Klondike Turn 3 for Spider players: a stock-steering playbook

Spider players are used to a stock that drops cards where it likes. Klondike Turn 3 is the opposite: the seven columns are ordinary Klondike, but the draw pile hands you cards three at a time and lets you use only the uppermost of each trio. The cards underneath can stay out of reach for pass after pass, and the only thing that changes which card surfaces is what you take from the waste.

You will not find win rates here — those live on the Solitaire Turn 3 page, next to the rest of our Spider-based difficulty ladder. This page is the practical side: the mechanics of the three-card draw, how to reshape the stock on purpose, when a waste card is worth taking, and one worked position from start to finish.

How the three-card turn works

After the deal, 24 cards sit in the stock. A click flips three of them onto the waste in a little fan; you can read all three, but the one on top is the only card in play. Run out of stock and the waste flips back to become the stock again, as many times as you want on spidersolitaire.us.

What matters is reach. Take nothing from the waste and a full pass splits the 24 cards into eight trios, and only positions 3, 6, 9, 12, 15, 18, 21 and 24 are ever on top. The sixteen cards in between stay buried. A second untouched pass shows those same eight cards in the same order; the stock only changes shape when you remove a card from the waste.

Steering the stock

Every card you take from the waste shifts the grouping of every card after it by one place. That is the whole art of Turn 3: using waste plays deliberately to bring different cards to the top.

Playing one card moves everything behind it

Suppose you play the top card of the second group. The rest of the current pass carries on as before, because the stock itself has not changed. But when the waste turns over for the next pass, it is one card shorter, and every card that came after the one you played now sits one position earlier. The groups of three fall differently, so different cards land on top. A single well-timed waste play can bring up a card that would otherwise never have been reachable.

Count the stock when you are stuck

If a pass shows the same cards as the last one, nothing you are doing is changing the stock. Look for any card at all you can take from the waste, even a modest one onto the tableau, just to shift the groups. Undo lets you try this without risk: take a card, go through a pass, and undo if it did not help.

Near the end, the stock gets friendlier

As the stock shrinks, fewer cards hide in each pass. When the stock and waste together hold only a few cards, almost every card can be reached, so a game that looks stuck early can open up late.

Waste or tableau: making the call

• Take it when it uncovers something. If the waste card lets you move a column and turn over a face-down card, take it.
• Take it when it steers the stock toward a card you need. If you know a 2 of hearts is sitting just below in the next group, one waste play can bring it up.
• Leave it when it only fills space. A card that lands on the tableau and blocks a column you are trying to clear can cost more than it gains.
• Do not rush it to the foundation. In Turn 3 a card taken off the waste changes the stock order for good. Send it up when that also helps, not just because you can.

One stuck position, solved by steering the stock

Positions like this one turn up again and again in the three-card game.

1. On the right of your tableau a red 6 sits on top of three face-down cards. To move it you need a black 7 somewhere on the table, and there is none. There is a red 8 in another column that could take one.
2. You go through the stock. The playable cards this pass include a red 9 and a black 5, but no black 7. You spot a black 7 at the bottom of the fourth group, the tenth card of the stock, out of reach under two others.
3. The red 9 can go onto a black 10 in the tableau. It was the sixth card of the stock, so playing it from the waste moves every card after it up by one place for the next pass.
4. On the next pass the black 7 is the ninth card, the top of the third group, and it is finally playable. You play it onto the red 8, move the red 6 onto the black 7, and the card underneath the 6 turns face up.
5. If the shift had not brought the 7 up, one tap of Undo puts the red 9 back and you try a different waste card instead.

That habit — watching the hidden cards in each group and using waste plays to bring the right one to the top — is what separates a strong Turn 3 player from someone just clicking the stock.

What the test bench says about Turn 3

Two results from our study back up everything above. A program that plays only the visible cards, never rewinds and takes waste cards by rote won just 10.6% of 10,000 three-card deals, and when it did win it had cycled the stock 3.8 times on average. Three-card games are won by patient, repeated passes, not a single sweep. For the full Turn 1 comparison and the Spider rungs, go to the Solitaire Turn 3 page.

The Turn 3 toolkit

The three-card game costs nothing and opens straight in a browser on desktop, tablet or phone; there is no installer and no login. The tool that matters most for this playbook is Undo, which is unlimited: play a waste card, run the next pass, and rewind if the shift did not pay off. You also get Hint, Restart, Auto Play, a zoom control for bigger cards, and a choice of dragging or two-tap moves. Mute, fullscreen and felt color live in the settings.

How we tested Turn 3, step by step

The figures on this page come from the spidersolitaire.us study run in September 2026.

1. Confirmed the engine. Under the hood this page and Solitaire Turn 3 are one and the same three-card game, with the stock recycling as often as you like.
2. Built the game as a program that only lets the top waste card be played, exactly as on screen.
3. Dealt 10,000 numbered hands, each from a seed so the test can be repeated exactly.
4. Chose a floor player: a bot that sees only face-up cards, follows a priority list and never undoes.
5. Played every deal to the end and counted wins and stock passes.
6. Checked the literature. The published study by Blake and Gent (2026) estimates that 81.945% of Turn 3 deals can be won by a player who knows every card, far above what a simple player achieves.
7. Calculated the margin of error: the bot's 95% range is 10.0% to 11.2%.
8. Noted the limits. The bot works through a fixed list of priorities instead of planning waste plays around the next pass, which is the skill this playbook teaches, so a thoughtful player should be able to beat it.

How the three-card test was coded

Everything is plain Python 3.11; nothing beyond the standard library was installed. The Klondike logic sits in klondike.py, and in three-card mode the program may only touch the uppermost card of each group, just as you can. Our runner split the 10,000 deals across 19 processes, and the whole batch finished in seconds. For the code and the wording of this playbook we had help from Claude, the AI assistant built by Anthropic; the figures were generated by running the program, not by the AI.

Three-card draw: the questions we hear most

What are the two cards under the top one for?

That is the rule of Turn 3. The other two show you what is coming once the top card is gone.

Why does every pass look the same?

Because you have not played anything from the waste. Take any card from it and the groups shift.

Should I switch to Turn 1 if I keep losing?

It is a good way to learn the tableau. The Klondike guide covers Turn 1 from the ground up.

Does the order of the stock ever reshuffle?

No. The stock keeps its order between passes; only your waste plays change how it is grouped.

Where does the three-card draw sit on the ladder?

It sits between one-suit and two-suit Spider on our ladder. Two-suit Spider is the next step up.

By Admin, spidersolitaire.us (Solitairen OOD). Research run September 2026.