A big thank you to Timothy, Clem and Julia for their thoughtful contributions.
This is the first one and all the team chose to make the Golden Anchor.
Falafel actually chose the hit, which is a small error, after considerable thought. Would our panel have done the same over the board, i.e. not presented as a problem? Nobody will ever know, so kudos to the panel there.
Timothy and Julia both liked 9/3*, but I am afraid that that is way off beam. With four men back, White's blitzing options are very poor, due to shortage of ammunition. Falafel correctly covered the 9pt and played 23/21. This is second best, a medium error and the best play is 20/18, 13/9. With this, White blocks sixes from the back and the midpoint, forcing Black to play them deep into his home board, generating lots of shots for White. This is an unusual situation, as it's rarely correct to switch from the 20pt to the 18pt this early in the game. Hard to find over the board or as a problem. I can't really say I would have hit on it.
Lastly, the play from game 9. Mochy lead 1-away, 3-away, Crawford. Black trails in the pipcount 146-152 before the roll.
A unanimous verdict from our panel, as indeed there was from the peanut gallery on GridGammon. Hit twice is correct and Mochy's play of 13/8, 9/8 got the bird. It isn't terrible, a small error at worst, but 8/7*. 6/1* is best, better than hitting once because fewer shots.
Although it's the Crawford game, Mochy should make an anti-positional play and hit. He needs to put White off-balance for a roll, otherwise White will just move ahead in the race and improve his poition at leisure.
To illustrate the dangers of generalising from this though, look at two more positions. The first comes from the first game of this fascinating match. Mochy is on roll with what looks very like the one above. It's game one, so no score to 13.
Now the answer is 14/8 by a very small margin over hitting twice. Here Black leads in the race 157-169 and has three men back against one, encouraging him to play safe. The big surprise though, is that if you choose to double hit, 8/7*/2* is correct! When you play 8/7*, 6/1*, you duplicate your own sixes and also leave four blots instead of three. This counts for a lot when it comes to tidying up.
You all see this position often.
Everybody knows that you hit twice with 6-5, but what about 5-1? Then it's incorrect and just 13/7* will do nicely. At GammonGo the double hit is almost equal. Note again the duplication of your own sixes after the double hit.
I hope you learned as much as me from these positions. You can see all of the Mochy v. Falafel match on an XG file, with rollouts and annotated by me, with comments from Mochy and Falafel. It costs $10 US, 7-50 euros or 6-50 pounds sterling. It is also available as a file that can be read with Gnu or Snowie and as a Word doc. for the more "hands on" types who like to learn on a real board. For the Word doc., add 50%, which will also include the XG or Gnu file.
Paypal to dorbel(at)gmail(dot)com please.
Until the next time, enjoy the game!
Subscribe to: Post Comments (Atom)
Position (2) - once you'd said about Black's sixes after 13/9 20/18, and I sat down and worked out how awful they were for him, as well as some other rolls like 51, the lightbulb came on. An excellent position to study, clearly!
It is an interesting position, but not perhaps one from which to draw general conclusions. You don't usually choose to shift from the 20pt to the 18pt, which is probably why falafel didn't choose it.
Nice problems. I actually didn't think that Problem 1 had much QF; I'm pretty confident I would have anchored OTB.
Problem 2 is interesting! I think that one should not primarily view this as an example where you want to switch from the 20pt to the 18pt; rather, it's that 13/9 is mandatory and 20/18 is the least bad 2. I would guess that if we were given the option of not playing the 2 at all (of course this is illegal), then we would opt not to play it. The important part of the play, I think, is that White needs to contain Black's straggler, and blocking with 13/9 is much more effective than hitting here. This is an error I need to work on correcting. I quadruple-whoppered a similar-themed problem that Chuck Bower posted to BGOnline recently.
For problem 3, I'm curious as to whether Mochy's safe play is best at a normal match score. I'd guess that the double hit loses more gammons than it wins in this position, but this is tricky to estimate.
Thanks Timothy, Position 1 I included because falafel thought about making the anchor. "How bad can it be not to hit?" was his take on it and then hit. A mistake that he makes is always interesting.
In position 3 the double hit is correct by a wider margin earlier in the match. As you say, both positions lose more gammons than they win, but the proportion is better after the bolder play. Of course the cube is then in play. After the double hit, the opponent almost has a double from the bar! Although technically a small error, I would do it at 0-0 in a 13pt match. A gammon win from here (23%)might put him on tilt for the rest of the match. I've seen weaker doubles than this passed too. After the safe play, the opponent has a correct double.
Post a Comment