This happened in a match on safe harbor today. The cube action analysis surprised me.

No way I'm doubling at this score when so many of my wins are gammons. Regardless of whether White should take or not, it's like you're throwing away all the extra equity you get on gammons by doubling. If the take is close enough and the gammons are low enough, you can justify cubing at the score, but it appears the gammon pct. is way too high here to justify a cube, regardless of the closeness of the take.

If doubled as White, I'm pretty sure I have the raw 25% needed to take. Cause if White anchors soon, at worst he has an OK deuce point game, and there's plenty of play before then. If Black hits loose (which he should), hitting back should push White way above 25%. So I think these scenarios are likely enough to give White >25% on average.

Big pass for money but no double/take at the score.

Here's a way to think about this that some might find useful. Say you set up a position at this score where White has a close take and Black wins 0% gammons. Obviously this is a good time for Black to double, and if the position is reasonably volatile, it's going to be a pretty big error for Black to not double.

But now modify the position, so White's win % stays the same, but now Black wins a few % gammons (and the overall volatility of the position stays about the same). What happens the size of the error of not doubling?

It actually goes down. Because now some of the time, when Black wins a gammon, doubling doesn't actually help him, and doubling and losing the game hurts him just as bad.

Up the gammon % a little more (still keeping White's winning % the same), the error size goes down some more. Up the gammon % enough and the error "flips". It actually becomes an error to double, even though White still has the same close take!

Understanding this stuff on some level is key to handling the cube well when nearing the end of a match.

Quite interesting that it seems too good for Black and possibly a take for White at the same time. You gotta love tournament play!

And how about when black is 3 away v 2 away?

White needs to be very cautious at that score, his take point is 29% + the effect of gammons.

So this would be no double/pass (aka too good to double). Because if White greatly improves by anchoring or Black hits loose and White hits back, it looks like Black will often still have a strong double at the score (he's maybe even still too good to double in some of these variations).

So Black can try for the gammon without much risk of White getting an easy take in if things go bad for Black. And there's obviously plenty of gammon chances here.

Indeed, the evaluation (confirmed by rollout) was too good to double / take. I wonder how common that is in match play.