When to take a double, and when to pass

Somebody turns the cube. Now you have a decision, and it's worth more than most of the checker plays you'll make all evening.

Take, and the game is worth double. Pass, and you hand over what it's worth right now and set up again.

You only need to win one game in four

Start with what each choice costs.

Pass, and you give up the game now. That costs you whatever the cube currently says: usually one point.

Take, and you play on at double. You'll either win two points or lose two.

So passing is a small certain loss, and taking is a gamble at twice the stake. The only question is how often you'd need to win for that gamble to pay.

Imagine the same position four times over, and you win one of the four:

Pass every time: -1, -1, -1, -1. Four points lost.

Take every time: -2, -2, -2, +2. Four points lost.

Identical. So one win in four, or 25%, is the break-even line. Win more often than that from such positions and taking is better. Win less often and passing is.

Which is why a position can look bad and still be a take. You don't need to be winning. You need a one-in-four chance.

One extra nudge towards taking: once you accept, the cube sits on your side of the table and only you can double next. That's worth enough to take a shade under 25%.

So the common mistake is passing too early

Most beginners drop doubles they should take. A position can look grim and still be a comfortable take.

The usual reason is confusing "I'm losing" with "I can't win". Being behind in the race while holding an anchor, with your opponent still having checkers to bring home safely, is often a perfectly good 30%.

The flip side: when gammons change the maths

The 25% rule assumes you'll lose one point if you lose. If you're in danger of being gammoned, you'll lose two.

That makes passing right at better than 25%. If you're stuck on the bar against a strong board with no checkers off yet, drop the cube and keep your losses to one point. A gammon risk is the main reason to pass a position you'd otherwise take.

And from the other side of the table

If you're the one thinking about doubling, the window is roughly 70% to 78% winning chances: strong enough that your opponent should still take, not so strong that they'll pass and deny you the game.

Double too early and you give away your cube for nothing. Double too late and they simply pass, and you've won one point instead of two. Good players aim to double just before the position becomes a pass.

How to estimate, without doing maths at the table

Three questions get you most of the way:

• Can I still win the race? Count the pips, add eight for whoever is on roll.
• Do I have contact? An anchor, a shot to wait for, checkers of theirs still to come home.
• Could this become a gammon? Checkers trapped, nothing off yet, their home board strong.

Yes to the first two, no to the third, and you almost certainly have a take.

The short version

• Take if you'll win about one game in four. That's 25%.
• You can take a shade under, because owning the cube is worth something.
• Passing too early is the most common beginner error.
• Gammon risk raises your take point: pass sooner when you could lose double.
• Doubling is right at roughly 70% to 78%.
• Ask: can I win the race, do I have contact, could this become a gammon?