Search This Blog

October 27, 2013

Chris Yep game 1 - another backgame

Some positions from first game of match against Chris Yep. This was played at the Michigan versus Ohio challenge. This time we look at the position from the defenders point of view.

This is the first time I have ever wrote about my opponent's mistakes rather than my own. I apologize in advance if Chris objects. Two reasons for this.

1) I was confused at first and mixed up who was who. I wrote up all these positions and then realized I was the other side!!

2) Backgames have been on my mind lately.


66 to play


is Bill

score: 0
pip: 128
                         
7 point match
                          pip: 172
score: 0

is Chris Yep
XGID=-----CCABa--dB--b--f-Bb-B-:0:0:1:66:0:0:0:7:10
to play 66

1.XG Roller++21/9*(2) eq: +0.400
Player:
Opponent:
59.30% (G:19.06% B:0.62%)
40.70% (G:11.72% B:0.75%)
2.XG Roller++21/9* 13/7(2)eq: +0.342 (-0.059)
Player:
Opponent:
57.37% (G:20.26% B:0.79%)
42.63% (G:12.05% B:0.87%)
3.3-ply21/15 21/9* 13/7eq: +0.288 (-0.113)
Player:
Opponent:
57.84% (G:17.84% B:0.59%)
42.16% (G:13.88% B:1.09%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

Chris moved 21/9*, 13/7(2).

This makes a 4 1/2 point block but leaves his 3 back men too stranded.

21/9*(2) just looks better. It also leaves a 4 1/2 point block but with good prospects of filling in the gap to make a 5 prime. Chris's back men have some connectivity with the midpoint, and at any rate my structure isn't geared towards priming here so Chris has little to fear here. Basically I need to pop a seven from the bar this roll or it's over. And even if I do Chris may hit back when I am back in the soup. At worst, I will have a mountain of work (even if I hit) before I can double. My stacky and gappy structure (six men on the 6 point) is just too ugly.

*****************

There has been a lot of hitting since the first position. And now I have a 1-4 backgame, not the best game but a game. 41 to play from the bar for Chris.

is Bill

score: 0
pip: 172
                         
7 point match
                          pip: 141
score: 0

is Chris Yep
XGID=-b-BbCBBBA-ab--a---dA-baAA:1:-1:1:41:0:0:0:7:10
to play 41

1.XG Roller++Bar/20eq: +0.514
Player:
Opponent:
68.76% (G:28.69% B:2.68%)
31.24% (G:5.71% B:0.21%)
2.XG Roller++Bar/21 24/23* eq: +0.455 (-0.059)
Player:
Opponent:
64.08% (G:34.14% B:4.71%)
35.92% (G:5.61% B:0.24%)
3.3-plyBar/21 9/8eq: +0.415 (-0.099)
Player:
Opponent:
65.07% (G:29.59% B:3.53%)
34.93% (G:7.48% B:0.37%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

Chris hit with Bar/21, 24/23*

Best was the no-hit play Bar/20.

This is an instructive moment. Why is no hit better?

Timing: My timing is suspect here. I am only down about 30 pips. That is simply not enough to hold the back points and build an offense and not crash.

Advanced home board: I already have 7 men in my homeboard. This is way too far advanced for a backgame. Moreover, my man on the deuce point totally blows.

"It's not that you made that move, it's that you wanted to make that move!!" This was a typically funny yet insightful comment GM Ben Finegold gave me in a chess lesson several years ago. I think it applies here.

It's not that hitting the checker on the 2 point is bad (it is). Rather, it's that Chirs even wanted to hit it. This is so bad it should not even be considered.

The checker is a liability for me and hitting it has no upside. Well, actually, there are some variations where Chris sweeps up all the blots and rolls home to a gammon. But most likely all Chris does is remove my badly placed checker while giving me some desperately needed additional timing.

The proper technique in fighting a poorly timed backgame is to challenge it head on. Force me to try and win going forward immediately. Force me off my advanced anchor, at which time Chris hopefully wins the battle for that key home board point. It will prove extremely difficult for me to win going forward. ie, escape and attack at the same time.

*****************

33 to play.

is Bill

score: 0
pip: 191
                         
7 point match
                          pip: 136
score: 0

is Chris Yep
XGID=-b-BcCBBBA-ab--a---dAAbA--:1:-1:1:33:0:0:0:7:10
to play 33

1.XG Roller++23/20 21/15* 5/2eq: +0.512
Player:
Opponent:
65.67% (G:35.18% B:4.58%)
34.33% (G:4.70% B:0.18%)
2.XG Roller++23/20 21/15* 9/6eq: +0.461 (-0.051)
Player:
Opponent:
62.90% (G:36.74% B:5.27%)
37.10% (G:4.08% B:0.16%)
3.XG Roller+23/11* eq: +0.515 (+0.003)
Player:
Opponent:
65.03% (G:36.68% B:5.19%)
34.97% (G:4.41% B:0.23%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

Chris moved 23/20, 21/15*, 9/6.

Better was 23/20, 21/15*, 5/2. This is another common anti-backgame move which Chirs may or may not know about.

The point is that Chris's spare on top of his prime is useless and misplaced. Slotting the two point is a ploy to gain timing at virtually no cost. If I roll a poor number like 26, 25, 23, 22 then I am forced to hit. This reduces my timing disadvantage, while allowing Chris to move his man around the board and ultimately target the key point on the board right now --> the 9 point. Chris would love to extend the prime if he could. Again, the main point is this play has no downside since the spare on the 5 is pointless right now anyway.

***************

64 to play.

is Bill

score: 0
pip: 203
                         
7 point match
                          pip: 124
score: 0

is Chris Yep
XGID=-abBcCCBB--ab--A---dB-b---:1:-1:1:64:0:0:0:7:10
to play 64

1.XG Roller++15/11* 7/1* eq: +0.440
Player:
Opponent:
62.10% (G:40.52% B:3.05%)
37.90% (G:5.30% B:0.24%)
2.XG Roller++15/9 5/1* eq: +0.354 (-0.087)
Player:
Opponent:
61.40% (G:34.68% B:2.25%)
38.60% (G:6.16% B:0.28%)
3.XG Roller+20/14 5/1* eq: +0.409 (-0.031)
Player:
Opponent:
62.33% (G:37.22% B:2.74%)
37.67% (G:5.80% B:0.28%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

This one I am less confident about the reasons ...

Chirs moved 15/9, 5/1* perhaps with similar logic to the last comment. He slots the back of the prime and doesn't really care too much if he gets hit.

The computer likes 15/11*, 7/1*. Maybe the computer is giving up on trying to bust the backgame timing. Is this strategy too likely to fail at this point? (a guess) ....

Therefore it's better to switch gears with 15/11*, 7/1*. I assume the idea is to keep me on the bar for a while. This allows Chris to scramble his checkers home while I am on the bar. If this is successful Chris will win more gammons, and also by delaying my offensive development I will be less ready to contain a late shot should I be fortunate to hit one.

****************

Things took a bad turn since we last saw this game. I redoubled. Take or Pass?

is Bill

score: 0
pip: 144
                         
7 point match
                          pip: 107
score: 0

is Chris Yep
XGID=-BcBBBCB--a-----aaabbbb-AA:1:-1:-1:00:0:0:0:7:10
on roll, cube action?

Analyzed in XG Roller++ No redouble Redouble/Take
  Player Winning Chances: 70.87% (G:20.12% B:0.86%) 71.27% (G:20.45% B:0.96%)
  Opponent Winning Chances: 29.13% (G:9.99% B:0.65%) 28.73% (G:10.25% B:0.57%)
  Cubeless Equities +0.529 +1.074
Cubeful Equities
     No redouble:+0.764 (-0.159)
xg Redouble/Take:+0.922
     Redouble/Pass:+1.000 (+0.078)
 
Best Cube action: Redouble / Take
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

Chris passed and I don't blame him. I think I would have passed too.

It looks real bad. Most likely I make the bar point and then Chris enters and starts cracking.

But the computer thinks he has enough to take. Most likely isn't all the time. Sometimes I don't make the bar. Sometimes Chris dances for a while and I don't escape his 5-prime. Sometimes Chris anchors and counterprimes. Sometimes Chris anchors and wins much later.








October 17, 2013

Ali - Backgame

Finally to the last game!! It's DMP and there are too many blunders too count.


42 to play.

is Ali

score: 6
pip: 95
                         
7 point match
                          pip: 226
score: 6

is Bill
XGID=-----ABAA--aaBB--abcbcbDB-:0:0:1:42:6:6:0:7:10
to play 42

1.XG Roller++8/4 7/5eq: -0.072
Player:
Opponent:
46.38% (G:1.53% B:0.07%)
53.62% (G:31.59% B:11.32%)
2.XG Roller++14/10 7/5eq: -0.094 (-0.022)
Player:
Opponent:
45.28% (G:1.42% B:0.07%)
54.72% (G:33.28% B:12.59%)
3.XG Roller+8/4 6/4eq: -0.071 (+0.001)
Player:
Opponent:
46.45% (G:1.18% B:0.04%)
53.55% (G:31.80% B:11.54%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

Not a big mistake to start, but an instructive one. At least it would be instructive if I knew what the proper lesson was. I;m afraid I don't know what's up with this one.

I played 14/10, 7/5. Make the points in order and I don't care if I get hit.

But should I care? Timing is a major consideration and I should have enough here ... Does getting hit here help or hurt? It might help if I get hit and dance for a while. It might help if I get hit and recycle. It could hurt if I come in right away and never roll a six. I'd say it's not clear on the timing issue.

Probably the main factor is I already have six men back and that's enough!! The more men back, the more split my army becomes. After all, I want to have a good offense in place and this will be very difficult to assemble with seven men back! But this is just a guess on my part.

**************

53 to play.

is Ali

score: 6
pip: 27
                         
7 point match
                          pip: 187
score: 6

is Bill
XGID=-----BBB-B-AA---A--c--cBB-:0:0:1:53:6:6:0:7:10
to play 53

1.XG Roller++16/13 11/6eq: -0.178
Player:
Opponent:
41.11% (G:0.00% B:0.00%)
58.89% (G:34.58% B:23.83%)
2.XG Roller++11/6 9/6eq: -0.181 (-0.004)
Player:
Opponent:
40.94% (G:0.00% B:0.00%)
59.06% (G:35.59% B:24.81%)
3.XG Roller++23/20 16/11eq: -0.276 (-0.099)
Player:
Opponent:
36.18% (G:0.00% B:0.00%)
63.82% (G:41.19% B:28.62%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2


The computer likes any play which keeps the double anchor. My preference would be for 16/8 which slots the next point in the prime but XG doesn't analyze this move.

I panicked and 23/20, 16/11. This abandons the double anchor but gives Ali 4's to play. He's down to six men and I am feeling the pressure to hit something. On principle my play makes "sense" to me but apparently my gut is just way off here.

Let's try to see how the numbers pay after splitting 23/20 compared to holding the anchors ....

65. Wash. Good roll either way.
64. +Split. Disaster for Ali if I split. Good roll if I hold anchors.
63. Wash. Good roll either way.
62. Wash. Disaster for Ali either way.
61. Wash. Disaster for Ali either way.
54. +Split. I get immediate shot on split. Roll is stuck is hold anchors.
53. Wash. Good roll either way.
52. Wash. I get double shot either way.
51. +Anchors. 6/5*/off if I split. 6/5/off good roll with the anchors but at least I am not in the air.
43. +Split. I get an immediate shot if I split. Plays safe if I hold anchors.
42. Wash. Plays safe for Ali either way.
41. +Anchors. Point on head if I split. Double shot if I hold anchor.
32. Wash. Double shot either way.
31. +Anchors. I get an almost double shot from bar if I split. Double shot at two different blots if I hold anchors.
21. +Anchors. If I split, pick/pass/score 3/2*/off. If hold anchor it plays safe but I keep the anchors and am on the ground.
66. Wash. Killer roll for Ali either way.
55. Wash. Doesn't play either way.
44. +Anchors. Superjoker if I spilt. Roll doesn't play if I hold anchor.
33. Wash. Great roll either way.
22. +Split. A good roll but at least I get a shot by splitting. Awesome roll for Ali if I hold anchor.
11. +Anchors. A pretty good roll either way as I get a shot but I think having the anchor is better long-term.

So tallying up ...

Anchors win on 4 rolls plus double 4 and double 1 or 10 rolls total.
Splitting wins on 3 rolls plus double 2 or 7 rolls total.


I guess that's the reason. Anchoring wins on net 3 rolls and eliminates 44 as a superjoker.

**************

53 to play.


is Ali

score: 6
pip: 36
                         
7 point match
                          pip: 171
score: 6

is Bill
XGID=--a--BBB-B-BA-A----a-AbAaA:0:0:1:53:6:6:0:7:10
to play 53

1.XG Roller+Bar/20 5/2* eq: +0.082
Player:
Opponent:
54.10% (G:0.00% B:0.00%)
45.90% (G:27.45% B:11.53%)
2.XG Roller+Bar/20 9/6eq: +0.016 (-0.066)
Player:
Opponent:
50.82% (G:0.00% B:0.00%)
49.18% (G:20.62% B:6.99%)
3.XG Roller+Bar/20 12/9eq: +0.011 (-0.071)
Player:
Opponent:
50.56% (G:0.00% B:0.00%)
49.44% (G:21.83% B:6.61%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

One of the few that I got right. I am desperate to get the second checker so Bar/20, 5/2* is called for.


******************

41 to play.


is Ali

score: 6
pip: 59
                         
7 point match
                          pip: 199
score: 6

is Bill
XGID=-aa--ABB-B-AA-A----aAAbAAA:0:0:1:41:6:6:0:7:10
to play 41

1.XG Roller++Bar/24 9/5eq: -0.096
Player:
Opponent:
45.20% (G:0.00% B:0.00%)
54.80% (G:28.63% B:14.67%)
2.XG Roller++Bar/21 20/19* eq: -0.412 (-0.316)
Player:
Opponent:
29.39% (G:0.00% B:0.00%)
70.61% (G:46.34% B:24.81%)
3.3-plyBar/24 23/19* eq: -0.053 (+0.043)
Player:
Opponent:
47.36% (G:0.00% B:0.00%)
52.64% (G:37.31% B:17.67%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2


This is where I blew the game. I have actually studied these positions before and "knew" the right answer but I couldn't resist the forbidden fruit when push came to shove in the heat of a real game.

The correct strategy for this game is as follows:

I need to hit the first checker to have any chance of winning.

I also want to hit a second checker. Why? Because Ali already has 10 checkers off. If I manage to close Ali out with a single checker captured I will still be an underdog.

After I hit the second checker I absolutely do not want to hit the third man!! This is fundamental. The correct strategy is to hit two men and close out two men. If Ali anchors, then I simply break the prime, let Ali jump out and then proceed to close him out and scoop up the second man. Simple!!


Again I knew this yet went down the wrong path anyway. I really made only one conceptual mistake but it translates into checker play whopper almost every move.

The problem with hitting the third man is that once Ali anchors it is very difficult to force him to break the anchor. He can always keep throwing the third man around the board and not really care if I hit it or not. (Of course I shouldn't hit it at all).

So the best play is simply Bar/24, 9/5, This ignores the third man while remaking the key 5-point. Now I can concentrate on coaxing Ali's blots out to the midfield where I can eventually pick them up and close them out.

***************

42 to play.

is Ali

score: 6
pip: 71
                         
7 point match
                          pip: 201
score: 6

is Bill
XGID=-a--aaBB-B-BA-A-----ABbAA-:0:0:1:42:6:6:0:7:10
to play 42

1.XG Roller++14/10 12/10eq: -0.078
Player:
Opponent:
46.11% (G:0.00% B:0.00%)
53.89% (G:19.31% B:5.53%)
2.XG Roller++9/5* 7/5eq: -0.271 (-0.193)
Player:
Opponent:
36.44% (G:0.00% B:0.00%)
63.56% (G:40.92% B:21.95%)
3.3-ply23/21 9/5* eq: +0.173 (+0.250)
Player:
Opponent:
58.63% (G:0.00% B:0.00%)
41.37% (G:33.33% B:19.64%)

eXtreme Gammon Version: 2.02, MET: Kazaross XG2



I made the "auto" play of 9/5*, 7/5 remaking the 5 point. What could be more logical?


This is another example of the wrong strategy. Again I want to force Ali out and around so I can focus on two blots only.


*******

51 to play.


is Ali

score: 6
pip: 67
                         
7 point match
                          pip: 213
score: 6

is Bill
XGID=---aaBBa-A-BA-A-----ABbAAA:0:0:1:51:6:6:0:7:10
to play 51

1.XG Roller++Bar/24 14/9eq: -0.323
Player:
Opponent:
33.85% (G:0.00% B:0.00%)
66.15% (G:39.63% B:19.65%)
2.XG Roller++Bar/24 12/7* eq: -0.484 (-0.161)
Player:
Opponent:
25.81% (G:0.00% B:0.00%)
74.19% (G:54.37% B:33.14%)
3.XG Roller++Bar/19eq: -0.552 (-0.229)
Player:
Opponent:
22.42% (G:0.00% B:0.00%)
77.58% (G:57.23% B:32.41%)

eXtreme Gammon Version: 2.02, MET: Kazaross XG2



I won't show every position but here is the theme in action again. The computer likes Bar/24, 14/9 ignoring the third man and starting to build a block against the remaining back two men.

I of course hit with Bar/24, 12/7*

***********

22 to play


is Ali

score: 6
pip: 70
                         
7 point match
                          pip: 181
score: 6

is Bill
XGID=---abCCA-B----A---A-A-bAB-:0:0:1:22:6:6:0:7:10
to play 22

1.XG Roller++23/21 14/12 9/7(2)eq: +0.294
Player:
Opponent:
64.72% (G:0.00% B:0.00%)
35.28% (G:10.92% B:3.78%)
2.XG Roller++14/12 9/7 9/5eq: +0.265 (-0.029)
Player:
Opponent:
63.25% (G:0.00% B:0.00%)
36.75% (G:12.45% B:4.44%)
3.XG Roller++18/14 9/7 5/3* eq: +0.100 (-0.194)
Player:
Opponent:
55.00% (G:0.00% B:0.00%)
45.00% (G:16.43% B:5.93%)

eXtreme Gammon Version: 2.02, MET: Kazaross XG2


One more with the same theme.

I made the bar, moved in the outfield 18/14 and hit loose 5/3*

The right idea is to make the bar 9/7 and any other deuces are OK. Don't hit!! Wait for him to break the anchor so you can focus on closing out two blots!!

**********

63 to play.

is Ali

score: 6
pip: 68
                         
7 point match
                          pip: 165
score: 6

is Bill
XGID=----cBCCB--A--------AAaaB-:0:0:1:63:6:6:0:7:10
to play 63

1.XG Roller++20/17 11/5eq: +0.414
Player:
Opponent:
70.71% (G:0.00% B:0.00%)
29.29% (G:6.89% B:1.88%)
2.XG Roller++20/14 11/8eq: +0.410 (-0.004)
Player:
Opponent:
70.49% (G:0.00% B:0.00%)
29.51% (G:6.67% B:1.85%)
3.XG Roller++21/15 20/17eq: +0.374 (-0.040)
Player:
Opponent:
68.68% (G:0.00% B:0.00%)
31.32% (G:7.57% B:2.07%)

eXtreme Gammon Version: 2.02, MET: Kazaross XG2



Actually I don't really care about the right play here. I want to focus on the winning percentage. I am about 70/30 to win after the best play. Pretty decent but I had closed two men out and Ali having 10 men off I would be in the low to mid 80's to win. a big difference!!