Home / Answers

Should I have a backup hand in American Mah Jongg?

No. A backup hand costs attention and gives almost nothing back, because you cannot predict which of the two directions will die. Commit to your single strongest option and let the table tell you when to move. The move has two triggers and neither of them is a feeling.

Answered by Prim. Last reviewed 2026-08-21.

Why two hands is worse than one

The case for a backup sounds sensible: keep a second direction alive so you are not stranded. In practice you pay for it on every single pick. Tiles that would be obvious discards get held because they serve the other hand, your discards slow down and give away more than they should, and you arrive at the middle of the wall committed to nothing in particular.

The reason it does not pay is that you cannot know in advance which hand is going to die. So you are spending real tiles and real attention to insure against something you cannot predict, which is the definition of a bad price. Pick your strongest option, pass by pass and pick by pick, and put everything behind it.

The two triggers to switch

You move when your hand is weakened, meaning the tiles you still need are going down in the discards, or wounded, meaning they have been taken up into somebody's exposures. Both of those are things you can see. Neither of them is available to you in advance, which is exactly why holding a backup does not help.

What this gives you is a clean test to apply instead of a vague sense that things are not going well. Look at what you still need. Ask how many of each are already gone. If the answer has quietly become none left, that is your signal, and it does not require a second hand to have been sitting in your rack for twenty minutes.

Most people who say pivot mean something else

A pivot is abandoning a hand you had committed to. If you never committed in the first place, and you were staying fluid while the tiles pointed you somewhere, that is not a pivot at all. It is adaptive play, and it is a perfectly good way to open a hand.

The distinction matters because the two get very different advice. Staying fluid early is cheap and often right. Abandoning a committed hand late is expensive and should need a reason you could say out loud. Calling them by the same word makes players think the expensive one is normal.

If you are going to move, move early

There is a window for this. Switching directions is cheap while there are still plenty of tiles left to draw and almost nothing has been exposed, and it gets more expensive with every pick after that. A workable marker is somewhere around seventy tiles remaining: if you are going to change your mind, that is roughly when.

Past that point the arithmetic turns. A late switch means starting a new hand with a shorter wall and a table that has already shown you what it is holding, and by then the honest question is usually not which hand to move to but whether this hand is one to fold.