Baker's Solitaire: the game where losing is usually not your fault

Baker's Solitaire looks like FreeCell's twin: all 52 cards face up, four free cells, four foundations. Then you notice two rules that change everything. Cards in the columns must follow suit, and on this site only a King can move into an empty column.

Those two rules make Baker's one of the toughest games here. We asked a solver to play 10,000 deals with perfect foresight and to keep going until it had either won or proven that no win exists. It got a definite answer on every single deal. Only 31.8% can be won. The other 68.2% are impossible from the moment they're dealt.

That is the most important thing a new Baker's player can know. If you lose two games out of three, you're not doing anything wrong — that is simply the shape of this game.

Baker's at a glance

The rules here, and why they bite

• Eight columns, all face up: four of seven cards and four of six.
• Four free cells that take a single card apiece.
• Four foundations, filled by suit from Ace to King.
• Build down in the same suit in the columns: the 8 of clubs on the 9 of clubs, and nothing else.
• Only a King can go into an empty column on this site.
• Moving runs uses your empty free cells; empty columns don't add to how many cards you can move at once.

Following suit means each card has exactly one place it can go in the columns, instead of two in FreeCell. And the Kings-only rule means an empty column is far less flexible than in the classic game. Our comparison run shows the cost: switching the rule so any card can fill an empty column lifts the winnable share from 31.8% to 76.0%. One rule more than halves your chances.

Casual play vs. careful play in Baker's

Because every deal in our test was settled for certain, Baker's is a rare case where we know the exact ceiling. Even perfect play wins about three deals in ten. A casual player won't hit that, so a realistic aim is to recognize the winnable deals and not waste energy on the rest.

Unlike most games on this site, careful thought here is mainly about diagnosis. Within the first minute you can often tell whether a deal is in trouble.

How to get more out of each deal

Locate every King before moving

Kings are the only cards that can use an empty column, so where they are decides a lot. A King buried deep in a long column, with low cards of the same suit trapped under it in other columns, is a warning sign.

Check for "locked" suits

Because you must build in suit, a card sitting on top of a lower card of its own suit is a problem: the lower card can't come out until the higher one has somewhere to go. Spot these knots early.

Only empty a column when a King is ready

An empty column that no King can reach is almost wasted space here, since nothing else can go in it. Clear columns when you have a King, and ideally its run, waiting to move in.

Keep two cells free for as long as you can

With empty columns unable to help with multi-card moves, your cells do all the work of moving runs. Two empty cells let you move three cards at once; one lets you move two.

Use Undo to test a plan, then commit

Everything is visible, so a line of play can be tested fully. Try it; if it jams, undo back and choose again.

Common ways Baker's games are thrown away

• Filling free cells with cards that have no route out.
• Clearing a column with no King to put in it.
• Playing a card to a foundation that a lower card of the same suit still needs to sit on.
• Playing on too long in a deal that is clearly jammed — two in three deals can't be won.

Playing Baker's Solitaire here

Free, in any modern browser, on any device. You won't need to download anything or register. Large, clear cards with a zoom control; Hint, unlimited Undo, Restart and Auto Play; drag and drop or tap-to-move; a tally of wins, time and moves for the current game; sound on or off, fullscreen and different table colors. Nothing to buy, and no gambling.

If Baker's feels punishing, FreeCell uses the same layout with friendlier rules and almost every deal winnable. Alaska Solitaire is another follow-suit game, but everything moves in groups. Find everything else on the games page.

The research behind these numbers, step by step

These tests were run in September 2026 by the play-solitaire-free-online.us team at Solitairen OOD, using the Baker's rules implemented on this site.

1. Rules from our game code. The site's Baker's Solitaire builds down in suit, allows only Kings into empty columns, and doesn't count empty columns toward multi-card moves. The classic Baker's Game in the literature lets any card fill an empty column; we tested that separately, only for comparison.
2. Game rebuilt in code. We programmed Baker's alongside FreeCell, since the two share most of their rules.
3. Numbered deals. Seeds 1 to 10,000 for the site rules and 1 to 2,000 for the classic rule. They're our own numbers.
4. The player. A solver that knows every card and searches for a win, always trying the most promising position first. Here the searches ran to completion, so every deal ended as either won or proven impossible, with none left undecided.
5. Running it. All 10,000 site-rule deals, plus 2,000 classic-rule deals.
6. Published comparison. For the classic rule, Blake & Gent (2026) report 75.053%; we got 76.0%, with a range of 74.1% to 77.8% that includes their figure. That match is how we know the solver works. We didn't find a published figure for the site's harder Kings-only version to compare with.
7. Margin of error. The "95% range" shows how far off our sample could be. For the site rules, the true winnable share is between 30.9% and 32.8%.
8. Cross-check and limits. A second, completely separate brute-force program re-checked deals 1 to 5 and agreed. The solver doesn't use the move that brings a card back off a foundation, which the site allows; that can only make our figure slightly low.

The code behind the test

The simulations are plain Python 3.11 with the standard library. Baker's shares freecell.py with FreeCell; the rules are switched by a setting. The solver is a best-first search — it ranks positions by cards still to go home and cards blocking lower cards — with a table of seen positions and only provably safe automatic moves. For Baker's, the search space is small enough that it finished on every deal. runner.py spread the work over 19 processes on a 20-core machine: about 16 minutes for the 10,000 site-rule deals, about 3 for the 2,000 classic ones. Error bars come from the Wilson score method.

Anthropic's Claude, an AI assistant, took part in writing and reviewing this code and in drafting the page. The figures came from running the solver, not from the AI, and were checked against the published study.

By Admin, play-solitaire-free-online.us (Solitairen OOD). Research dated September 18, 2026.

Questions new players ask

Why is Baker's so much harder than FreeCell?

Building in suit gives each card only one possible home, and on this site only Kings can fill empty columns. Our numbers: about 99.7% of FreeCell deals were solved, against 31.8% of Baker's deals.

So most Baker's games can't be won?

Right. 68.2% of deals in our test were proven impossible even with perfect play.

How can I tell early whether a deal is winnable?

Look for Kings sitting on top of low cards of the same suit, and for suits tangled into knots in one column. Several of those together usually spell trouble.

What's the difference between Baker's Solitaire and Baker's Game?

They're the same family. The classic Baker's Game lets any card into an empty column; our version allows only Kings, which is why it's harder.

Is it worth playing if I'll lose most games?

Many players love it for exactly that reason: winning one feels earned. Just treat each deal as a puzzle to diagnose, not a test to pass.