Search This Blog

March 24, 2014

Fleet Underwood v Faris

Continuing to look at mistakes from the Fleet Underwood tournament.  These are taken from second round match against Faris.

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

9 away 9 away.  65 to play.



is Faris

score: 0
pip: 164


9 point match

pip: 151
score: 0

is Bill

XGID=-aa---EaD---dD--Ab-cbAa---:0:0:1:65:0:0:0:9:10
to play 65

1.XG Roller++21/16 13/7* eq: +0.174

Player:
Opponent:
54.70% (G:14.40% B:0.65%)
45.30% (G:12.65% B:0.42%)

2.XG Roller++13/7* 6/1* eq: +0.087 (-0.087)

Player:
Opponent:
51.87% (G:17.14% B:0.81%)
48.13% (G:14.79% B:0.79%)

3.XG Roller++13/7* 7/2* eq: +0.075 (-0.099)

Player:
Opponent:
51.44% (G:16.28% B:0.99%)
48.56% (G:14.27% B:0.59%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I hit twice with 13/7*/2*

Best is hit and outside anchor, 13/7*, 21/16.  While this play is unlikely to work, it has the biggest upside when it does and the least downside when it doesn't.

Note that 13/7*, 6/1* is the best of the hit two plays.  Even though it's more blots, this gives the chance to "steal the bar" in an exchange of hits.  If missed altogether I am best diversified to do something constructive next time, including just making the bar point.

So While I will probably be hit after 21/16, 13/7* I may not be.  And if I'm not, then racing is the game plan here.  So get the backmen moving.  If I am hit at least I have return shots from the bar.

Hitting twice slows him down but I am outboarded so a hitting contest favors him.  Also, even if he misses I have liabilities everywhere (four blots) which need cleaning up. 

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

8 away 8 away.  Cube action?



is Faris

score: 1
pip: 147


9 point match

pip: 151
score: 1

is Bill

XGID=--a-ABD-C-A-eA---c-eaB--A-:0:0:1:00:1:1:0:9:10
on roll, cube action?

Analyzed in XG Roller++ No double Double/Take
  Player Winning Chances: 63.45% (G:27.51% B:0.78%) 63.55% (G:27.74% B:0.78%)
  Opponent Winning Chances: 36.55% (G:6.84% B:0.26%) 36.45% (G:6.92% B:0.25%)
  Cubeless Equities +0.507 +1.079
Cubeful Equities
     No double:+0.698 (-0.073)
xg Double/Take:+0.771
     Double/Pass:+1.000 (+0.229)
Best Cube action: Double / Take
eXtreme Gammon Version: 2.10, MET: Kazaross XG2

Once again I missed a strong double.  I have a good position but the race is close but I have three men back, and I haven't hit that second man yet, and I have lots of scattered blots.  And even if I do hit I have a bit of work to do, or so I figured, so I rolled.

The key to this cube, and a recurring theme lately, is that Faris's position is so weak.  No board, and still only 3 points.  Totally stacked midpoint and six point.  Even when things go bad, it will take a long time before Faris can ever recube.  Also, the gammons are significant.  This was not obvious to me at first blush.  I think this can be traced back to the weakness of opponent's game.  If I do hit there are some gammon and or prime possibilities.  Also, Faris is so stacked that he may be coughing up blots as the game continues.

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

Next roll.  Cube action?



is Faris

score: 1
pip: 167


9 point match

pip: 144
score: 1

is Bill

XGID=a-a-BBD-C---eA---c-eAA--A-:0:0:1:00:1:1:0:9:10
on roll, cube action?

Analyzed in XG Roller++ No double Double/Take
  Player Winning Chances: 73.73% (G:41.44% B:1.07%) 73.39% (G:44.55% B:1.13%)
  Opponent Winning Chances: 26.27% (G:4.64% B:0.19%) 26.61% (G:5.22% B:0.25%)
  Cubeless Equities +0.898 +1.943
Cubeful Equities
     No double:+1.096
     Double/Take:+1.706 (+0.610)
     Double/Pass:+1.000 (-0.096)
Best Cube action: Too good to double / Pass
Percentage of wrong take needed to make the double decision right: 13.6%
eXtreme Gammon Version: 2.10, MET: Kazaross XG2

My doubling window tuning fork is out of tune!!  After this joker sequence (hit and dance) I am too good but just cashed.  Oh well.

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

I lead 7 away 8 away.  52 to play.



is Faris

score: 1
pip: 139


9 point match

pip: 135
score: 2

is Bill

XGID=---b-CD-BC-Ab-a--bbbbBb---:0:0:1:52:2:1:0:9:10
to play 52

1.XG Roller++9/4 6/4eq: +0.271

Player:
Opponent:
58.09% (G:10.76% B:0.27%)
41.91% (G:7.59% B:0.22%)

2.XG Roller++21/14* eq: +0.215 (-0.056)

Player:
Opponent:
57.18% (G:13.47% B:0.58%)
42.82% (G:12.12% B:0.56%)

3.2-ply11/9 6/1eq: +0.042 (-0.229)

Player:
Opponent:
52.37% (G:9.43% B:0.25%)
47.63% (G:8.67% B:0.20%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I hit the blot but should have made the point.

I am slightly up in the race so figure I don't want to be priming here.  But the 52 does play well by making the 4 point.  This establishes a broken 5 prime, and I have some stall time to try and escape later.    Also, the hit now is a bit risky as most entering 2's, 3's and 4's hit me back at a time when I am outboarded.

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

I lead 5 away 8 away and have already doubled.  63 to play.



is Faris

score: 1
pip: 179
                         


9 point match
                         
pip: 147
score: 4

is Bill

XGID=-ba-BaD-BA--bDA--d-e----A-:1:-1:1:63:4:1:0:9:10
to play 63

1.XG Roller++9/3 6/3eq: +0.185

Player:
Opponent:
64.81% (G:23.44% B:1.69%)
35.19% (G:7.17% B:0.26%)

2.XG Roller++14/5* eq: +0.103 (-0.082)

Player:
Opponent:
62.42% (G:24.72% B:2.39%)
37.58% (G:8.80% B:0.38%)

3.XG Roller++8/5* 8/2* eq: +0.072 (-0.114)

Player:
Opponent:
61.57% (G:27.51% B:3.41%)
38.43% (G:10.31% B:0.66%)

4.XG Roller++14/8 9/6eq: +0.069 (-0.116)

Player:
Opponent:
62.09% (G:20.65% B:1.38%)
37.91% (G:7.65% B:0.25%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I thought I needed to try and conquer the 5 point, so hit twice 8/5*, 8/2*.  This play is just too big though.  This is an all eggs in one basket approach.  I have five blots exposed after all.  Faris may turn the game around quickly. 

Just the steady improvement play 9/3, 6/3 was called for.  Although Faris will likely anchor on the 5 point, he will still need to do a lot to win from there.

I have a 3 point board and Faris's offense is totally undeveloped.  My gameplan should have been to simply come around clean and win the race.  Some times you want to blitz and or bust the backgame but the dice don't cooperate.  I should have accepted that and switched plans to a simple race.

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

Later in the same game.  15 to play.



is Faris

score: 1
pip: 163
                         


9 point match
                         
pip: 178
score: 4

is Bill

XGID=-ba-B-D--B-abC---c-bcBA-aA:1:-1:1:51:4:1:0:9:10
to play 51

1.XG Roller++Bar/24* 21/16eq: -0.324

Player:
Opponent:
52.85% (G:19.12% B:1.84%)
47.15% (G:13.69% B:0.77%)

2.XG Roller++Bar/24* 13/8eq: -0.475 (-0.151)

Player:
Opponent:
49.97% (G:18.47% B:1.84%)
50.03% (G:14.84% B:0.74%)

3.2-plyBar/24* 9/4eq: -0.518 (-0.193)

Player:
Opponent:
47.76% (G:16.66% B:1.56%)
52.24% (G:14.40% B:0.64%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I moved Bar/24*, 13/8. 

Best was Bar/24*, 21/16

I was probably afraid to leave the security of the anchor.  But Faris only has seven men
in the offensive zone so I am not in much danger.  The priority here is keeping men circulating, in play.

Another consideration is I give Faris good sixes from the bar.  This is usually a backgammon sin and here is no exception.

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

Still later in the same game.  64 to play.



is Faris

score: 1
pip: 134
                         


9 point match
                         
pip: 123
score: 4

is Bill

XGID=---aCbCbBBB--C-----bbabab-:1:-1:1:64:4:1:0:9:10
to play 64

1.Rollout110/6 10/4eq: +0.334

Player:
Opponent:
70.32% (G:13.31% B:0.71%)
29.68% (G:3.99% B:0.13%)
Conf.: ± 0.008 (+0.326...+0.342) - [100.0%]
Duration: 3 minutes 45 seconds
2.Rollout113/3* eq: +0.240 (-0.095)

Player:
Opponent:
69.26% (G:13.78% B:0.93%)
30.74% (G:6.53% B:0.11%)
Conf.: ± 0.010 (+0.230...+0.249) - [0.0%]
Duration: 4 minutes 22 seconds
3.3-ply8/4 8/2eq: +0.157 (-0.178)

Player:
Opponent:
66.38% (G:10.21% B:0.30%)
33.62% (G:5.14% B:0.02%)

 
1 1296 Games rolled with Variance Reduction.
Moves and cube decisions: 3-ply
eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I hit loose with 13/3*.   Logic was to keep my structure and try and get Faris to crack.  I didn't think I was in much danger since Faris has a double blotted board.

And yet the computer thinks clearing a point with 10/6, 10/4.  Probably I am up the race so just race.  This clears a point in front of a double anchor and gives some spares to play.  Faris must play somewhere next time and may crack anyway.  So with a spare or two I may be poised to attach next time if I can't clear the midpoint safely.

*****************
I lead 3 away 8 away.  54 to play.



is Faris

score: 1
pip: 171
                         


9 point match
                         
pip: 151
score: 6

is Bill

XGID=-aa---EaD-A-dC---c-eB-----:0:0:1:54:6:1:0:9:10
to play 54

1.XG Roller++10/1* eq: +0.111

Player:
Opponent:
57.22% (G:16.35% B:0.44%)
42.78% (G:7.61% B:0.26%)

2.XG Roller++6/1* 6/2* eq: +0.047 (-0.064)

Player:
Opponent:
55.54% (G:17.43% B:0.51%)
44.46% (G:8.72% B:0.31%)

3.XG Roller+13/8 10/6eq: +0.106 (-0.004)

Player:
Opponent:
57.17% (G:16.23% B:0.44%)
42.83% (G:7.65% B:0.24%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2


I hit twice with 6/2*, 6/1*

I am anchored and Faris so getting hit isn't the end of the world.  But a double hit is the wrong idea.  A blitz is very unlikely here and besides the best gameplan is simply to race.

So simply 10/1* leaving the minimum number of shots and flushing Faris forward if he misses.

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

Later in the same game and I am on roll.  Cube action?



is Faris

score: 1
pip: 128
                         


9 point match
                         
pip: 110
score: 6

is Bill

XGID=-ABB-BB-A-c-cC-B---ccba---:0:0:1:00:6:1:0:9:10
on roll, cube action?

Analyzed in XG Roller++ No double Double/Take
  Player Winning Chances: 69.02% (G:2.64% B:0.03%) 69.30% (G:2.64% B:0.04%)
  Opponent Winning Chances: 30.98% (G:2.52% B:0.07%) 30.70% (G:2.24% B:0.05%)
  Cubeless Equities +0.384 +0.784
Cubeful Equities
     No double:+0.473
     Double/Take:+0.259 (-0.214)
     Double/Pass:+1.000 (+0.527)
 
Best Cube action: No double / Take
Percentage of wrong pass needed to make the double decision right: 22.4%
eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I know Stick's rule at 3 away many away with contact is "never double."   I thought this might be an exception since it is likely to switch to a race where I have a big lead.  So I doubled.

For money this is a borderline double but is a terrible blunder at this score.  The problem is there is contact and a long race ahead.  Faris should aggressively recube to 4 if he begins to turn things around.  I only need 3 points to win so a 4 point game is certainly not what I want.

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

I lead 3 away 4 away and am on roll.  Cube action?



is Faris

score: 5
pip: 134
                         


9 point match
                         
pip: 114
score: 6

is Bill

XGID=---bBDC-B----Dba-abcbb----:0:0:1:00:6:5:0:9:10
on roll, cube action?

Analyzed in XG Roller++ No double Double/Take
  Player Winning Chances: 76.41% (G:3.63% B:0.06%) 77.16% (G:3.72% B:0.07%)
  Opponent Winning Chances: 23.59% (G:1.25% B:0.02%) 22.84% (G:1.07% B:0.02%)
  Cubeless Equities +0.577 +1.637
Cubeful Equities
     No double:+0.954 (-0.046)
     Double/Take:+1.158 (+0.158)
xg Double/Pass:+1.000
 
Best Cube action: Double / Pass
eXtreme Gammon Version: 2.10, MET: Kazaross XG2

My cube errors are all over the place.  Last time too aggressive; this time too wimpy.

I didn't double here.  This time I was afraid of getting a recube for the match at the wrong time.  So I wanted to lock the game up first.

Trouble is the game actually is locked up already!!  I am up 20 pips, with good structure and Faris is stuck back on the 3 point anchor.

For money this is borderline Take for Faris.  But at this score he should prefer to drop and play from 4 away 2 away where he can play aggressively for gammons.   This is better than the most likely outcome, where Faris is down 4 away Crawford.





2 comments:

  1. Great positions. In position 1, Magriel's safe/bold criteria point pretty clearly to a safe play. In position 2, I think that it's important that your numbers are well-diversified: 1's and 4's hit, 2's and 6's cover, and even a 3 can usefully be played 13/10. Only 5's play poorly. In position 4, hitting looks like the preferred play at gammon go. In position 5, you're not necessarily switching plans to a race because you may still be able to attack; it's just that with this roll you can't leave all those blots because that's the best way for your opponent to get back into the game. I like your analysis of position 7, where I also wanted to play 13/3*. In position 8, I'd recommend rolling out 13/8 10/6 as well, since that's the play that truly leaves the fewest shots. In position 10, I wanted to take, but of course I forgot that when trailing 4a3a one should be quicker to drop low-gammon holding games than for money.

    ReplyDelete
  2. Thanks for your insight Tim. Nice to see you popping up on Stick's forums occasionally as well. I still have several more matches to put up here, plus I'll probably get some more from Cleveland this weekend. Will keep me busy.

    ReplyDelete