GM Joel on the Question He Always Wanted
By GM Joel Benjamin   
February 2, 2010
Joellead.jpg
GM Joel Benjamin, Photo Betsy Dynako, May 2009
All good things come to an end.  And so it will be with Ask GM Joel.  Instead, I will be "checking in" periodically with topical blogs on subjects like the U.S. Championship, National Scholastics, U.S. Chess League, and whatever else comes up.

In the beginning, I was excited about the prospects for a Q&A column.  I wrote one for a few years in the nineties for Chess Life (Theoretically Speaking), and that seemed to produce an awful lot of interesting queries.

This time around, an awful lot of questions weren't appropriate for the column.  They were requests for information that should have been directed elsewhere within the USCF.  Occasionally I sent replies if I knew an answer and it would take very little time, or I felt it particularly important to do so, but most of these messages went unanswered.  Please note that I am not a full-time employee of the USCF; the column is all I do for the federation, but only a fraction of my overall workload.  Some other questions called for my professional expertise...and personal replies (free coaching?) as well.  So a lot of people just didn't seem to get the purpose of the Ask GMJoel e-mail address.

Even among the legitimate questions, I felt there wasn't enough variation.  More than half of the questions were forms of the same basic question:  How do I get better?  To illustrate, let me present a composite (not an actual) question:

I had a relatively easy time getting from 1200-1500, but now I've been stuck there for a few years and I'm wondering what I need to work on to get my rating up to 1800?


The biggest problem with this question...like a doctor who is asked to diagnose without meeting a patient, I don't really know what you need to study/learn just from knowing your rating or one or two other facts about you.  I answered a few questions by giving common sense advice about what to study, but after awhile there just wasn't a lot left to add. 

I had hoped to receive more "hard" chess questions on game analysis or opening theory, but I suppose those just come less frequently in the computer era.  If you can run a position on a chess engine, you won't feel a need to ask a grandmaster.  [In fact, I would check any position on a chess engine, too, or else readers would surely find some errors.]  I did occasionally get some fine questions on opening theory, but too many questions were of this form:  "Is the [insert your favorite opening here] a good opening?"   There isn't much I can do with that one.

I would like to thank all of those readers who contributed frequently and followed my answers on a regular basis.  Let me close the column by presenting an example of a clever, thought-provoking question that I didn't actually receive.  But if I did, I would have had a good story to tell.

Have you ever written intentionally faulty analysis or held back something in order to catch someone in a trap in a future game?

Excellent question!  I wish we had more contributors like you.

I would never publish something intentionally dishonest, but last year I did choose to avoid discussing certain details of a game for selfish reasons. 

I wrote a report on the 2009 World Open for New In Chess magazine.  NIC articles are somewhat lengthy, and offer the author the opportunity to cover a lot of topics.  I wrote a bit about my tournament, especially since I had a very interesting game against Gata Kamsky to annotate (a near victory which ended in a draw).  In the interest of completing the narrative, I thought I should mention how my promising tournament ended just two rounds later, after a marathon draw against Ray Robson and a crushing defeat at the hands of UTD student Jacek Stopa.

4149

Jacek Stopa-Joel Benjamin [C41]
World Open Philadelphia USA (7), 2009

1.e4 e5 2.Nf3 d6 3.d4 exd4 4.Nxd4 Nf6 5.Nc3 Be7 6.Bf4 0-0 7.Qd2 d5 8.Ndb5 Bb4 9.0-0-0 c6 10.Nc7 Nxe4 11.Qd4 Bxc3 12.bxc3 Qe7 13.Kb2 g5 

ASKGMJOELg5.jpg
14.Bg3!?
This move came as a real surprise.  I had analyzed this position at home (okay, my computer analyzed and I watched), and concluded that White had nothing better than taking the rook in the corner, with an unclear position.  I fell victim to a common problem when analyzing with a computer; since 14.Bg3 is actually unsound, it never showed up on my screen.  I never thought to check what would happen if White tried this quite forcing move.
14...Nxg3
Black has no choice but to grab the piece, as otherwise 14...g5 gashed my kingside for no reason.
15.hxg3 Qxc7 16.Qf6 Qd8??
Honestly, I took a lot of time trying to find a defense for Black.  The right idea just never clicked in my brain.
17.Qh6 Bf5 18.Bd3 Qd7 19.Qxg5+ Bg6 20.Rxh7 Kxh7 21.Qf6 1-0
after16qf6ask.jpg

As soon as I resigned, Stopa informed me that he had blundered in the piece sacrifice, overlooking a simple defense.  When I saw what I could have done, I wanted to crawl in a hole and hide.  How many players had walked by the board wondering why I didn't simply refute the piece sacrifice?

The game actually makes a pretty good story, and normally I would have written it up for my article.  But I was so steamed about blowing this opportunity (5 out of 7 would have left me in decent position for a top prize), I wanted to see if I could someday obtain some compensation for this humiliation.

Enter Alec Getz, my opponent in the fifth round of the New York State Championship.  I was not having a great time to this point, having lost to a player I've beaten about fifty times before, Jay Bonin.  The penultimate round at open tournaments often produces dodgy preparation.  It's the "checkout" (of the hotel) round.  So my preparation consisted of transferring all my stuff from my room to my car.  Presumably Alec's father took care of business, allowing him time to search the computer a bit, but not time to evaluate the data very critically.  [Unless he anticipated the pairing the night before.  I can only say that I didn't.]

In these frenetic circumstances, I wasn't thrilled about facing a young talent who had already dispatched Lenderman earlier in the tournament.

Alec contributed fairly lengthy annotations of this game to CLO, so I will incorporate his work into my commentary.  

4150

Alec Getz-Joel Benjamin [C41]
New York State Champs (5), 07.09.2009

1.e4 e5 2.Nf3 d6
I just played a hunch here.  Either he wouldn't know much about this opening, or he would be following my game against Stopa.
3.d4 exd4 4.Nxd4 Nf6 5.Nc3 Be7 6.Bf4
This is the most challenging line, but I was now certain he was following the afore-mentioned game.  Alec figured I might well have an improvement, but was willing to deal with it over the board.
6...0-0 7.Qd2 d5
after7d5.jpg
Now we are in a full-fledged poker game.  I thought the most likely situation was that Getz was blindly following my game with no idea that Stopa's play was unsound.  It was possible, however, that he knew more about the game.  Shortly after the World Open Dennis Monokroussos recorded a video about my game.  I didn't watch it (videos sometimes crash my computer) but I figured Dennis spilled the beans about my trap and maybe suggested some improvements.
Black has other moves here-7...Nc6, 7...a6, and 7...c6 are all playable-but certainly more pleasant for White.  So I decided to play the percentages.  I also took a little extra time to let Getz think that I was an idiot just realizing now that my opening doesn't work and my opponent might know why.  Getz:  "So far we're following Stopa - Benjamin. Benjamin spent quite some time deciding on 7...d5 and I optimistically considered the thought that perhaps Benjamin didn't find an improvement over that Stopa game and wasn't sure if he should risk going into the line again."
8.Ndb5 Bb4 9.0-0-0 c6
There seem to be a lot of possibilities but practice indicates that this position is virtually forced after 7...d5.
10.Nc7 Nxe4 11.Qd4
Getz played this move quickly, indicating he was following Stopa-Benjamin.  Otherwise he would have considered the possibly superior 11.Qe3.  Now I banged out the next few moves to demoralize my young opponent.  Getz naturally became suspicious, but this is a difficult time to work out complications over the board. 
11...Bxc3 12.bxc3 Qe7 13.Kb2
Alec wondered about the alternate defense to the threat of 13...Qa3+ 14.Kb1 Nc3+, namely 13.Rd3.  It is true that 13...g5 14. Bg3 Nxg3 15.Rxg3! would be out of the question, but Black has a good solution:  13...Be6 14.Nxa8 c5 15.Qe3 c4 16.Rd4 Qa3+ 17. Kd1 Nxc3+ 18.Kd2 Nb1+ 19.Kd1.  Here Black took the perpetual with 19...Nc3+, in (Nijboer-Haslinger, Essent Open 2008, but 19...Nc6 20.Nc7 Nxd4 21.Qxd4 Nc3+ 22.Kd2 Ne4+ 23.Ke1 Qa5+ 24.c3 Qxa2 looks very promising for Black.
13...g5 14.Bg3 Nxg3 15.hxg3 Qxc7 16.Qf6
after16qf6ask.jpg
16...Qb6+
All is becoming clear to Mr. Getz.  This is what Stopa told me after the game.
17.Ka1 c5 18.Qxg5+ Qg6 19.Rxd5
Getz thought for a long while and decided he didn't have enough compensation for the piece, but it was the best he could do.  But now I was kicking myself, because I hadn't worked out exactly how to continue from here.  While still in Philadelphia I confirmed that Black is indeed very well off here, but the position is still very complicated and I could easily go wrong.
19...Qxg5
I felt like 19...Be6 should be right, but I realized White can force a draw with 20.Rh6 Qg7 21.Bd3 Re8 22.Bxh7+ Kf8 23.Qe3! Bxd5 24.Qxc5+ Re7 25.Qc8+.   Black can try other moves but it seems that trading queens is the strongest.
20.Rxg5+ Kh8 21.Bd3
My Rybka seems to prefer 21.Rgh5 Kg7 22.Rxh7+, but the humans involved prefer Alec's choice.
21...f5 22.g4 Nd7 23.gxf5 Nf6
Although determined to fight, Alec wasn't optimistic about his chances.  My computer is also telling me good things here.  But I wasn't happy at all.  White's pawns offer substantial counterplay, and he doesn't even have to hurry to push them.
24.g4 Bd7 25.Rg6 Rac8
I wanted to provoke c3-c4 to make sure his bishop didn't land on that square later.  Getz made the right call:  26.g5 c4! 27. Be2 (27. gxf6 cxd3 28. Rg7 Bxf5) and now 27..Ng8 is probably stronger than Getz's 27...Nd5 because the rook on g6 does not have the h6 square to run to.
26.c4 Rf7 27.f3!
White avoids the premature advance 27.g5? Ng4.  The silly computer wants me to play 27...Bc6 28.Rh3 Bxf3!? 29.Rxf6 Rxf6 30.Rxf3, but I would be very surprised to actually win that position.
27...Ng8
Alec could sense my growing anxiety.  I was facing a real possibility that I would suffer another failure in this opening.
28.f6
Alec gives this move a "former oyster" (exclam), and indeed, it isn't easy to counter White's activity.  However, after the sedate 28.Be4, how am I supposed to win this position?
28...Bc6
 I grasped that 28...Nxf6 29. Rxf6 Rxf6 30. Rxh7+ Kg8 31. Rxd7 wouldn't win for me, but I intended 28...Rxf6, noting that there the discoveries don't hurt me after 29.Rxh7+ Kxh7.  However, I noticed at the last moment that 29. Rxf6 Nxf6 30. g5 would hurt, whereupon Black has to fight for a draw.
29.g5 Re8
I had to avoid 29...Bxf3 30.Rg7! Rxg7 31.fxg7+ Kxg7 32.Rxh7+ Kg8 33.g6 Ke8 (33...Bc6 34.Bf5 Rd8 35.Be6 Rd1+ 36.Kb2 Rg1 37.Rf7+ Ke8 38.g7 with a draw in the offing) 34.Bf5 Rd8 35.Be6 Nf6 36.Rf7 Bh5 37.Rxf6 Ke7 38.g7 Kxg6 39.g8 (Q) Rxg8 40.Bxg8 with a dead draw.  Even if Black wins the c4-pawn, there's no way to win the bishop ending. after29.Re8askgm.jpg

Now White has to liquidate, because the rook on g6 has no way to back out.
30.Rg7 Rxg7 31.fxg7+ Kxg7 32.Rxh7+ Kf8 33.f4 Re7 34.Rh8 Be4 35.f5 Bxd3 36.cxd3 Re5
askgmjoelafterre5.jpg
37.Rh7 Rxf5 38.Rxb7 Rxg5 39.Rxa7 Ne7
Getz thought I was winning here.  I wasn't sure, but I felt a heck of a lot better than before all the simplification.
40.Rc7 Ke8 41.a4 Kd8 42.Rb7 Nc6 43.Kb2 Kc8 44.Rh7 Re5 45.Kc1 Nd4 46.Kb1 Re1+ 47.Kb2 Re2+ 48.Kb1 Nb3 49.Rg7
askgmjoel49rg7.jpg
49...Rd2?
Both players were tired by now, and the subtleties were not easy to grasp in the sudden death time control.  This move is actually a substantial error because in some tactics Black needs this square for the knight.
 50.Rh7
Getz correctly concluded that 50.Rg3 Kc7 was a slow but sure death, but he didn't see the hidden tactic 50.a5!  Now 50...Nxa5 51.Rg5 Nb3 52.d4! actually forces a draw (52...cxd4 53.Rb5 or 52...Rxd4 53.Kc2).  R+N v R offers some practical chances but should certainly be drawn.  So Black should try 50...Kb8 51.a6 Ka8 but it will be a tightrope to eat White's pawns without losing my one and only.
50...Rxd3 51.Kc2 Re3 52.Rg7 Nd4+ 53.Kb2 Rb3+ 54.Ka2 Rb4 0-1

So in the end I got my slightly sneaky win.  The moral of the story is to beware of preparing with your finger.  That's when you keep pressing the cursor button until you get to the end of the game, instead of analyzing the moves critically.  When you see a game end in brutal fashion, it's prudent to turn on a chess engine to make sure the game ended like it was supposed to.