Search This Blog

September 25, 2012

Ramezsamir: Noteboom follow up

I got to play Schandorff's move in the ICC 5 minute pool today.  I got a great position out of the opening, but blew it with a tactical blunder.  Instead of my careless 28. g4 ?? which loses, 28. h4 would have locked up the win.




(2142) gammonman (2298) - ramezsamir (2198) [D31]

ICC Internet Chess Club, 25.09.2012

[Bill]

Disconnection will count as a forfeit. 1.d4 e6 2.c4 d5 3.Nc3 c6 4.Nf3 dxc4 5.a4 Bb4 6.e3 b5 7.Bd2 Bb7 8.axb5 Bxc3 9.Bxc3 cxb5 10.b3 a5 11.bxc4 b4 12.Bb2 Nf6 13.c5 Diagram



XABCDEFGHY
8rsn-wqk+-tr(
7+l+-+pzpp'
6-+-+psn-+&
5zp-zP-+-+-%
4-zp-zP-+-+$
3+-+-zPN+-#
2-vL-+-zPPzP"
1tR-+QmKL+R!
xabcdefghy


 
13...0–0 14.Bb5 Bc6 15.Ba4 Qe8
the first non-book move. 15... Bxa4 and 15 ... Qc7 are covered in my previous blog post. 16.0–0 Bxa4 17.Qxa4 Qxa4 18.Rxa4 Nc6 19.Nd2 Nd5 20.Nc4 Rfb8 21.Rfa1 A niggling endgame advantage - just what White is aiming for. 21...Nc3? 22.Bxc3 bxc3+- 23.Nb6 c2 24.Rc1 Nb4 25.Nxa8 Nd3 Tricky. I didn't see that, but White should win anyway. 26.Rc4? I panicked. [26.Rf1 Rxa8 (26...Rb1 27.Ra1+-) 27.Rc4 c1Q 28.Rcxc1 Nxc1 29.Rxc1+-] 26...Rb1? [26...Nxc1 27.Rxc2 Ne2+ 28.Kf1 Nxd4 29.exd4 Rxa8 30.Ke2 equal per the computer] 27.R4xc2 Nxc1 28.g4?? [28.h4!] 28...Nd3+ 29.Kg2 Ne1+ 30.Kg3 Nxc2 31.Nb6 Nb4 32.Kf4 f6 33.e4 e5+ 34.dxe5 Nd3+ 35.Ke3 Nxc5 36.exf6 Rxb6 {Game 1408 (gammonman vs. ramezsamir) gammonman resigns 0–1



 

September 15, 2012

Schandorff's new line against the Noteboom

Schandorff suggests the new line with 13 c5 against the Noteboom in his excellent work Playing 1.d4 The Queen's Gambit (2012).  Below is his analysis of this line, plus a few of my own comments interspersed.




(2) Noteboom - Schandorff's 13 c5!? [D31]

15.09.2012

[Bill]

1.d4 d5 2.c4 e6 3.Nc3 c6 4.Nf3 dxc4 5.a4 Bb4 6.e3 b5 7.Bd2 a5 8.axb5 Bxc3 9.Bxc3 cxb5 10.b3 Bb7 11.bxc4 b4 12.Bb2 Nf6 13.c5!?
Diagram



XABCDEFGHY
8rsn-wqk+-tr(
7+l+-+pzpp'
6-+-+psn-+&
5zp-zP-+-+-%
4-zp-zP-+-+$
3+-+-zPN+-#
2-vL-+-zPPzP"
1tR-+QmKL+R!
xabcdefghy


This is Schandorff's new idea. A positional approach, he hopes to route the bishop to blockade the a4 square in two moves via b5-a4 instead of three moves via d3-c2-a4. If White can succesfully thwart Black's counterplay, he can later turn his attention to the center eventually expanding with f3 and e4 with advantage. [13.Bd3 0–0 14.0–0 Nbd7 15.c5 according to Schandorff, Dreev has tried this strategy in his games. 15...Bc6 16.Bc2 Qc7 17.Ba4 Thus, the genesis of Schandorff's concept. White can potentially get positions like this, but being on move. Not much, but it's something.] 13...0–0 14.Bb5 Bc6 [14...Qc7 15.0–0 Bd5 (15...Ng4 16.g3) 16.Qe2 Nc6 17.Rfc1 Ne4 18.Ne1 good for White per Scherbakov. He wrote a book for Black on the Triangle System (and noteboom) which I don't have.; Note White in these lines plays Rfc1 and Qe2 in the first phase. 14...Qd5 15.0–0 Nc6 16.Qe2 Ne7 17.Rfc1 Rfc8 18.Qf1 Bc6 19.Ne5 Bxb5 20.Qxb5 Rcb8 21.Qa4 Qe4 22.Qb3 Nf5 23.Re1 Qh4 24.e4 Ne7 25.g3 Qh3 26.f3 J. Johansson-Gunajew, Olomouc 2011; 14...Nc6 15.0–0 Qd5 transposes to 14...Qd5] 15.Ba4 Bxa4 [15...Qc7 16.0–0 Nbd7 this would be the same position as in the note to move 13 but with White on move. This hardly looks like the end of the world for Black, but nevertheless is what White is aiming for.] 16.Qxa4 Qd5 [16...Qc7 17.0–0 Nc6 18.Nd2 Nd5 19.Nc4 f5 20.Nd6 Rf6 analysis by Scherbakov with evaluation unclear. Schandorff continues 21.Rfe1 Rd8 22.e4 fxe4 23.Rxe4 Rdf8 24.f3 good for white according to Schandorff] 17.0–0 Nc6 18.Rfd1 Ne4 a mistake since White drives Black back anyway, gaining time [18...Rfc8 19.Nd2 Ne4 20.Nf1 f5 (20...e5 suggested as a better try for black) 21.f3 Nf6 22.Nd2 with a slight advantage for White] 19.Ne1 1 Blockade those pawns 2 Prevent / anticipate Blacks ...e5 break 3 Drive black backward and prepare f3/e4 4 keep in mind the knight maneuver Nd2/c4/d6 (offensive) and Ne1/d3 or Nd2/f1 (defensive) 5 Never play Bxc6 as you need to keep the a4 blockade 19...f5 20.Nd3 Qd7 21.f3 [21.d5! is tactically superior as Schandorff points out 21...exd5 22.f3 Ng5 (22...Nf6 23.Bxf6 gxf6 24.Nf4) 23.Nf4 Qe8 24.Nxd5 Ra7 25.Rd3] 21...Nf6 22.Nf2 Nd5 23.Rd3 Rad8 24.Nh3 Nf6 25.Rad1 Kh8 26.Qb3 Qe8 27.Re1 h6 28.Rdd1 Ra8 29.Qa4 Ne7 30.Ra1 Qxa4 31.Rxa4 Rfd8 32.Nf4 Ra6 33.Rea1 Rda8 34.Nd3 Nc6 35.Kf2 Kg8 36.Ke2 Kf7 37.Kd2 R6a7 38.Kc2 Rb8 39.Kb3 g5 40.Re1 Rd7 41.Kc4 Ke7 42.e4 fxe4 43.fxe4 Kf7 44.Rf1 Kg6 45.Rxf6+ Kxf6 46.d5+ Ke7 47.dxc6 Ra7 48.Be5 Black resigns. Lomineishvili-Melnikova Dresden 2004 Line



 

September 8, 2012

65R-42P-XX

The plays after 65R-31P-XX all seemed so logical I thought the same exact plays would apply to 65R-42P-XX.  Let's see.
 
But first a bit more on 65R-31P-43.  I think the main reason safe is wrong, besides being ugly, is that it doesn't improve the position at all.  Stacking a sixth man on the 6-point, while safe, does nothing else for the position.
 
My plays (same as 65R-42P-XX) are underlined below, and assume correct unless another answers appears in bold.  If bold, then my play is wrnog and the Bold play is correct
 
 
is Player 2

score: 0
pip: 161
                         
Unlimited Game
Jacoby Beaver
                          pip: 156
score: 0

is Player 1
XGID=-b----E-C---eF---b-d-b--A-:0:0:1:00:0:0:3:0:10
on roll, cube action?

eXtreme Gammon Version: 2.02
*************
65R-42P-XX
*************
General thoughts:

Gameplan is still to race, though we are ahead less than after 31P

Opponent’s board is weaker than after 31P  – getting hit not as costly.
Opponent still has a priming game formation.


65. Run 24/13

64. Run 24/14

63. Run 24/15

62. Run 24/16

The more I look at this, with the sixes, probably the main thing is there is simply a lack of playable alternatives.  Regardess of startegy and big picture concerns, running is all you got since the alternatives look so lame.

61. Point 13/7, 8/7

54. Down 13/9, 13/8
 
Run 24/13 is right by .02 so down is a small error.  Apparently you would prefer to Run against 31 too except that it is impossible in that case so two down is the only option.  Is backgammon really so simple that "up in the race, race" is the only factor here?

53. Point 8/3, 6/3

52. Down 13/8, 13/11

51. Split 24/23, 13/8

43. Down 13/9, 13/10.  Safe is too ugly and non-constructive.

42. Point 8/4, 6/4

41. Tower 13/8

32. Down 13/11, 13/10
 
Note Tower 13/8 is essentially tied.  Unstacking and diversifying is highly desriable, yet you also really don't want ot get hit so the decision is close.

31. Point 8/5, 6/5

21. Down 13/10
 
Split 24/23, 13/11 is correct by .015.  Perhaps the blocked 62 is a factor, compared to the position after 65R-31P

66. 24/18, 13/7(3). 

55. 13/3(2)

44. 13/5(2)

33. 13/10(2), 6/3(2)

22. 13/11(2), 6/4(2)

11. 24/22, 6/5(2)

** 8/7(2), 6/5(2) is only off by .01.  The safe ace play is correct but by a smaller margin than after 65R-31P.  Could it be that getting hit, while still a joker, is less costly since opponent's board is weaker?  Having 6-4 points is worse than 6-5 points.  It's still wrong to leave the 7 shot joker, but note that the upside is still there while the downside is lessened a bit by the weaker board - hence the bold play is reasonable.
 
All in all, I'd say the logic behind 65R-31P-XX held up pretty well compared to 65R-42P-XX.  All plays but one (54) were exactly the same, and a couple other plays were really close.

September 6, 2012

Pittsburgh (Roberto)

I was fortunate to have three matches recorded from Pittsburgh Open last weekend. Below are slected errors from one of those matches, against Roberto.

******

First, score is 9 away 9 away and I am being recubed. Take or Drop?

is Roberto

score: 0
pip: 128
                         
9 point match
                          pip: 130
score: 0

is Bill
XGID=--a-C-DaBC--cB----bbbb-b-A:1:-1:-1:00:0:0:0:9:10
on roll, cube action?

Analyzed in XG Roller++ No redouble Redouble/Take
  Player Winning Chances: 70.91% (G:27.78% B:0.46%) 71.32% (G:27.38% B:0.49%)
  Opponent Winning Chances: 29.09% (G:5.99% B:0.27%) 28.68% (G:5.90% B:0.26%)
  Cubeless Equities +0.655 +1.319
Cubeful Equities
     No redouble:+0.888 (-0.112)
     Redouble/Take:+1.098 (+0.098)
xg Redouble/Pass:+1.000
 
Best Cube action: Redouble / Pass
eXtreme Gammon Version: 2.02, MET: Kazaross XG2Roberto

I took and it turns out to be a pretty good sized pass.

Let's look at the basics.  I am down a bit in the race, outboarded and on the bar against a four point board.  His 'threat' is simply that I dance a couple rolls, or I come in and he sucessfully attacks me or simply rolls home clean.  There are some gammons in the mix too.

On the positive side, Roberto still has 1.5 men back so he still has escape work to do.  My offensive strucure is decent, so if I come in right away I may be able to counterattack or counterprime.  Also, of his ten in the zone, 8 are tied to made points and the other 2 are not diversified to attack.

PRAT would indicate a pass - after all I am behind in the race, have the worse position (outboarded) and I am on the bar.  So I am behind on all phases of the game.  Yet due to the above, I have reasonable winning chances.  The kicker seems to be the gammons.  I simply get G-balled too often to justify a take.

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

Later in the same game. 22, a great shot. How to play it?

is Roberto

score: 0
pip: 108
                         
9 point match
                          pip: 138
score: 0

is Bill
XGID=-a-AC-D-BA--bB----bcbc-bB-:2:1:1:22:0:0:0:9:10
to play 22

1.XG Roller++24/22(2) 13/11(2)eq: -0.424
Player:
Opponent:
30.52% (G:4.38% B:0.12%)
69.48% (G:20.79% B:0.57%)
2.XG Roller++24/22(2) 9/7 3/1* eq: -0.486 (-0.062)
Player:
Opponent:
28.22% (G:3.81% B:0.08%)
71.78% (G:20.77% B:0.52%)
3.3-ply24/22(2) 6/4 3/1* eq: -0.512 (-0.088)
Player:
Opponent:
26.86% (G:3.31% B:0.10%)
73.14% (G:19.63% B:0.59%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

24/22 is for sure, what about the other two?

I went with 9/7, 3/1* trying to slow him down by putting him on the bar.

This looks like the wrong idea.  For starters I don't really want the ace point, which never be part of a prime with the 8-point.

Also, compared to the superior play, 13/11(2) my play leaves more shots.  The computer play leaves any 2 plus 44 for 13 hitters.  With my play, any ace hits from the bar as well as 52 and 43 for 15 hitters.

So my play takes more risk for less upside.  The computer play, when missed, leaves me with a better more compact structure which can improve easier - in short more upside.

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

Things have turned my way.  I am on roll, holding a 4 cube, score is 9 away 9 away.  Cube action?

is Roberto

score: 0
pip: 59
                         
9 point match
                          pip: 160
score: 0

is Bill
XGID=a--BBBBA-----A--A--cAbBdA-:2:1:1:00:0:0:0:9:10
on roll, cube action?

Analyzed in XG Roller+ No redouble Redouble/Take
  Player Winning Chances: 75.98% (G:0.00% B:0.00%) 74.47% (G:0.00% B:0.00%)
  Opponent Winning Chances: 24.02% (G:10.15% B:0.87%) 25.53% (G:11.32% B:1.19%)
  Cubeless Equities +0.419 +0.936
Cubeful Equities
     No redouble:+0.869 (-0.045)
xg Redouble/Take:+0.914
     Redouble/Pass:+1.000 (+0.086)
 
Best Cube action: Redouble / Take
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

I wasn't sure and so held back.  Classic mistake per Wollsey's law.  When unsure if opponent has a take or drop, then double as he might make a mistake!  Simple.

I wanted to make the 5-prime first, as Roberto has 5 men off, which obviously gives him a certain reservoir of counterplay.  I still have many men back (six on his side of the board still) so it seems he will have many many rolls to fight with into the future.

But, that is only part of the story.  First, I am threatening to make a 5-prime with any six with half the rolls (any six plus 55).  This is a real threat.  Besides this, are some 'hidden' threats.  Once I make the 5-prime there is a decent chance that Roberto will enter and crack, allowing me to pick up a second checker and basically put the game away.  This is the powerful 'threat beneath the surface' that I knew was there but didn't fully appreciate.

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

Fortunately for me I didn't double since I made the 5-prime and Roberto immediately jumped it with 26 from the bar!!  OK, so how do I play this lemon 31 roll?

is Roberto

score: 0
pip: 51
                         
9 point match
                          pip: 154
score: 0

is Bill
XGID=---BBBBBa-------A--cAbBdA-:2:1:1:31:0:0:0:9:10
to play 31

1.XG Roller++20/16eq: +0.078
Player:
Opponent:
53.92% (G:0.00% B:0.00%)
46.08% (G:31.95% B:1.16%)
2.XG Roller++20/17 16/15eq: +0.064 (-0.014)
Player:
Opponent:
53.31% (G:0.00% B:0.00%)
46.69% (G:33.07% B:2.24%)
3.XG Roller++20/17 7/6eq: -0.025 (-0.103)
Player:
Opponent:
50.59% (G:0.00% B:0.00%)
49.41% (G:35.59% B:2.72%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2

I went for maximum coverage with 20/17, 7/6.  A large mistake.

A superior execution of my idea would have been 20/17, 16/15.  I am guessing there are the same number of 7's and 8's so this play keeps maximum defensive coverage while holding hte strength of my position - the 5-prime.

The best play was to button up with 20/16.  The main viture of this play is that it gets gammoned the least often.  Most likely Roberto comes around the board with his lone man and I get a single or double direct shot anyway.  So perhaps the put'em where you want em is overkill and just too risky?  Not sure really.

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

Here is one last bonehead play.  I lead 4 away post-Crawford and have boxes to play.

is Roberto

score: 5
pip: 165
                         
9 point match
                          pip: 114
score: 8

is Bill
XGID=aB-B-dDBA-B-----bb-bBbb---:1:1:1:66:8:5:0:9:10
to play 66

1.XG Roller++20/8(2)eq: +0.655
Player:
Opponent:
84.49% (G:43.00% B:0.23%)
15.51% (G:1.66% B:0.06%)
2.XG Roller++20/14(2) 10/4(2)eq: +0.526 (-0.129)
Player:
Opponent:
79.18% (G:47.80% B:0.45%)
20.82% (G:2.74% B:0.11%)
3.3-ply20/8 10/4(2)eq: +0.544 (-0.111)
Player:
Opponent:
79.73% (G:50.55% B:0.81%)
20.27% (G:2.42% B:0.07%)
eXtreme Gammon Version: 2.02, MET: Kazaross XG2
I moved 20/14(2), 10/4(2).  This play has the merit of flushing him forward, by taking away the four point and also wins more gammons.  Of course, gammons are worthless here.

Roberto has a strong offense so I really don't want to get hit.  Why give him the 53 and 26 hitting jokers if I don't have to?

Best is simply 20/8(2) which moves a lot of freight while leaving no shots.  I should just try to roll home clean and win the match.

September 1, 2012

Psycho-Cowboy

At the Pittsburgh Open. I got skunked 10-0 in the first round. I got Gammoned followed by Backgammoned. So I go back to the room and relax with a game of chess. And what happens? I end up crushing a Grandmaster. Go Figure. Here is the game. Note I am higher rated than GM Simon Williams here (psycho-cowboy) which is stone cold ridiculous. Enjoy.