Poll
Question: Move? 
Option 1: 13/6
votes: 0
Option 2: 13/9, 8/5
votes: 0
Option 3: 13/10, 8/4
votes: 3
Option 4: 13/10, 13/9
votes: 2
Option 5: 8/1*
votes: 2
Title: Game 1, move 3: Herd 4-3
Post by: diane on March 17, 2012, 10:20:01 PM
Post by: diane on March 17, 2012, 10:20:01 PM
(http://www.fibsboard.com/fibsboard-forum-match-12/game-1-move-3-herd-4-3-4995/?action=dlattach;attach=7374;image)
Title: Re: Game 1, move 3: Herd 4-3
Post by: ah_clem on March 18, 2012, 02:20:36 AM
Post by: ah_clem on March 18, 2012, 02:20:36 AM
We're behind in the race, but we have a great anchor. We can afford to make bold moves. Bringing two down threatens to do serious damage to our opponent. If he escapes or hits, we're no worse off. But if he fails to hit or escape we're positioned to clobber him.
Bring down two - only 7 shakes hit for our opponent, but many many more are devastating for our side. And we can recover quite nicely from a hit.
Bring down two - only 7 shakes hit for our opponent, but many many more are devastating for our side. And we can recover quite nicely from a hit.
Title: Re: Game 1, move 3: Herd 4-3
Post by: diane on March 18, 2012, 02:55:03 AM
Post by: diane on March 18, 2012, 02:55:03 AM
You can tell when things just got a whole lot trickier when you have 4 votes and 4 different opinions! ;)
Someone will have to have a rethink - or make their case.
[note - if the tie isnt broken, my favourite move gets it - unless you have talked me out of it!]
Someone will have to have a rethink - or make their case.
[note - if the tie isnt broken, my favourite move gets it - unless you have talked me out of it!]
Title: Re: Game 1, move 3: Herd 4-3
Post by: stog on March 18, 2012, 10:43:37 AM
Post by: stog on March 18, 2012, 10:43:37 AM
Spoiler
ok i'm switching my original vote of 13/9 8/5
8/1* no need to do this at this stage
13/10 8/4 and 13/10 13/9 both do what my original choice attempted to do, but 13/10 13/9 strips our midpoint, so i go with 13/10 8/4
8/1* no need to do this at this stage
13/10 8/4 and 13/10 13/9 both do what my original choice attempted to do, but 13/10 13/9 strips our midpoint, so i go with 13/10 8/4
[close]
Title: Re: Game 1, move 3: Herd 4-3
Post by: ah_clem on March 18, 2012, 03:06:00 PM
Post by: ah_clem on March 18, 2012, 03:06:00 PM
13/10 8/4 is a reasonable play and may well be best - it's bold where bold play is called for. If we're not hit, any 2 4 5 or 9 covers it. But I like bringing down two better (D) - slotting the four leaves 14 shakes to hit while D only gives up 7. Granted slotting gives more chances to make the four point, but D gives almost 75% to make either the 4 or the bar point. My take is that the slightly smalller chance of making a good point next time is worth paying to leave half of the hitting numbers. D also brings an additional checker into the zone, making a blitz more realistic.
Either must be better than hitting loose on the ace - with only nine checkers in the zone a blitz is premature.
Either must be better than hitting loose on the ace - with only nine checkers in the zone a blitz is premature.
Title: Re: Game 1, move 3: Herd 4-3
Post by: ah_clem on March 20, 2012, 01:39:13 PM
Post by: ah_clem on March 20, 2012, 01:39:13 PM
Well, glad that we finally got this move sorted out.
rollout:
Here are the top four plays. Hitting loose is not among them and 2-ply analysis has it as a double whopper so I didn't bother to roll that out.
gnu likes D best, with the quiet 13/9 8/5 not too far behind. Perhaps Diane was trying to do us a favour by making that move instead.
1. Rollout 13/10 13/9 Eq.: +0.221
0.541 0.131 0.003 - 0.459 0.067 0.003 CL +0.156 CF +0.221
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.004 CF 0.008]
2. Rollout 13/9 8/5 Eq.: +0.177 ( -0.044)
0.527 0.125 0.003 - 0.473 0.059 0.003 CL +0.130 CF +0.177
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.004 CF 0.008]
3. Rollout 13/6 Eq.: +0.111 ( -0.110)
0.511 0.112 0.003 - 0.489 0.057 0.003 CL +0.088 CF +0.111
[0.001 0.002 0.000 - 0.001 0.001 0.001 CL 0.003 CF 0.007]
4. Rollout 13/10 8/4 Eq.: +0.104 ( -0.117)
0.514 0.130 0.003 - 0.486 0.088 0.003 CL +0.077 CF +0.104
[0.001 0.002 0.000 - 0.001 0.001 0.001 CL 0.004 CF 0.009]
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]
rollout:
Spoiler
Here are the top four plays. Hitting loose is not among them and 2-ply analysis has it as a double whopper so I didn't bother to roll that out.
gnu likes D best, with the quiet 13/9 8/5 not too far behind. Perhaps Diane was trying to do us a favour by making that move instead.
1. Rollout 13/10 13/9 Eq.: +0.221
0.541 0.131 0.003 - 0.459 0.067 0.003 CL +0.156 CF +0.221
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.004 CF 0.008]
2. Rollout 13/9 8/5 Eq.: +0.177 ( -0.044)
0.527 0.125 0.003 - 0.473 0.059 0.003 CL +0.130 CF +0.177
[0.001 0.002 0.000 - 0.001 0.001 0.000 CL 0.004 CF 0.008]
3. Rollout 13/6 Eq.: +0.111 ( -0.110)
0.511 0.112 0.003 - 0.489 0.057 0.003 CL +0.088 CF +0.111
[0.001 0.002 0.000 - 0.001 0.001 0.001 CL 0.003 CF 0.007]
4. Rollout 13/10 8/4 Eq.: +0.104 ( -0.117)
0.514 0.130 0.003 - 0.486 0.088 0.003 CL +0.077 CF +0.104
[0.001 0.002 0.000 - 0.001 0.001 0.001 CL 0.004 CF 0.009]
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]