2010/09/26

Backgammon is an interesting game

I have never been board of playing BG. It has consistently given me new interesting positions, surpriseness and awareness.
is Player 2

score: 0
pip: 70
                         
Money session
Jacoby Beaver
                          pip: 86
score: 0

is Player 1
XGID=-ABBBBCB---------bbbbbbabA:1:1:1:22:0:0:3:0:10
to play 22

1.XG Roller+Bar/23* 6/4 3/1(2)eq: -0.512
Player:
Opponent:
22.17% (G:4.06% B:0.06%)
77.83% (G:6.51% B:0.14%)
2.XG Roller+Bar/23* 7/5 7/3eq: -0.643 (-0.131)
Player:
Opponent:
19.25% (G:2.82% B:0.14%)
80.75% (G:11.50% B:0.25%)
3.XG Roller+Bar/23* 7/5(2) 6/4eq: -0.655 (-0.142)
Player:
Opponent:
18.61% (G:2.56% B:0.12%)
81.39% (G:10.93% B:0.23%)
4.3 plyBar/23* 7/5 3/1(2)eq: -0.691 (-0.179)
Player:
Opponent:
16.17% (G:2.66% B:0.07%)
83.83% (G:9.23% B:0.17%)
5.3 plyBar/23* 5/3 5/1eq: -0.692 (-0.180)
Player:
Opponent:
15.18% (G:1.77% B:0.03%)
84.82% (G:6.28% B:0.12%)

eXtreme Gammon Version: 1.21

31 32 34 is key numbers. It would be even game after those rolls. It is much better than breaking bar point and hoping that something (unrealistic) good will happen.

2010/09/19

Simple bear off

is akaisuisei

score: 2
pip: 26
                         
11 point match
                          pip: 36
score: 9

is mochy
XGID=----BBC------------ac-aa--:1:1:1:52:9:2:0:11:10
to play 52

1.Rollout16/4 5/0ffeq: -0.503
Player:
Opponent:
24.87% (G:0.00% B:0.00%)
75.13% (G:0.00% B:0.00%)
Conf: ± 0.000 (-0.503...-0.503)
Duration: 0.4 second
* 2.Rollout15/0ff 4/2eq: -0.525 (-0.022)
Player:
Opponent:
23.76% (G:0.00% B:0.00%)
76.24% (G:0.00% B:0.00%)
Conf: ± 0.000 (-0.525...-0.525)
Duration: 0.4 second
 
1 Truncated at 11 moves using 4 ply.
2592 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

After 11 years of professional career, I still do not know how to play simple bear off. What a fish.
Black has a clear advantage and Yellow needs to roll something good. Yellow needs to prepare to roll something good. It is time to prepare for fours, not deuce.
is akaisuisei

score: 4
pip: 139
                         
11 point match
                          pip: 79
score: 9

is mochy
XGID=bCCBBCa-------b--Abbbb-bA-:1:1:1:54:9:4:0:11:10
to play 54

1.Rollout117/8eq: +0.064
Player:
Opponent:
55.84% (G:29.68% B:0.21%)
44.16% (G:4.41% B:0.13%)
Conf: ± 0.012 (+0.052...+0.076)
Duration: 44.4 seconds
* 2.Rollout117/12 5/1eq: -0.004 (-0.068)
Player:
Opponent:
53.36% (G:28.88% B:0.23%)
46.64% (G:5.92% B:0.20%)
Conf: ± 0.014 (-0.018...+0.010)
Duration: 39.4 seconds
 
1 Truncated at 11 moves using 4 ply.
544 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

I chose 17/12 5/1 with the thought that I would have more pick and pass numbers because I could use six as a picking number while two as a stepping up number. That is true to 64 63 61. What I missed here is that the importance of placing my spare as useful as possible.

-If I roll deuce in next shake, it would be much better to have a spare checker on five point and a lone checker closer to home board. It is much more flexible.

-Small numbers (11, 13, 14) do not allow me to kill my five and six.

-65 force me to move full eleven pips instead of only five with 17/8.(So that 17/8 is better for immediate 65)
You should not missed to see whole picture of the position.

2010/09/16

is Kiyokazu Nishikawa

score: 0
pip: 133
                         
11 point match
                          pip: 133
score: 0

is Masayuki Mochizuki
XGID=-Ba-B-D-B---cC---cbcc-B---:0:0:1:65:0:0:0:11:10
to play 65

1.Rollout122/11eq: -0.340
Player:
Opponent:
40.74% (G:11.89% B:0.26%)
59.26% (G:16.61% B:0.62%)
Conf: ± 0.014 (-0.354...-0.326)
Duration: 2 minutes 53 seconds
* 2.Rollout113/2* eq: -0.442 (-0.102)
Player:
Opponent:
37.26% (G:11.22% B:0.12%)
62.74% (G:12.08% B:0.56%)
Conf: ± 0.015 (-0.457...-0.427)
Duration: 2 minutes 32 seconds
3.XG Roller+22/16 13/8eq: -0.454 (-0.114)
Player:
Opponent:
37.97% (G:11.80% B:0.30%)
62.03% (G:20.47% B:0.78%)
4.2 ply8/3 8/2* eq: -0.635 (-0.295)
Player:
Opponent:
32.34% (G:8.06% B:0.06%)
67.66% (G:14.44% B:0.60%)
5.1 ply8/2* 6/1eq: -0.710 (-0.370)
Player:
Opponent:
31.09% (G:6.89% B:0.07%)
68.91% (G:14.10% B:0.58%)
 
1 Truncated at 11 moves using 4 ply.
290 Games rolled with Variance Reduction.
Dice Seed: 9586
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

2010/09/15

is Player 2

score: 0
pip: 147
                         
Money session
Jacoby Beaver
                          pip: 142
score: 0

is Player 1
XGID=aaBBA-BBB---a---bbce-C-A--:1:-1:1:53:0:0:3:0:10
to play 53

1.XG Roller+23/15eq: -0.108
Player:
Opponent:
50.37% (G:17.97% B:0.93%)
49.63% (G:11.40% B:0.49%)
* 2.XG Roller+7/4 7/2eq: -0.302 (-0.194)
Player:
Opponent:
43.22% (G:16.46% B:0.75%)
56.78% (G:12.15% B:0.50%)
3.XG Roller+8/3 7/4eq: -0.345 (-0.237)
Player:
Opponent:
42.19% (G:16.80% B:0.74%)
57.81% (G:13.92% B:0.61%)

eXtreme Gammon Version: 1.21

Quite difficult for me.

2010/09/14

The need to avoid "Double Jeopardy" is rarely exist

(Special Thanks to Karam Ashoo, who kindly corrected my english.I highlighted my mistake for a future study (of english)).
is Player 2

score: 0
pip: 91
                         
Money session
Jacoby Beaver
                          pip: 54
score: 0

is Player 1
XGID=-CCCCbB-----A---a--bbbbbb-:1:-1:1:62:0:0:3:0:10
to play 62

1.Rollout112/6 2/0ffeq: +0.921
Player:
Opponent:
95.55% (G:2.17% B:0.01%)
4.45% (G:0.00% B:0.00%)
Conf: ± 0.001 (+0.920...+0.922)
Duration: 43.6 seconds
2.Rollout112/4eq: +0.910 (-0.011)
Player:
Opponent:
95.38% (G:1.01% B:0.00%)
4.62% (G:0.02% B:0.00%)
Conf: ± 0.001 (+0.909...+0.911)
Duration: 14.6 seconds
3.XG Roller+12/6 4/2eq: +0.905 (-0.016)
Player:
Opponent:
94.90% (G:1.98% B:0.01%)
5.10% (G:0.00% B:0.00%)
4.XG Roller+12/6 3/1eq: +0.901 (-0.020)
Player:
Opponent:
94.66% (G:2.03% B:0.01%)
5.34% (G:0.00% B:0.00%)
 
1 2592 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21

I liked Kit Woolsey's article about Avoiding Double Jeopardy July-August issue 1995. Avoiding Double Jeopardy means a play that stripping the six point and leaving 6-1 and 5-1 instead of having three checkers on six point and leaving 6-5 only. ADJ is actually better is from the perspective of minimizing a shot chance in the long run.
However, Kit investigated lot lots of positions and conclude concluded that the need to avoid Double Jeopardy is very rare because of race potential and missing a chance to win Gs.

Today I revisited the positions provided by Kit and can conclude that it is even more rarer rarer than he might have thought. Position above is from Kit's article and he showed it as an example of where ADJ is right.Extreme gammon thinks the opposite. Even here ADJ is not right simply because he couldn't win as many gammons as he would with 2/off.

Even so, you should be paying attention to the score. In double match point or sort of close to it, it could easily be right to play 12/4 and ADJ, all in all, it is the safest play in terms of leaving a shot. A position below is a good example.
is Player 2

score: 0
pip: 102
                         
1 point match
                          pip: 54
score: 0

is Player 1
XGID=-CCCCcB-----A------bbbbbb-:1:-1:1:26:0:0:0:1:10
to play 26

1.Rollout112/4eq: +0.920
Player:
Opponent:
95.99% (G:26.00% B:0.00%)
4.01% (G:0.00% B:0.00%)
Conf: ± 0.001 (+0.919...+0.921)
Duration: 13.9 seconds
2.Rollout112/6 2/0ffeq: +0.913 (-0.006)
Player:
Opponent:
95.67% (G:30.81% B:0.01%)
4.33% (G:0.00% B:0.00%)
Conf: ± 0.001 (+0.912...+0.914)
Duration: 35.9 seconds
3.XG Roller12/6 4/2eq: +0.922 (+0.002)
Player:
Opponent:
96.09% (G:5.75% B:0.01%)
3.91% (G:0.00% B:0.00%)
4.XG Roller12/6 3/1eq: +0.920
Player:
Opponent:
96.01% (G:5.26% B:0.01%)
3.99% (G:0.00% B:0.00%)
 
1 1332 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

Also if two point is empty, then 6/4 would be right. It is just the safest and 4/2 won't gain as much as 2/off would.
is Player 2

score: 0
pip: 102
                         
Money session
Jacoby Beaver
                          pip: 56
score: 0

is Player 1
XGID=-D-DDcB-----A------bbbbbb-:1:-1:1:26:0:0:3:0:10
to play 26

1.Rollout112/4eq: +0.958
Player:
Opponent:
95.87% (G:4.77% B:0.00%)
4.13% (G:0.00% B:0.00%)
Conf: ± 0.001 (+0.957...+0.959)
Duration: 12.4 seconds
2.Rollout112/6 4/2eq: +0.945 (-0.013)
Player:
Opponent:
95.14% (G:5.16% B:0.00%)
4.86% (G:0.00% B:0.00%)
Conf: ± 0.001 (+0.944...+0.946)
Duration: 34.2 seconds
 
1 1508 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21

2010/09/11

Concept was right but not enough

is Player 2

score: 1
pip: 116
                         
3 point match
Crawford
                          pip: 149
score: 2

is Player 1
XGID=--aBBAB-Bb---Ca--b-cBbbAb-:0:0:1:64:2:1:1:3:10
to play 64

1.Rollout120/14* 14/10eq: -0.731
Player:
Opponent:
45.39% (G:14.75% B:0.93%)
54.61% (G:31.96% B:2.32%)
Conf: ± 0.005 (-0.736...-0.726)
Duration: 5 minutes 36 seconds
2.Rollout113/3eq: -0.784 (-0.053)
Player:
Opponent:
38.19% (G:10.60% B:0.61%)
61.81% (G:27.41% B:0.56%)
Conf: ± 0.004 (-0.788...-0.780)
Duration: 6 minutes 45 seconds
3.Rollout120/14* 5/1eq: -0.850 (-0.119)
Player:
Opponent:
40.64% (G:12.22% B:0.65%)
59.36% (G:33.15% B:1.93%)
Conf: ± 0.005 (-0.855...-0.845)
Duration: 5 minutes 21 seconds
4.Rollout220/16 20/14* eq: -0.859 (-0.127)
Player:
Opponent:
43.11% (G:13.00% B:0.85%)
56.89% (G:36.06% B:2.66%)
Conf: ± 0.005 (-0.864...-0.854)
Duration: 5 minutes 46 seconds
 
1 Truncated at 11 moves using 4 ply.
2376 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

2 Truncated at 11 moves using 4 ply.
2375 Games rolled with Variance Reduction.
Dice Seed: 1341
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

My opponent hit the checker here. Even though score aruges for safe play , hitting is right. Problem with non-hitting play is that you can still lose gammon with two blots against four point board. So hitting is the right play, but he hits me wrongly. The best play is hit and run. You should remain double slotted on 23 and 20 point so that you can re-create it in case you have got return hit.