Playing With Three — and With Two, Five or Six
Yes, and there is an official way to do it. There is also one official rule that almost every table quietly ignores.
You can play the American game with three, and the League has published how. Build all four walls, deal nobody into the empty seat, and give each player twelve tiles before the last ones go out. Then the official rule says to skip the Charleston — and that is the rule almost every table refuses to follow.
The three-player deal
Build four walls as usual. The empty seat still has a wall; it simply never gets tiles.
Each player takes four tiles in turn, exactly as in the four-player game, until everybody holds twelve. The last tiles are where it gets interesting. Three procedures are in circulation, and they are not rival opinions — they replaced each other.
All three end with the dealer holding one more tile than everyone else, which is the point. Sloper reckons they are effectively equivalent. The dice already broke the wall, so nobody can engineer which tiles they get. His practical advice is not to argue about it — pick one procedure and use it for the whole session.
That matters more than it sounds. Three players at a table can easily have learned three different deals — one from the 2024 newsletter, one from the 2020 rulebook, one who only ever played with four.
The rule everybody ignores
After the deal, the official rules say to go straight to the dealer's first discard. No Charleston.
Sloper is blunt about it: many if not most American players reject skipping the Charleston, and in the great majority of his own three-player games the others insisted on having one.
So the rule is clear and the practice is clear and they disagree. We are not going to pretend otherwise, and we are not going to tell you your table is wrong.
A three-player Charleston is a table rule, which means nobody can tell you the official way to run it, because there is not one. If your group invents one, these are the decisions Sloper says you have to make:
- Do you deal into the empty chair and let players pass to and from it?
- Or do you take Charleston tiles from the empty seat's wall instead?
- Must tiles be taken in a set order, or may a player take whichever they like?
Answer those three and write the answers down. The failure mode of a table rule is not that it is unofficial — it is that two people at the table remember it differently.
Two players
The same shape. Sit opposite each other and build all four walls. Deal to two players only, with no dummy seats: the dealer holds fourteen tiles and the other player thirteen, as usual. No Charleston. The dealer discards first.
Five players
The normal answer is that the fifth player sits out. Everyone rolls the dice at the start, and the low roller sits out the first hand. At the end of each hand the dealer steps away and the player who was out takes that seat. The rotation looks after itself.
The fifth player looks at everyone's hand once the Charleston is done, decides who will win, and writes that name down away from the table. They then say nothing — any comment at all voids the bet.
At the end of the hand they collect exactly as their chosen player collects, or pay exactly as that player pays. The League sells a bettor for $12, so this is not a fringe practice.
Six players
Two sit out and rotate in. The previous dealer stands up, one of the sitting-out players takes the seat, and it goes round in order. It is the five-player rotation with one more person in the queue.
When you have read it all: the rules workbook, 105 questions with an answer key that cites the rule behind every answer, $19.
Where this comes from
All of it comes from Tom Sloper's FAQ 13A, which sets out the deals for two, three, five and six players. It cites the League's own documents for each — the January 2024 newsletter, the 2020 edition of the official rulebook, and the rulebook's pages on other numbers of players.
We have not read those newsletters and rulebook editions ourselves. This page reports what Sloper says they contain, which is one credible source rather than two, and we would rather say that than imply we checked the primary documents.
That most tables reject the no-Charleston rule is Sloper's observation of his own games and of the players who write to him. It is an observation, not a rule, and it is his rather than ours.