Spider Solitaire, two suits: the deals are still there, the easy moves are not

Two-suit Spider uses the same 104 cards and the same ten columns as the one-suit game, but half the cards are hearts and half are spades. That one change splits the game in two. The layout still looks friendly. The moves that made one-suit Spider flow no longer work nearly as often.

Our research asks how much of a game is winnable at best. For Spider, we did not build a perfect-play solver, so we measure the other end, the floor, and compare it with the best published ceiling. For two suits the gap between them is striking. Our simple computer player won just 5.7% of 4,000 deals. With every card visible, published research finds about 98.5% of even four-suit deals winnable. Two-suit deals are not lost causes; they just demand much better play.

You can play two-suit Spider here for free, in any modern browser, with nothing to install.

Two suits against one

The published figure is from Blake & Gent's 2026 study in the Journal of Artificial Intelligence Research. No published figure exists for two-suit Spider on these rules.

Why one extra suit costs so much

In Spider, any card may be placed on a card one rank higher, whatever the suit. But a group of cards can only be moved together if it is all one suit. With one suit, every ordered run can move as a block. With two, a mixed run such as 9 of spades, 8 of hearts, 7 of spades is stuck: you have to take it apart one card at a time.

That is why our player's results fall off a cliff. It still makes plenty of legal moves, but it creates mixed runs it cannot later move, and it never goes back to try a cleaner order. Completed suits dropped from almost five per game to under one. The deals themselves have not become impossible; the room for error has shrunk.

Rules for two-suit Spider on this site

• 104 cards (hearts and spades) in ten columns: four of six and six of five, top card face up.
• 50 cards in the stock. Each deal puts one card on every column.
• A card may sit on any card one rank above it, whatever its suit.
• Only same-suit runs move as a group.
• Any card may go into an empty column.
• A finished King-to-Ace run of one suit is removed automatically.
• If a column is empty when you click the stock, the game shows a warning.

Playing for the ceiling, not the floor

Prefer same-suit moves

When you can place a card on either suit, choose the one that matches. A same-suit link can move later; a mixed one is a knot you must untie.

Be slow to mix

Mixed runs are sometimes necessary to reach a face-down card. Make them when they uncover something, not just because they are legal.

Value an empty column enormously

Our player cleared a column before the first deal in only 0.5% of games, but won 15 of the 21 games in which it did. An empty column lets you untangle mixed runs.

Aim for your first complete suit

Each finished suit leaves the table and frees space. Our player averaged under one per game; getting even one early changes the rest of the game.

Undo mixed moves that led nowhere

Undo is what separates you from our player. If a mixed run blocked a column, step back and try the other order.

What we can and can't say about the two-suit ceiling

For most games on this site we can tell you how many deals are winnable. For two-suit Spider we can only frame it. Here is what is solid:

• The published ceiling for four suits, with every card visible, is 98.487%, with a margin of ±1.513%.
• Our no-peeking player's results rise steeply as suits are removed: 0.2% at four suits, 5.7% at two, 52.8% at one.
• solitairex.io reports real-player win rates of 70–84% for one-suit Spider and 11% or less for four suits.

What we cannot give you is a measured two-suit ceiling. It would take a proper Spider solver, which is a much bigger project than the solvers we built for other games. Until someone publishes one, the best reading is that the deals themselves are very rarely the obstacle.

Inside the test, one step at a time

The solitaire-online.us team ran this study in September 2026 to measure the floor for two-suit Spider on this site and compare it with the published ceiling.

1. Reading the rules from our game code. Layout, stock, same-suit group moves, empty-column rule and automatic removal of finished runs all come from the site's game files. We assumed that you cannot deal onto an empty column; the site shows a warning, but we did not test it in a browser.
2. Rebuilding the game. The same Spider program as for one suit, with hearts and spades.
3. Dealing the cards. Seeds 100,001 to 104,000, kept apart from seeds 1 to 30 that were used to tune the player.
4. Choosing the player. A look-ahead player searching 250 positions a turn over visible cards only, never undoing. That gives a floor. We built no Spider solver, so we report no ceiling of our own.
5. Running it. 4,000 deals, each played until won or until no useful move remained.
6. Checking against published research. There is no published two-suit figure for these rules. The four-suit value of 98.487% from Blake & Gent (2026) is the nearest ceiling.
7. Working out the margin of error. The 95% range means the true value would sit inside it in 19 repeats out of 20. For our player's win rate: 5.0–6.5%.
8. What the numbers cannot tell you. Our player is weak at two suits. Its 5.7% is not how winnable the deals are, and not what a careful person with Undo achieves.

The code behind the ceiling

We wrote the study in Python 3.11, standard library only. The Spider game and player are in spider.py; runner.py runs the deals in parallel on 19 worker processes on a 20-core machine. Each deal is regenerated from its seed, so the same shuffle comes out every time.

Each turn, the player runs a best-first search of up to 250 positions over the visible cards and scores them: same-suit links earn far more than mixed links, empty columns and newly turned cards earn a bonus, and face-down cards cost points. It plays toward the best position it found and deals only when nothing improves. The two-suit run of 4,000 deals took about 14 minutes. Ranges use the Wilson score interval.

Our code and this page were written with help from an AI assistant, Claude by Anthropic, which we used to draft and to review. The figures are not the AI's: each one comes from running the code or from the published research we link to.

Because each shuffle comes from a numbered seed, anyone with the same code gets exactly the same deals and the same verdicts.

By Admin, solitaire-online.us. Study date: September 18, 2026.

Ceiling questions from players

Is two-suit Spider winnable?

Most deals very likely are. Published research finds about 98.5% of four-suit deals winnable with every card visible. No one has published a two-suit figure for these rules.

What makes the second suit so costly?

Because only same-suit runs can be moved together. Mixed runs have to be broken up card by card.

How many moves did the two-suit wins take?

Our player's wins averaged about 155 moves, including deals from the stock, compared with about 118 at one suit.

Should I switch back to one suit?

If two suits feels frustrating, one-suit Spider is the best place to practice. When you're ready for the hardest version, try four suits.

Is there more research on the site?

There are more findings from our study on the blog, and every game is on the games page.