Poll
Question: Move? 
Option 1: 24/23, 13/8
votes: 2
Option 2: 24/23, 8/3
votes: 0
Option 3: 13/8, 6/5
votes: 2
Option 4: 24/18
votes: 0
Option 5: 8/3, 6/5
votes: 0
Title: game 1, move 0: Herd 5-1
Post by: diane on March 14, 2012, 10:52:08 AM
Post by: diane on March 14, 2012, 10:52:08 AM
Welcome to fibsboard match 12 all, I hope our newest worthy opponent enjoys this beat..I mean match ;)
Ok - we are off..I met with stiefnu on fibs and agreed his opening 5-3 move would be the standard one. So - to ease you gently in... how would the herd like to respond?
Ok - we are off..I met with stiefnu on fibs and agreed his opening 5-3 move would be the standard one. So - to ease you gently in... how would the herd like to respond?
Title: Re: game 1, move 1: Herd 5-1
Post by: ah_clem on March 16, 2012, 01:40:36 AM
Post by: ah_clem on March 16, 2012, 01:40:36 AM
Somehow I was asleep at the switch, and I missed this play.
I'm not sure how the tie was broken, but splitting is much better than slotting here. Slotting with an opening of 51 is doubtful, but with the opponent having a second homeboard point it's probably a blunder to slot.
The problem with slotting with 51 is that there are no builders, so even when you're not hit you can't cover the blot often enough to make the risk worthwhile. Add in the additional danger of being on the roof against a two point board and slotting is severely counterindicated.
I'm not sure how the tie was broken, but splitting is much better than slotting here. Slotting with an opening of 51 is doubtful, but with the opponent having a second homeboard point it's probably a blunder to slot.
The problem with slotting with 51 is that there are no builders, so even when you're not hit you can't cover the blot often enough to make the risk worthwhile. Add in the additional danger of being on the roof against a two point board and slotting is severely counterindicated.
Title: Re: game 1, move 1: Herd 5-1
Post by: ah_clem on March 16, 2012, 08:16:03 PM
Post by: ah_clem on March 16, 2012, 08:16:03 PM
rollout
Splitting is the right play, but "blunder" is too strong a word to describe slotting. Splitting wins only a few more games, has the same gammon rate, and only loses a few fewer gammons. The difference between the plays is only a third of a percent MWC. I wish all my errors were this small.
1. Rollout 24/23 13/8 Eq.: -0.203
0.456 0.122 0.007 - 0.544 0.168 0.007 CL -0.143 CF -0.203
[0.001 0.002 0.001 - 0.001 0.002 0.001 CL 0.004 CF 0.009]
2. Rollout 13/8 6/5 Eq.: -0.246 ( -0.043)
0.450 0.122 0.006 - 0.550 0.178 0.014 CL -0.177 CF -0.246
[0.002 0.002 0.001 - 0.002 0.003 0.002 CL 0.008 CF 0.019]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 910706162 and quasi-random dice
Play: world class 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]
Spoiler
Splitting is the right play, but "blunder" is too strong a word to describe slotting. Splitting wins only a few more games, has the same gammon rate, and only loses a few fewer gammons. The difference between the plays is only a third of a percent MWC. I wish all my errors were this small.
1. Rollout 24/23 13/8 Eq.: -0.203
0.456 0.122 0.007 - 0.544 0.168 0.007 CL -0.143 CF -0.203
[0.001 0.002 0.001 - 0.001 0.002 0.001 CL 0.004 CF 0.009]
2. Rollout 13/8 6/5 Eq.: -0.246 ( -0.043)
0.450 0.122 0.006 - 0.550 0.178 0.014 CL -0.177 CF -0.246
[0.002 0.002 0.001 - 0.002 0.003 0.002 CL 0.008 CF 0.019]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 910706162 and quasi-random dice
Play: world class 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]
[close]