Free the game.

We didn't guess our way to a good card. We studied twenty-one years of American mahjong cards — how the hands are built, how the sections move, what makes a year feel fresh — and then mapped the full design space a card can draw from — 86,977 hand designs, every one scored for difficulty. Every card we publish is drawn new from that space. Nothing recycled, nothing copied, nothing you've played before.

Then we measured difficulty the honest way: by playing. Every hand is run through simulation — dealt and drawn over and over — before it earns a line on the card, so the easy hands are truly easy and the showpieces earn their points. We ran the numbers so you can run the table.

We read every card so you don't have to

For twenty-one years — 2006 through 2026 — American mahjong has gotten a new card every spring. We tracked down all of them and catalogued every line: 995 printed hands, each recorded as plain fact — what wins, what it's worth, whether it has to stay concealed. Every year was cross-checked against multiple sources and graded for confidence, because the world contains misprints and marketplace fakes and we refuse to build on either. Then we added fourteen more cards from independent publishers, for range. As far as anything published shows, no archive like this existed. Now it does.

What 86,977 actually means

A card looks like seventy arbitrary hands. It isn't. Every line ever printed is written in a quiet grammar — group shapes, number relationships, suits standing in as roles — and once you've catalogued twenty-one years of it, you can map everything that grammar allows. Stay inside the traditional American style and the map holds 16,370 possible hand designs. In twenty-one years, the annual cards have printed only a few hundred distinct ones. Open the map to patterns that independent publishers and the world's other mahjong traditions already play, and it grows to 86,977.

Precision matters here, so here is exactly what that number is: 86,977 distinct hand designs, each one catalogued and each one rated for difficulty. We call it the scored design space. It is not every random pile of fourteen tiles — that number is far larger, and mostly nonsense no card would print. It is every design a real card could plausibly carry. And we scored all of it.

How we know “largest ever” isn't just talk

Our manifesto calls this the largest analysis of the game ever run. This page is the receipts. Before we said it, we searched — books, academic literature, player forums, publisher materials — for any prior mapping of the game's design space, and found none published anywhere. So the claim rests on numbers you can hold in your hand: twenty-one years of cards catalogued line by line, 86,977 designs mapped and rated, millions of simulated deals behind the ratings. If someone surfaces a bigger body of work, we'll update this page and thank them. Until then, it stands.

How we measure difficulty: by playing

The honest way to learn how hard a hand is: play it until it tells you. So we did — at computer speed. For every design in the space, the computer deals tiles and draws toward the hand, over and over, solitaire-style, counting how often it comes together and how fast. Speed matters, because a real game ends — the table does not wait politely while you assemble a masterpiece. Every design gets hundreds of simulated deals at minimum; the rare, stubborn ones get thousands, until the number settles.

Then we check the machine against the real world: published win-rate statistics drawn from millions of actual games, and independent difficulty ratings — both from sources outside this project. Where our numbers disagree with reality, ours move.

One finding worth your time: printed point values track difficulty only loosely. Inside a single point class we found a sixty-five-fold spread in actual win rates — same points, wildly different odds. You have felt this at the table. Now it's measured.

What this buys you at the table

Three things. Point values that mean something — when our card says a hand is worth more, the data says it's genuinely harder. Difficulty you can trust — the easy sections are truly easy, and the showpieces earn their keep. And new hands forever — at roughly seventy hands a card, 86,977 designs is more than a thousand years of fresh cards. We'll pace ourselves.

What we don't claim

Honesty is the whole point of the project, so, plainly:

• We are independent. Not affiliated with, sponsored by, or endorsed by any mahjong league, association, or card publisher.
• Our point values are estimates. Good ones — simulated at scale and checked against real-game statistics — but estimates, and they will keep improving as play data grows.
• We never reproduce anyone's card. Every card we release is a new selection from the design space, machine-checked against the full archive for overlap before it ships.
• A simulation is not your weekly game. Four players, table talk, the friend who always calls your tile — real play is richer than any model. We measure what can be measured and label the rest honestly.

One set. Every variant. That's the whole pitch. The tiles that play your Tuesday American game play Hong Kong on Thursday and Riichi when your nephew visits — no second set, no adapter kit, no asterisks. We're building it the way you'd want your own set built: heavy enough to feel serious, pretty enough to leave out, and finished paying for itself the day it arrives. It's coming. The waitlist hears everything first.

Also mugs, tees, and a kitchen towel that says “In this house, the card is free” — because some opinions deserve to be laundered and hung back up. Everything we print passes one test: would you carry it into your own game without explaining yourself. The card money's in the snack fund now; the snack fund deserves a nice apron.

Let's say the gracious thing first, because it's true: without the League, there is no American mahjong. Someone had to standardize the hands, settle the arguments, and mail the card every spring — and for nearly ninety years, they did. That is stewardship. Decades of order, of care, of a game kept alive and kept in tune. We thank the League for ninety years of it, and we mean every word.

Here is where we part ways, and it's smaller than it sounds. When the annual card was born, printing one and mailing it to every player in America was real work, and a small fee was a fair way to pay for it. That arrangement belonged to its era — the era of the typewriter and the postage stamp. Ours isn't it. A rulebook to a ninety-year-old game should cost what rulebooks to old games cost: nothing.

Because the game was never the card's. The game belongs to the players — to the standing Tuesday, the snack table, the same seats, the woman teaching the new girl the Charleston. The renewal has done its job. The game can carry itself from here.

So here is what we're doing about it.

Free cards, forever. Built from the largest analysis of the game ever run, with every hand's difficulty measured by simulation before it earns a line. Print them at home. Play them online. A new card every season, at the same price every year: nothing.

One set, every variant. American, Chinese classical, Hong Kong, Riichi — one beautifully made set that plays them all. Bought once, finished paying for itself the day it arrives.

Tools, not gatekeepers. We show our work, publish how the cards are made, and hand you everything you need to play. No memberships. No renewals. No permission slips.

Play the card you love — nobody at this table scolds anybody. Just know that from now on, there is always a free seat, and the card that comes with it costs the same this year, next year, and every year after that.

Free the game.