Problem of the Week #55: March 21


Cash game, White owns the cube.




Black to play 6-2.


Note: All ‘cash game’ problems assume the Jacoby Rule is in effect. That is, you can’t win a gammon unless the cube has been turned.

I like 23/17 10/8.

I like 23/15, even if hit twice I will probably enter quickly and have a good game, otherwise he will probably hit once and the other checker has a good chance to escape and there will be lots of fly shots as well which are pretty strong and add to gammon chances.

Well, it looks like the priority is now to bear-in quickly. Black would have liked to scoop another blot for gammon chances, but he cannot hit with this roll. Black doesn't want to alter his home board. He also doesn't want to break any of his outer board points. That leaves plays involving the back checkers. It doesn't gain much to slot the 9-point. It leaves Black with 23/15. When leading in the race, run.

23/15

At first glance, it looks like the 10-point is important as it blocks White 6's. But Black's lead is 48 pips (56 after the roll), so even with White rolling 6-6, he would still be leading by 30+ pips. Anchoring on the 17-point also has the merit of enhancing Black's chances to hit another checker. However, keeping White in a holding game seems more important, especially since White holds the cube and Black cannot double him if he moves the back checkers with 6-6 (if Black plays 23/17 10/8).

So there's no clean way to play a 2, and nothing nice to play after 17/11 (we have to leave a shot off the 4 and in the back). So we're left with 23/17. Then what? 17/15 is basically conceding a shot with 3 hitting numbers. Not that we can't withstand it, but why would we beg for it? I'm never messing up my board playing 5/3 over 10/8 or 8/6. I prefer 8/6 for two reasons. It gets a guy in, and also, if white does hit, it's harder for him to clean up his 3 blots since the 10 threatens the blot on the 8, but not vice versa.

23/17 8/6

23/15

Besides running (which is what you need to do now), duplicating 3s and 5s can lead to white becoming awkward much more quickly than he would like, since he's got 84 pips tied up in that anchor and he needs those numbers to get those guys moving.

Since black has a stronger board I would go with 23/15. If hit, reentering seems easy. I think breaking the 10 and 8 point would be a bad decision.

For the six, there seems to be only one possibility: make the outfield point, 23/17. The only other play would be to run one man, 17/11, surrendering outfield control. After running, no matter how the two is played, one checker would be left stranded in White's innner board. Let's reject that.

So, that leaves 23/17 as the best six, and Black need only choose a two. Here, there are two reasonable choices:

• 10/8 - This move gives White 14 shots, but keeps all four points in Black's home board.
• 5/3 - The alternative maximizes safety, at the cost of busting - for the moment - Black's home board. White would have 11 shots.

Breaking the board seems premature. Black should keep his strong inner board as a threat to White's blots in the outfield. If he goes for safety instead, he'll have to forego most of his hitting options on the next turn.

My solution: 23/17, 10/8

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 39a tossup, my record at this writing is 54% correct.
19 Correct: 28a, 29, 30, 32, 35, 36, 38, 39a, 39b, 42b, 43, 44, 45, 47, 48a, 48b, 50, 52b, 53. 16 Incorrect: 28b, 31, 33a, 33b, 34, 37, 40a, 40b, 41, 42a, 46, 48c, 49, 51, 52a, 54.

Having now read the posts of others, I'm embarrassed that I didn't even consider leaving a triple shot in the outfield. The duplication of White's escape numbers (threes and fives) might just work. White could hit with twenty-seven shots (any 1, 3 or 5). All the even numbers miss.

• 25% - White misses both blots, and Black is suddenly a big favorite.
• 50% - White hits one blot.
• 25% - White hits both blots, perhaps swinging the momentum in his favor.

I can't say I'm convinced this is best, but I can certainly see the merit. As with so many of these problems, the difference between best and second best is a small percentage often measured in single digits.

Grunch: 23/17, 8/6.

leave the 10 point seems ugly to me . white is bad in timing and need to move some back cheker quickly in order to gain some flexibility. thats why in black pants i like the 23/15 . 5 and 3 are duplications of rolls white needs to move from back, so it's a very good position( indeed i think in next roll white should'nt hit but escape ) .

Ugly roll.
The safe play doesn't look correct in this type of position where white is way behind and has made only one point.
I'm not sure what is best here but I'm gonna go with 23/15.
The probable result is that one checker will get hit and one will escape. Black doesn't mind getting hit and losing a few pips. It is more important to retain the 10 point and not to mess up his structure with a play like 8/6, 8/2

After wondering for weeks, I finally looked it up:

grunch - To reply to an original post on a web forum without first reading the other replies.
Source: http://www.urbandictionary.com/define.php?term=grunch

Yours is another credible solution worthy of consideration. Tom Cowley also chose this move. All in all, this is one tough problem. In my response, I chose 23/17, 10/8 over this because it yields a better distribution, and breaks from the rear. The eight point will be easier to clear later than the ten point.

Interesting you found it in the urbandictionary. There is actually a poster called Grunch on these forums who would reply to strategy posts without looking at anyone else's answers.