Baker's Solitaire: when planning meets a hard limit
Baker's Solitaire uses FreeCell's table. All 52 cards are face up in eight columns, with four free cells and four foundations. Two rules make it much stricter. Cards build down in the same suit only, and on this site only a King may fill an empty column. Those two rules change the question strategy has to answer. In FreeCell it is "how do I win this deal?". In Baker's it is often "can this deal be won at all?".
Our solver settled every one of 10,000 deals on this site's rules. Only 31.8% can be won (95% range 30.9%–32.8%). The other 68.2% are proven impossible from the deal (95% range 67.2%–69.1%). No amount of thinking ahead rescues those. For the winnable third, though, planning is everything, because the solver found a way through every one of them.
We did not run a greedy player on Baker's, so this page has no greedy-versus-planner pair. The closest thing we have is a rule comparison: it shows how much of the game is decided before you play, and how much by what you do.
At a glance: what the rules leave for strategy
| Version | Winnable | Proven impossible | Deals |
|---|---|---|---|
| As played here (Kings only in gaps) | 31.8% (30.9%–32.8%) | 68.2% (67.2%–69.1%) | 10,000 |
| Classic rule: any card in a gap | 76.0% (74.1%–77.8%) | 24.0% (22.2%–25.9%) | 2,000 |
| FreeCell, same table, alternating colors | 99.7% (99.5%–99.8%) | 0.0% | 10,000 |
Every deal was fully settled, so these are true shares of winnable deals, not bot scores. The classic figure agrees with the published 75.05% from Blake and Gent (2026). The Kings-only rule removes more than half of the winnable deals. It is the single biggest factor in this game.
How Baker's is played on this site
- Layout: eight columns, four of seven cards and four of six, all face up.
- Free cells: four, one card each.
- Foundations: four, one per suit, Ace up to King.
- Building: down in the same suit only, such as the 6 of hearts on the 7 of hearts.
- Empty columns: only a King, or a same-suit run headed by a King.
- Group moves: ordered runs can move together, limited by your empty free cells. Empty columns do not raise that limit.
What thinking ahead changes in Baker's
Plan around the Kings first
Only a King can use a cleared column. Before your first move, find all four Kings. A King sitting deep in a column is a liability. A King near the bottom of a short column is your way to reuse space.
Never clear a column you cannot fill
A greedy player empties a column because it can. A planner empties it when a King is ready to move in. An empty column with no King available does nothing for you.
Same-suit links are the only links
Every card placed on a different suit has to leave again through a free cell. Treat each off-suit placement as a debt, and pay it back before you take on another.
Solutions are short and exact
When a deal can be won, the solver's first solution averaged 39.1 moves (95% range 38.8–39.4, over 3,184 solved deals), shorter than FreeCell's 62.2. There is little slack. One wasted cell can close the only line.
Half the time you start by digging
49.7% of our deals (95% range 48.7%–50.6%) had an Ace free to play on move one. In the rest, pick the dig whose cards all have a same-suit home or can go straight up.
Know when to stop
With 68.2% of deals impossible, some walls cannot be climbed. If a Restart with a different plan hits the same wall, deal a new game. That is not giving up; the numbers say it may be the right call.
Features on spidersolitaireonline.us
Baker's Solitaire is free in any modern browser on computer, tablet or phone, with nothing to install and no account to create. Large cards and a zoom control, Hint, unlimited Undo, Restart, Auto Play, and drag-and-drop or tap are all included. The game keeps count of wins, time and moves, and you can switch sound off, go fullscreen or change the table color.
How we ran the study
The spidersolitaireonline.us team ran this research in September 2026 on this site's own Baker's rules, using simulations written in Python. An AI assistant, Claude by Anthropic, helped write and review the code and draft the page. Every figure comes from running the code.
- Rules read from the game code: FreeCell's layout with same-suit building, Kings only in empty columns, and group moves limited by free cells alone.
- Two versions in one program: the site rule and, for comparison, the classic any-card-in-a-gap rule.
- Numbered deals: 1 to 10,000 on the site rule and 1 to 2,000 on the classic rule, all from fixed seeds.
- An exhaustive solver. All cards are visible, so the solver sees exactly what you see. It finished its search on every deal. Each deal is therefore either won or proven impossible, with none undecided.
- Published check. Blake and Gent (2026) report 75.05% for classic Baker's Game; ours is 76.0%. No published figure exists for the Kings-only rule.
- Cross-check. A separate brute-force program replayed deals 1 to 5 and agreed on every verdict. The site allows a card back from a foundation, which the solver never uses, so if anything the true share is slightly higher.
- Margins of error are 95% Wilson score intervals.
Baker's questions
Why do I lose so often?
Because most deals cannot be won on these rules. Only 31.8% have a solution, even for a program that sees every card and searches every line.
Would better strategy win more?
On winnable deals, yes: the solver won all of them. On the other 68.2%, no strategy can help.
Why is the classic game so much easier?
Letting any card into an empty column turns gaps into general workspace. On our rule a gap only takes a King, and the winnable share drops from 76.0% to 31.8%.
What should I play instead?
Eight Off has the same suit rule with eight cells, and almost every deal is winnable. FreeCell is the forgiving cousin. For hidden-card planning, try Spider. All games are on the home page.
By Admin, spidersolitaireonline.us (Solitairen OOD). Research run September 2026.