| |
BGonline.org Forums
early 64 play
Posted By: Chase In Response To: early 64 play (bob koca)
Date: Tuesday, 27 May 2008, at 11:05 a.m.
I spent some time with Gnu trying to figure out what the bots see that that the humans are missing. These are all "after-the-fact" observations, of course, trying to make sense of the rollout rather than coming by them honestly.
According to Gnu 2-ply, here is what White is looking at, after each of Blue's moves, as far as the number of blots to shoot at, single-hitters, double-hitters, and no-hitters.
If Blue plays 24/18 13/9:
blots: 2
sh: 12 34 36 = 6
dh: 11 13 15 16 23 33 66 = 11
nh: 14 22 24 25 26 35 44 45 46 55 56 = 19If Blue plays 24/18 8/4*:
blots: 3
sh: 11 12 13 15 16 22 23 25 26 33 35 44 45 46 = 24
dh: 14 34 36 56 = 8
nh: 52 55 66 = 4
What jumps out is the trade-off in the number of return hits. By playing 13/9, Blue gives White 3 more double-hitters in exchange for reducing Whites single hits by 18. In a position where the fur is about to fly, that seems huge--especially considering that, at this stage of the game, a double-hit isn't all that much worse than a single hit. Blue is behind no matter what, so he simply makes an advanced anchor (or two) and plays from there.
I was impressed by, after Blue's 13/9, how often White should pass up hits in order to improve positionally. It does makes sense, however. Again, Blue is already behind, and he can readily enter, so the hit is less damaging than usual.
13/9 is interesting in some other ways. First, it holds down Blue's blot count. Second, it often increases White's blot count, as he'll be hitting from his 8 and midpoint,creating new blots rather than hitting with existing blots. While Blue is playing with 2 blots, White could easily end up with 4 or 5. Getting hit may no be the end of the world for either player now, but it could be fatal in a turn or two.
Third, White is ahead in the race, and will likely hit and gain more ground. Thematically, he'd like to be moving his back checkers around the board. By leaving all his blots in White's outfield and making his blocking 9pt, Blue makes it difficult for White to follow his game plan; White will usually be forced to engage on his own side of the board.
Finally, 13/9 creates a solid offensive asset. Giving up the midpoint this early could be costly later, but the real battle right now is over the infield. After 13/9 Blue is guaranteed to have 11 checkers in the zone, whereas, after 8/4*, he'll likely have only 9, and possibly only 8.
So, what does a human take away form this in the way of general principles which will help him find 13/9 over the board? I think we need to start by looking at our assumptions. (I'm certainly not trying to pick on any of the posters. I made some of the same assumptions.)
"I don't want to miss my mid-pt so soon in order to make the not-much-important 9-pt."
The early game is about development and keeping game plans open. The position after 13/9 looks to give Blue the flexibility to move into a race, blitz, priming, holding, or back game. The loss of the midpoint is unfortunate, but the fight is for development and the immediate action looks to be up front.
"If you don't hit you are not fighting back, you just allow white to hit you AND advance forward."
The impact of hitting is a lot less when neither side has any development and when the hitting looks to continue over several turns. Positional advantages become more important.
"Surely blue will be hit after 24/18 8/4* also but it will be a more even blot hitting fight."
Maybe not. There is something to be said for locking up key assets, reducing your blot count while encouraging your opponent to increase his, and, of course, fighting the blot-hitting contest on his side of the board. These things are especially true if the hitting war looks to continue for a while.
"We have nine checkers into zone so blitz is not impossible."
Two of those checkers are blots, so this number may drop. Meanwhile, the alternative play guarantees that we will have 11 checkers in the zone.
"I can't say I understand why trading the mid point for the 9pt is good here."
My best guess is that it's a combination of the things observed above. And even at that, the plays are fairly close. I'd be interested if any of these ideas trigger any other thoughts from others.
Interesting position, Bob. Thanks for posting it.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.