How to play Pyramid Solitaire
Clear all 28 pyramid cards by removing pairs that add up to 13. Kings count 13 on their own; queens are 12, jacks 11, aces 1.
Twenty-eight cards form a seven-row pyramid, each card pinned down by the two below it. The other 24 cards wait face down in the stock.
- Tap two open cards that add to 13 - a 6 and a 7, a queen and an ace. Both disappear.
- Kings are 13 by themselves: tap one to remove it alone.
- A pyramid card is open only when both cards covering it are gone.
- Tap the stock to turn a card onto the waste; the waste's top card can pair too. You may run through the stock three times in all.
Tips
- Before pairing, check what each removal uncovers: the same rank is often buried in both arms of the pyramid.
- Don't burn waste cards carelessly; a recycled stock comes back in the same order.
- Count the deck: if a 6 you need sits under a 7 you must pair it with, plan another route early.
Pyramid looks nothing like Klondike, and it plays nothing like it either. There's no building runs, no chasing a card down seven columns. You get 28 cards stacked into a triangle, one row overlapping the next, and the whole game is pairing: find two open cards that add up to 13 and they're gone. A 9 and a 4. A queen, worth 12, and an ace, worth 1. A king is worth 13 by itself, so it comes off alone the moment it's open. That's the whole rule set, and it's also why the game is easy to misjudge early on.
The short rules above tell you what a legal move looks like. This is about what actually decides whether you clear the pyramid: which pairs to spend and which to save, how to use your two redeals instead of burning them by row three, and an honest number for how often a deal can be won at all.
Every card is pinned by two others
A pyramid card is only open, meaning you can touch it, once both cards resting on it below are cleared. The single card at the top starts open because nothing sits on it. Everything else stays locked until you clear a path to it, and a card near the bottom row can need four or five removals above it first. The real puzzle isn't spotting a 13, it's spotting which 13s actually free something up.
This is where new players stall. It's easy to grab the first legal pair you see, but a pair from the bottom two rows barely changes your options, while a pair from the top three can open three or four new cards at once. Before you tap a pair, glance at what's underneath the two cards you're about to remove. If the answer is nothing new, it's a fine move but not an urgent one.
The stock, the waste, and your two redeals
Cards outside the pyramid sit face down in the stock. Tap it and the top card turns onto the waste, where it can pair with an open pyramid card just like anything else. Once the stock runs dry, you can turn the waste back into a fresh stock and keep going, but only twice: three passes through those 24 cards, not an unlimited cycle. The cards come back in the same order every time, so a card you buried on pass one resurfaces in the same spot on pass two.
Because the order never changes, a wasted early redeal is a real cost, not a free do-over: tapping the stock hoping something turns up spends a resource you can't get back.
The waste's top card matters more than it looks, too. Bury it under the next draw and it's out of reach until the next redeal, so check whether it's the only match left for something buried deep in the pyramid before you draw again.
Watch for pairs that share a blocker
The trickiest way to get stuck is two cards that both need the same rank, and that rank is buried behind one of them. Say a 9 in the middle of the pyramid is your only route to open a 4 two rows down, but the only other 9 left in the deck is the one covering that same 4 from the other side. You can't use the 9 without freeing the 4 first, and you can't free the 4 without a 9. Deals like that aren't always dead ends, since the waste or stock might still hold the card you need, but they're worth spotting early rather than discovering them five moves too late.
A habit that helps: before clearing a row, count roughly how many copies of that rank are still unaccounted for. Track kings especially closely, since each one comes off alone with no way to pair it out of a tight spot.
How winnable is Pyramid, really
We ran our own solver, a program that checks every possible sequence of moves from a deal, against 500 seeded deals using the exact rules this page plays with: pairs from the pyramid or the waste top, two redeals. It proved 303 winnable and 15 unwinnable outright. The rest, 182 deals, hit a search limit before the solver could finish checking every line, because a handful of Pyramid deals branch into tens of millions of positions, and a few didn't finish even at 20 million.
That gives us a proven floor of at least 60.6% of deals winnable, not a point estimate. We won't round the undetermined bucket into an average, because the two outcomes aren't equally likely: proving a deal winnable just takes finding one working line, which tends to happen fast, while proving a deal unwinnable means ruling out every line, the slow case. So the undetermined deals skew toward unwinnable, and the true rate sits above 60.6 percent, likely well above it, even if we can't say by how much.
The deals on this page are seeded and random, dealt the same way for everyone, never filtered for winnability. Some you'll lose no matter how sharp your play is, with no way to tell which from the opening layout alone. The hint button leans on the same search: when it can prove a move keeps a winnable deal winnable, that's what it shows you, and when it can't, it says so instead of guessing.