, and that is a big part of why our no-Undo bot struggled. Try a promising line, see what it uncovers, and back out if it leads nowhere. It is how you close the gap between 4.5% and the two-thirds of deals the solver proved winnable.

How we did the research, step by step

The free-solitaire.nz team ran this study in September 2026. These are the steps we took, in order.

1. Read the rules from our own game. From this site's game code: seven columns of 1, 6, 7, 8, 9, 10 and 11 cards with the last five face up in each, no stock, any face-up card moving together with everything on top of it, building down in alternating colours, and only Kings into empty columns.
2. Rebuilt the game in code. We wrote a program that deals and plays Yukon by exactly those rules.
3. Generated numbered deals. Every deal comes from a fixed seed, so anyone running our code gets the same cards in the same order.
4. Ran a solver that can see every card. Researchers call this a “thoughtful” solver. It is a depth-first search allowed up to 50,000 positions per deal, and we ran it on 500 deals.
5. Sorted every deal into three groups. Solved: 67.0% (95% range 62.8–71.0%). Proven impossible, meaning the search checked every possibility: 3.4%. Undecided, meaning the budget ran out: 29.6%. So between 67.0% and 96.6% of deals can be won with perfect knowledge.
6. Ran a bot that only sees what you see. It looks up to 300 positions ahead on the face-up cards before each move and never undoes. It played 10,000 deals.
7. Counted what the bot did. It won 4.5% (95% range 4.1–4.9%) and got 7.4 cards home per game on average. Whether two or more Aces started face down made no real difference (4.4% against 4.5%).
8. Looked for outside figures. The 2026 study by Blake and Gent in the Journal of Artificial Intelligence Research covers 45 patience games but not Yukon, which is why we measured it ourselves.

What this means: the solver shows the most any player could win, and the bot shows the least a careful player should manage. The bot is weak, and a patient player using Undo will do far better. Ranges are 95% confidence intervals.

The technical side

Yukon is simulated in Python 3.11 with nothing outside the standard library, in yukon.py, which also covers Alaska. A shared runner.py spreads the work across 19 worker processes with Python's multiprocessing, and every deal is shuffled from a numbered seed.

• The bot looks up to 300 positions ahead each turn using only the cards a player can see. It keeps a set of positions it has already explored, and scores positions by cards home, hidden cards and correct links. The 10,000 deals took about 2 minutes.
• The solver sees every card and searches depth-first, remembering every visited position so none is searched twice. It uses Yukon's safe automatic foundation moves and stops after 50,000 positions. The 500 deals took about 8 minutes.
• The 95% range is a Wilson score interval.

The solver's only shortcut is a safe automatic move, so when it runs out of options that is a genuine proof the deal cannot be won, not a guess. An AI assistant (Claude, from Anthropic) helped us write and review the code and draft this page. The AI did not produce the figures: they come from running the code, and we checked them against published research.

Frequently asked questions

What share of Yukon games can be won?

In our test, a solver that could see every card won 67% of deals outright and proved only 3.4% impossible, with the rest undecided. So somewhere between about two-thirds and 97% of deals can be won.

Is Yukon harder than Klondike?

It feels harder because there are so many possible moves. But there is no stock to run dry and no luck of the draw, so a careful player can often dig out of positions that would be lost in Klondike.

Why can I move cards that are not in order?

That is Yukon's signature rule. Any face-up card can be moved along with everything on top of it, in any order. Only the card you pick up has to fit where it lands.

Can any card go into an empty column?

No, only a King, together with whatever sits on it.

What is Alaska Solitaire?

Alaska uses the same deal and big moves, but you build in the same suit, up or down by one rank. It is a good next step once Yukon feels familiar.

Do I need to sign up or download anything?

No. Open the page and the cards are dealt, free.

What else is worth trying?

For another game where nearly everything is in view, try FreeCell or Baker's Solitaire. If you would like a longer challenge, Spider is a good choice. Browse everything on the games page and read more on the blog. A good game for a quiet evening anywhere from Kaitaia to Bluff.