Yukon Solitaire: Spider's close cousin on the ladder
Yukon looks like Klondike at first glance: seven columns, alternating colors, foundations from Ace to King. But there is no stock at all. Every card is dealt onto the table, and you may pick up any face-up card along with everything stacked on top of it, whether or not those cards are in order. That freedom to move messy piles is what makes Yukon feel like a relative of Spider.
It also lands right next to two-suit Spider on our difficulty ladder. A simple program that sees only the face-up cards and never undoes won 4.5% of 10,000 Yukon deals, just under two-suit Spider's 5.7%. Yet a solver that could see every card proved at least 67.0% of deals winnable. Yukon is a game where the deals are generous and the planning is hard.
This page covers where Yukon ranks, how the rules work here, what our results suggest about playing it, and every step of the research.
Where Yukon ranks
| Game | Simple program won | Deals | Winnable with every card known |
|---|---|---|---|
| Spider, 1 suit | 52.8% | 4,000 | not measured |
| Spider, 2 suits | 5.7% | 4,000 | not measured |
| Yukon (this page) | 4.5% (4.1–4.9%) | 10,000 | 67.0–96.6% (500 deals) |
| Alaska | 1.5% | 10,000 | 75.6–92.8% (500 deals) |
| Spider, 4 suits | 0.2% | 2,000 | 98.5% (published) |
Our Yukon program is weak, and we say so plainly. It looks a short way ahead and never undoes, and in Yukon that leaves a lot on the table. Its 4.5% is a floor, not a verdict on how often people win. The right-hand column is the better guide to what the deals allow.
Reading the solver range
Our solver was given a fixed search budget for each of 500 deals. It found a win in 67.0% of them and proved 3.4% impossible. The other 29.6% were still open when it ran out of budget. So somewhere between 67.0% and 96.6% of deals can be won with perfect knowledge. The main published study we use, from Blake and Gent, does not cover Yukon, so we have no outside figure to compare against.
Yukon rules on spidersolitaire.us
- Layout: seven columns. The first has one card; the others hold six to eleven cards, with the last five face up.
- No stock: all 52 cards start on the table.
- Foundations: four, built up by suit from Ace to King.
- Building: a card goes on one rank higher in the opposite color.
- The big move: you can move any face-up card together with every card on top of it, even if those cards are not in sequence. Only the card you pick up has to fit.
- Empty columns: only a King, together with whatever sits on it.
What our results suggest
Don't worry about hidden Aces
On average 1.6 of the 4 Aces started face down. Our program won 4.4% of deals with two or more hidden Aces and 4.5% with one or none. No meaningful difference. In Yukon, the layout of the face-up cards matters more than where the Aces start.
Use the big move to uncover, not to rearrange
Moving a whole pile is powerful, but its best use is to expose a face-down card. Every card you turn gives you new options, and there is no stock to fall back on.
Expect a game of about a hundred moves
The games our program won took about 97 moves on average. A winning line is often long and indirect.
Plan the King columns
Only a King can take an empty column, and it brings its whole pile with it. Choose which King goes first based on what lies beneath it.
Use Undo freely
Our program's weakness is that it never takes a move back. Across all games it averaged just 7.4 cards on the foundations. With unlimited Undo, you can test a big move and reverse it, which is exactly what the solver's 67% or more depends on.
Features on spidersolitaire.us
Yukon is free on this site in any modern browser on computer, tablet or phone. There is nothing to download and no account, sign-up or email address. You get large clear cards with a zoom control, Hint, unlimited Undo, Restart, Auto Play, and drag-and-drop or tap to move. Wins, time and moves are tracked for the current game. Sound can be switched off, and fullscreen and table colors are available.
Behind the numbers: how we ran the study
The spidersolitaire.us team studied Yukon in September 2026 to find out where it sits relative to Spider, using the rules our own Yukon follows.
- Started from the code that runs our game. Seven columns with the last five cards face up, no stock, opposite-color building, any face-up card moves with everything above it, and Kings only in empty columns.
- Rebuilt the game as a program. Yukon and Alaska share one program, since they differ only in how cards build.
- Dealt numbered hands. Deals 1 to 10,000 for the simple player and 1 to 500 for the solver, each shuffled from a fixed seed so the test repeats exactly.
- Chose two players. A simple program that sees only face-up cards and never undoes, as a floor. And a solver that sees every card, as a ceiling nobody can reach at the table.
- Ran the deals. The simple player finished all 10,000. The solver had 50,000 positions per deal before a deal was marked undecided.
- Looked for published figures. Blake and Gent (2026) do not report Yukon, so our figures stand on their own.
- Calculated the margin of error. For the simple player's 4.5%, the 95% range is 4.1% to 4.9%. For the solver's 67.0%, it is 62.8% to 71.0%.
- Stated the limits. Our simple player is weak at Yukon. The solver sees hidden cards. And nearly 30% of solver deals stayed undecided, so we quote a range rather than one number.
The code behind the ladder
All the code is Python 3.11 with only the standard library; Yukon runs in yukon.py. The simple player searches up to 300 positions ahead on each turn using the visible cards and scores them: a big reward for every card on a foundation, a penalty for every hidden card, and small points for correct and incorrect links. The solver is a depth-first search that tries every legal move within its budget and plays cards to the foundations automatically only when that is safe under alternating-color rules. The 10,000 player games took about two minutes and the 500 solver deals about eight minutes, spread across 19 worker processes on a 20-core machine.
Ranges for percentages are Wilson score intervals. For writing and reviewing the code, and for drafting this page, we worked with Claude, an AI assistant built by Anthropic. It was not a source of any number: all figures come from running the simulation, and the team reviewed them before publishing.
Ladder questions, answered with data
Why can I move cards that aren't in order?
That is Yukon's defining rule. Only the card you pick up must fit; everything on top of it comes along.
Does Yukon beat Klondike for difficulty?
For our simple programs, much harder: 4.5% against 42.3% for Klondike Turn 1. For a patient player who plans, the gap is smaller, since most Yukon deals can be won.
What's the difference between Yukon and Alaska?
Alaska keeps Yukon's deal and pile moves but has you build within a suit, in either direction. Within the same search budget, our solver proved more Alaska deals winnable (75.6% against 67.0%), though the ranges overlap once undecided deals are counted.
Can every Yukon deal be won?
No. Our solver proved 3.4% impossible, even seeing every card.
What should I try after Yukon?
If you liked moving piles, try two-suit Spider. All the other games sit on our games page.
By Admin, spidersolitaire.us. Research run September 2026.