Problem of the Week #40: December 6


Cash game. Center cube. Black on move.




(a) Should Black double? If he doubles, should White take or drop?




(b) Should Black double? If he doubles, should White take or drop?

a)double/drop
b)double/take

I think they're both double/drop.

In (a), black has too many threats. Any 6 but 65 makes a good point, any 2 or 4 hits a blot. White doesn't look good in any of those scenarios.

In (b), white is way behind in the race without an advanced anchor, and his builders aren't working for him -- lots of 5s and 6s play badly already.

Problem 40(a)
There are a lot of bad things that can happen to White if Black is allowed to continue this game. Any roll that let's Black make his bar point (8 shots) should be counted as a market loser. In addition, any two (11 shots) gives Black a strong hit on the twenty-one point. And in case neither of these happen, Black has some random rolls that hit and make his two point (64, 66 and 44). Black's double is clear.

If White knew that Black would never make his bar point, he could accept a double. Gammon chances are low, and the anchor on the three point provides enough shot equity for a take, while giving White a launch point for escape. Without that guarantee, however, White should probably pass.

My solution in Part (a): Double, drop.

Problem 40(b)
Offensively, Black is much weaker in Problem 40(b), but so is White's defense. Rolls like 11, 61 and, perhaps, 44 that allow Black to make a broken five-point prime might be classed as market losers. Otherwise, Black has few immediate threats. In fact, offensive structure is nearly the same on both sides of board. Black's only big advantage is his racing lead. Does that, together with his threat to escape his lone backman give Black a double? I would wait. White's take is easy.

My solution in Part (b): No double, take.

For the Record
I am so often wrong that I like to post my record in these messages. It's kind of a truth-in-advertising thing. I have been answering these problems without the use of a bot, and before checking the excellent solutions of others, since Problem 28. Including the 39(a) tossup, my record at this writing is 60% correct.
Correct: 28a, 29, 30, 32, 35, 36, 38, 39a, 39b
Incorrect: 28b, 31, 33a, 33b, 34, 37

a) double/take
b) double/drop

(a) Double/Drop
(b) Double/Take

a) Double/take

The double seems obvious. Black has a lot going on, and he stands to lose his market if he hits followed by white dancing, or if he builds the 5 prime. These combinations are enough to warrant doubling now.

I don't like it much, but I can't quite give up as white from this position. (Note: I also take far too many cubes according to GNU.) Black's anchor is disconnected from the rest of the checkers, so he's going to have a tough time bringing those checkers around. White is not going to get completely primed in, so that 3 point anchor will always have daylight to make an escape. I think this gives white just enough to take.

b) Double/take

Black's double is pretty easy since white isn't in very good shape. With only one checker to clear over the 3 prime, it's not that daunting of a task. Black's home board is a work in progress, but unless something bad happens in the next couple rolls, he should have another builder or two ready to go.

White's take is a lot easier than on the previous problem. Black has *some* work to do to get the last checker out, and white has *some* space to either run a checker from the ugly 2 point or advance his anchor. Black isn't threatening to make his prime too much more menacing just yet.

a) no double ( very close btw ) , take
b) double take

Good arguments, Aaron. I didn't have time to give a detailed explanation, but I had it a borderline drop because Black is threatening to send back a 4th checker and also has a few rolls that would make the very important bar point.

On the other side, Black has lost his midpoint, which causes some disconnection like you said. All in all, it looks like a very efficient double.

you loose too many gammons in a.

We seem to have hit one of everything for both parts.

Yes, this problem seems to have hit some sort of sweet spot as far as doubling decisions go.

It would be even more interesting if the reasons for the decisions were posted!

Looks like D/T in both positions.

Black has a big advantage in both and will be crushing White if things go well, which argues for cubing. White can counterprime Black's straggler(s) if things don't go well, which argues for taking.

I'd post my reasoning but it looks like I almost certainly have part b wrong because no one else thinks that it is a double/drop. It appears that I'm underestimating white's chances of escaping out of black's home board in part b and that white has better winning chances in part b.

With all the different opinions, one thing is clear: no one is cheating by using a bot! It's nice to see the spirit of fun hasn't been overtaken by the desire to be right.

I'm confident that Part (a) is a double, and that Part (b) is a take. I'm less confident that that Part (a) is a drop, and that Part (b) is not a double.

In Part (a), I've betrayed myself in my choice of language. I didn't say that Black had already lost his market, rather I stated that the eight rolls that make the bar point are market losers. Hmm...

Yes, this problem seems to have hit some sort of sweet spot as far as doubling decisions go.

hmm no part (b) is definitely a double

I was thinking this, but went to
a) Double/drop
b) wait/drop