50 Years of Chess: Year 37

After my all-time best game, my win over IM David Pruess in 2006, I was invited by IM (and soon to be grandmaster) Jesse Kraai to record a lecture about the game for ChessLecture.com. The lecture instantly hit their top-ten list, and it was so popular that I was invited to be a regular commentator on the site, even though I did not have nearly the credentials of the other lecturers. While the other lecturers talked about grandmaster games and gave ultra-refined opening lectures, no one else was lecturing about the kind of games ordinary tournament players play, and the problems they face. That was the niche that I tried to fill.

For me, the number one perk of lecturing was that I also got a free lifetime membership at ChessLecture, and the number one perk of that was getting to hear Jesse Kraai’s lectures. He was at the site from 2005 to 2011, and his videos were the absolute epitome of the art. I have not heard anyone before or since, whether on YouTube or chess.com or anywhere else, whose lectures can hold a candle to Jesse’s. He had such a calm, rational, philosophical approach to the game that he made it sound easy, even when it was not. And his terminology! I still remember concepts like the “angry bishop” (a French Defense bishop after it has been released from its prison) or the “two bishops versus the Odd Couple” (bishop and knight).

In 2008 Jesse and I played at the U.S. Championship Qualifier in Tulsa. I believe, if I’m not mistaken, that 2008 was the only year in history when ordinary tournament players were given an opportunity to qualify for the national championship. I suppose it has also been possible to qualify by winning certain big tournaments like the World Open, but this was the only time that there was a tournament designated solely as a championship qualifier. The top eleven people got to play in the U.S. Championship that year.

Guess who was the winner? You’ve got it: Jesse Kraai, with 5.5 points out of 7. (He tied with Julio Becerra, Alex Yermolinsky, Dean Ippolito and John Fedorowicz.) Six other players also qualified, including Bay Area stars Sam Shankland and David Pruess.

As for me, let’s just say I was a long, long, long way from qualifying. I ended with a score of 2.5 out of 7. It was a pretty disappointing result, and my rating dropped from 2114 to 2101. But there was a silver lining. Jesse (who by then was a grandmaster) offered to analyze one of my games on ChessLecture. I sent him a game I lost in the second round against a FIDE Master from New Orleans, John Bick.

One condition was that I had to send Jesse a full annotation of the game that I did myself, without computer help or help from anyone else. That’s where I received my first, unexpected lesson. I sent him two handwritten pages of notes, densely packed (which would probably have been four pages typewritten, at least). He e-mailed me: “Is that all?”

I was floored. I had always thought that I was a crazy eccentric for keeping a binder full of my games, all of them carefully annotated. But Jesse makes me look like a complete slacker. Every single game gets at least ten pages of annotation. This is how you become a grandmaster.

His ChessLecture came out a couple weeks later, and I have to say it was merciless. Several subscribers (including some of my fans) were shocked at how harshly he criticized some of my moves. But that’s what you want from a chess coach, a sensei. You don’t want them to be polite and hold back.

And I have to say, in retrospect, his criticism overall was spot on. Spot. On. He identified my main weakness as “insensitivity to dynamic imbalances,” which he said was a common weakness for people in the 2000-2200 rating range (and even higher). It takes some unpacking and some examples to see what this really means, but basically it means asking simple questions about the position. When you make a particular move, what does it change about the position, plus or minus? Take my favorite variation against the Sicilian, 1. e4 c5 2. f4. What does the second move do? Positive: it increases my control over e5. It gains space on the kingside and it can be a prelude to a pawn storm. Negative: It doesn’t help me develop, it weakens my king position, it blocks my bishop at c1. The negatives outweigh the positives. So perhaps one shouldn’t play this move. But if one insists, then one should at least be sensitive to those imbalances and play in a way that minimizes the harmful ones.

It is exactly this approach, learning to think about a position by asking simple questions about it, that I learned over the next few years at Mike Splane’s chess parties. And it was no accident that my rating went up: from 2101 that year to 2203 just seven years later. And that’s at a time in life (my fifties) when most players are getting weaker. If I had been younger, or worked at it harder, perhaps I could have improved more. So Jesse was right: sensitivity to dynamic imbalances was exactly what my game needed.

So let’s take a look at my game with John Bick and see how this big-picture lesson looks in a specific case. I will juxtapose my comments with Jesse’s (paraphrased). Warning: This will be a long post! As Jesse says: ten pages minimum. If you haven’t done that much, you haven’t really analyzed the game. Also, let me mention that Jesse’s comments here are (mostly) not exact quotes; they are paraphrases of the comments in his lecture, because I didn’t write down the exact quotes.

John Bick — Dana Mackenzie, 3/29/2008

1. d4 Nf6 2. c4 d5

DANA: One of my favorite defenses, partly because it’s supposed to be bad. I maintain that it’s “bad” in the same sense that the Caro-Kann is bad, i.e.: not bad at all. In some lines Black has to bend a little bit, but his position is very resilient. More to the point: Black gives up some space in the center, but hopes to compensate for it with active piece play. It’s not easy to find defenses to 1. d4 that give Black open lines and active pieces.

JESSE: We have to ask, are we playing strange openings because we’re getting beaten in traditional openings where we have to be sensitive to dynamic imbalances? If that is the case, we are cheating ourselves. I saw this in my own chess for a while, when I used to play … e6 and … b6 to avoid traditional openings.

MY RESPONSE: In part, the answer is yes. I used to play the Nimzo-Indian against 1. d4 but there were too many things I didn’t like about it: the Catalan, the early 5. Ne2 lines, and transpositions into the Queen’s Gambit Declined after 1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3. Too many problems to solve them all. Still, the Nimzo is a gold-standard traditional opening. The problems should have solutions. I just didn’t want to put in the work to find them.

3. cd Nxd5 4. Nf3 Bf5 5. Qb3 e6 6. a3 Nbd7 7. Qxb7 …

FEN: r2qkb1r/pQpn1ppp/4p3/3n1b2/3P4/P4N2/1P2PPPP/RNB1KB1R b KQkq – 0 7

DANA: I was excited to finally play Fritz’s home prep, 7. … Bxb1 and 8. … Bxa3! Unfortunately, I didn’t have a clear idea of what to do next. In general, an important point about this variation (the Marshall Variation with 2. … d5) is that White’s Qxb7 is seldom a real threat.

7. … Bxb1 8. Rxb1 Bxa3!

DANA: Winning back the pawn, since 9. ba? would be met by 9. … Rb8. If 9. Bd2 Rb8 10. Qxa7 Bxb2 is great for Black, with the threat of … Nc3.

JESSE (his voice dripping with scorn): This is obviously a computer move. Humans don’t play moves like this.

In general, Jesse was very critical of my approach to openings, leaning on memorized computer lines rather than thinking about the dynamical imbalances in the position.

9. Qc6 …

Here was the first point where Jesse talked about the dynamic imbalances in the position.

JESSE: Black is clearly ahead in development, but there are three compensating factors for White. He has the better pawn structure, with a strong central pawn on d4. Black has a weak pawn on c7. And White has the two bishops. Black’s position is probably all right, but he has to be sensitive to these imbalances. That means, in particular, that he should strive for the pawn break … c7-c5, because that will rid him of his biggest weakness and also challenge White’s control over the center.

9. … Bb4+ 10. Nd2?! …

DANA: I was surprised by 10. Nd2, but White’s idea is to take over the long diagonal with his unopposed white-squared bishop. But can he really afford such a casual approach to development? I think not!

JESSE: This move led to Dana misevaluating the position, thinking that he was better or winning when he in fact was not.

10. … O-O 11. e3 …

FEN: r2q1rk1/p1pn1ppp/2Q1p3/3n4/1b1P4/4P3/1P1N1PPP/1RB1KB1R b K – 0 11

11. … N7f6?

DANA: “…”

JESSE: This is Dana’s biggest mistake of the game, and he doesn’t even say one word about it in his analysis. In this position, time is Black’s greatest advantage. But 11. … N7f6 is not a developing move. It’s not clear at all that the knight is better on f6; later on, Black will wish it were on d7 to support the … c7-c5 break. Black should start with 11. … Rb8, preparing to play … Rb6 and … c5. Perhaps Dana was worried about 12. Bd3 Rb6 13. Qc2, when the queen and bishop attack the h-pawn. But in a position like this, when you sense that White is down in time, you shouldn’t worry about the threat of Bxh7+. He should take advantage of the disorganization of White’s pieces with 13. … c5 or possibly even 13. … e5. The latter move might be more in Dana’s style.

MY RESPONSE: A fascinating comment for many reasons. First, I completely agree. 11. … N7f6 is a knee-jerk “developing move” that was played without considering what I am actually trying to accomplish in the position. And the fact that I didn’t even see it as being worth commenting on makes my failure even worse!

A second fascinating thing is how little Jesse relies on calculation of variations. He goes all of two moves deep here, maybe three, and he dismisses Bxh7+ on general principles without even analyzing any variations.

I actually suspect that Jesse does a great deal more calculation that he lets on. I have seen some ChessLectures by him where he would say that X was a “simple move,” when in fact you had to go several moves deep and navigate some tricky tactics to be sure of that. But he has great faith in his intuitive evaluation of the dynamic imbalances, and he only goes into a deep tactical calculation when his intuition tells him that there must be a tactical resource.

It’s also interesting that Jesse thought my main concern after 11. … Rb8 was 12. Bd3. I actually think my bigger worry was 12. Bb5, and that may be the reason I thought it necessary to avoid trouble by moving the knight away from d7 first. However, it turns out that 11. … Rb8 12. Bb5 is quite bad — but you need Jesse’s archenemy, the computer, to see why! Fritz finds the absolutely stunning move 12. … Qg5! It’s pretty clear that White can’t afford to take the knight because of 13. Qxd7 Qxg2, but what’s really amazing is that after 13. O-O Black simply picks off a pawn with 13. … Nxe3! This move discovers an attack by Black’s queen on the b5 bishop.

I don’t know if I would have found this over the board, but it IS a great example of the kind of active piece play Black is striving for in this opening.

12. Bd3? …

Jesse correctly criticizes Bick’s move because it’s a little bit “loose” — the bishop is undefended. Loose pieces drop off!

12. … Be7?!

DANA: This move was a crime against chess. First, from a strategic point of view, it removes an already well-developed piece from a good, active square — all for the chimera of a knight fork on b4. Second, the position demanded a deep look, because the critical moment has already arrived. 12. … e5! is a tremendous shot. If 13. de Nxe3! now Black is overwhelming White! If 14. ef Qxd3 with huge threats like … Nc2+ or … Qxb1 or … Bxd2+. Instead [of 13. de] White must play 13. O-O, but now on 13. … ed 14. ed Nf4 it’s already hard for White to hold his d-pawn. Wow! To think that I walked by a gold mine to pick up a chunk of fool’s gold.

JESSE: There’s an objectivity problem here. Dana is still under the illusion that Black should be better. After 12. … e5 13. de Nxe3 14. Be2 Black has two pieces hanging, and he does not get enough compensation.

MY RESPONSE: Actually, I think that Jesse has an objectivity problem here! It’s not true that White is winning a piece. After 12. … e5 13. d3 Nxe3 14. Be2 Black has 14. … Nf5! (a computer find, but nevertheless a legal move). The main point is that 15. ef? runs into 15. … Nd4 16. Qc4 Re8, and Black wins. More likely White would play 15. O-O, but after 15. … Bxd2 16. Rd1 Nd4 17. Qc4 Re8, I certainly wouldn’t mind playing this position and the computer calls it equal. Black’s active piece play has definitely triumphed over his pawn weaknesses up to this point.

This was the one point in the game where I found Jesse’s critique disappointing. I thought that 12. … e5! and 13. … Nxe3! were a great find, and I was expecting him to lavish praise on those moves. I felt aggrieved that he dismissed my idea so quickly. But let’s dig a little bit deeper. From the practical point of view, Jesse wins the argument, because I did not see the idea of 12. … e5 and 13. … Nxe3 during the game, when it really mattered.

This point is really important, and it gets to the question of why we even bother analyzing our own games. It just occurred to me, as I was writing the last paragraph, that the main purpose of analyzing your own games is NOT to determine the “objective truth” about a position. Finding the “objective truth” is only one step toward the real goal of analysis, which is to discover the subjective truth of what were the flaws in your thinking process. What were the ways in which you could have thought better in order to obtain a better outcome? You are not done with your analysis until you have answered this question.

That is why it’s so important to analyze your game first without the computer. The computer might tell you the objective truth (though we can debate that), but it for sure does not tell you the subjective truth about a position. A computer variation that you would never have thought of during a game is essentially irrelevant. A good example is that variation I just gave in my note to Black’s 11th move: 11. … Rb8 12. Bb5 Qg5! 13. O-O Nxe3! Objectively that just busts White’s 12. Bb5, or at least it wins a free pawn. Subjectively it’s irrelevant because “humans don’t play moves like this.”

When you analyze the game without the computer, using only your brain, you are at least getting closer to the subjective truth. In home analysis, I found 12. … e5! and 13. … Nxe3!, which means that I was capable of finding them over the board. Then the question becomes: Why didn’t I? And what can I do in the future that would enable me to play those moves?

Now, after all this philosophical meta-chess talk, we finally get to an important chess point. (Reader, I appreciate your patience!) I think that there is one question that would have led me straight to this variation. It ties in perfectly with Jesse’s theme of dynamic imbalances. It’s a question that is not on the list of eight to ten strategic questions that were developed at Mike Splane’s chess parties, but I think it should be on the list. The question is this:

Are there any time-sensitive imbalances in this position?

A time-sensitive imbalance is one that comes with an expiration date: Use it or lose it. Examples of such imbalances are: Undefended pieces. (Because the opponent will probably defend it or move it very soon.) Uncastled kings. And pins. (Once the pin is broken, it usually loses all of its significance.) Notice that in this position we have all three of those factors. A loose bishop on d3, a pinned knight on d2 (both on the d-file!) and an uncastled king. Also, we have to acknowledge that Black has invested a great deal in these time-sensitive imbalances. With 2. … d5 I gave away some central control. When I allowed 7. Qxb7, I gave away some pawn structure. I did these things in order to gain time. Therefore, we can conclude that if I allow my time-sensitive advantages to expire, I will most likely be in an inferior position. So right now, before White castles, I need to look very hard for a way to capitalize on them.

Obviously, with the move 12. … Be7, I was attempting to capitalize on the loose bishop on d3. However, in the process I gave away my second time-sensitive advantage, the pin on d2. While it’s attractive to threaten the knight fork … Nb4, the objective fact is that this threat is easily defended.

But is there some other way to harass that bishop, one that doesn’t require me to give away another advantage? Our eyes have to gravitate to that d-file, which has two ultra-sensitive targets on it, and ask whether there is some way to blast it open. That leads to my first move, 12. … e5, a move that also happens to contribute to my long-term strategic goal of weakening White’s d-pawn. The only problem is that it seems to give away a pawn, but lo! After 13. de that d-file is one step closer to opening. Now if I just move my knight out of the way, I will really have some nasty threats. And that suggests the idea of 13. … Nxe3, and after that the pieces really start to dance. You get this chain reaction where every time White defends one threat, two new threats emerge in its place. White plays 14. Be2 and now after 14. … Nf5 White has a problem on d2, a problem on d4, a problem on e2 and a problem on the e-file in general.

13. Qc4 Nb4 14. Be2 Rc8

DANA: I had to at least play 14. … c5! 15. dc Qa5! I looked a long time at this. Note that 16. Nb3?! Nd3+! 17. Kd1 Nf2+ is extremely risky for White. The line I was actually worried by was 16. O-O Qxc5 17. Qxc5 Bxc5 18. Nb3, because I thought I was “losing my advantage.” This was insane! Even if it’s true, this is still much better than the line I played, where I not only lost my advantage but had a significant disadvantage.

JESSE: Dana’s evaluation is still not objective. He thought that 14. … c5 was equal and tried to play for more. But in reality he never had an advantage, and in the line he gives, Black has a significant disadvantage because of White’s two bishops.

15. Nb3 …

JESSE: Establishing control over c5.

15. … Qd5

JESSE: Ordinarily you don’t want to trade queens when you’re at a disadvantage, but it’s hard to see what else Black can do here.

FEN: 2r2rk1/p1p1bppp/4pn2/3q4/1nQP4/1N2P3/1P2BPPP/1RB1K2R w K – 0 16

16. Qxd5 ed?

DANA: I already didn’t like my position. If I take with one of the knights, I though the weak pawn on a7 would eventually fall, as he will put his bishop on the long diagonal, play e4 to chase my knight away, and then just have a field day attacking all the weak pawns on my queenside. With 16. … ed I was hoping to keep his bishops somewhat bottled up.

JESSE: This move is really bad. Dana is not sensitive enough to pawn weaknesses. In this position, with the two bishops against the “Odd Couple,” Black shouldn’t be able to survive.

MY RESPONSE: I sadly have to agree, though I would phrase it differently. I was sensitive to pawn weaknesses but in a completely illogical way. After, say, 16. … Nbxd5, I was sensitive to the weakness of my a-pawn, as if the knight move had anything to do with that. But it doesn’t! That pawn is weak, period. The real issue is whether Black wants a position with one weakness, after 16. … Nbxd5, or two weaknesses, after 16. … ed. After the latter move, Black is stuck with two weaknesses, because either the c-pawn is going to be weak (backward pawn on an open file) or else I will have to play … c5 and then the d-pawn will be weak (isolated pawn on an open file).

For someone who used to be a mathematician, it’s mortifying that I couldn’t grasp such a simple concept as the fact that two weaknesses are worse than one. But really, the problem is that I over-thought it and came up with all of these irrelevant reasons why one is worse than two. I wonder if it has to do with the psychology of inferior positions. Look at the words I wrote: “I didn’t like my position.” That’s an emotional reaction, not a rational one. Once you’ve decided you don’t like a position, you’ll start looking for moves that will make you like it. When you do that, you’re not thinking rationally any more.

In an inferior position, you should be thinking hyper-rationally. Your job is not to make the position good, but to keep from making it worse. One way to make it worse is to give yourself two weaknesses instead of one. Don’t do that!

17. O-O c5 18. dc?! …

Jesse again points out that this is a slight misstep by Bick. He should play 18. Nxc5, so that I will be forced to trade my bishop for his knight.

18. … Ne4 19. Rd1 Nxc5 20. Nd4 …

FEN:2r2rk1/p3bppp/8/2np4/1n1N4/4P3/1P2BPPP/1RBR2K1 b – – 0 20

DANA: “…”

JESSE: Dana gets a chance and doesn’t notice it. Bick should have played 20. Nxc5; it’s a simpler route to an advantage. I think that with 20. Nd4 he was playing too routinely, too much like a typical isolated queen pawn position. He give Dana a chance to play 20. … Ne4. Simple chess. Black wants to play Nc2. You have to be active here and fight for d4, and also perhaps create the possibility of … Bc5.

MY RESPONSE: For the second time, a move that Jesse sees as a turning point did not even register on my radar as being remarkable! Black still does have a time advantage, because White has not finished developing, and Jesse is looking for a way to exploit that.

When you put the position on a computer, you find that Jesse’s comment is very debatable. On 20. … Ne4 the computer says White should play 21. f3, a move that Jesse didn’t even talk about. He says that 20. … Ne4 is “simple chess.” Well, after 21. f3 it gets complicated. I suspect that his grandmaster intuition told him that 21. f3 weakens the e-pawn and the light squares in general, and after 21. … Nd6 22. Bd2 Nc4 Black’s knight can come around and start putting pressure on those points. But the computer comes up with 23. Nf5!, after which things get really tactical.

Here again we have to back up and realize that our goal is not to find the objective truth about the position, which might not even exist. Our goal is to improve our human thought processes, imperfect though they may be. Jesse is telling us how a grandmaster looks at the position. He sees Black’s need to fight for squares, fight to keep his time advantage. He should look at 21. f3, granted, but instead of getting sucked into a deep tactical calculation, he says look at the weaknesses. Look at the dynamic imbalances. Yes, 21. f3 buys White some time to develop his bishop, but it also weakens his pawn structure and his dark squares. Black has to be happier with this position than with a static position where he stands worse and has no counterplay.

As humans we can strive to emulate Jesse’s thought process. Even though he’s a grandmaster he is also human, and thinks in a way that we can think. We cannot emulate the computer’s thought process, and we should not strive to.

Instead of Jesse’s move 20. … Ne4, I played an “automatic” move that the computer considers to be better than 20. … Ne4.

20. … a5 21. Bd2 Bf6 22. Be1 …

Jesse didn’t comment on this but I thought it was a nice move by Bick. He is finally getting his pieces out of each other’s way.

22. … Rfd8 23. Bg4 Ra8?!

Both Jesse and I missed a tactical point here. Black actually can play 23. … Rb8, which appears bad because 24. Bxb4 Rxb4 25. Nc6 forks the rooks. But the bishop on g4 is hanging! So 25. … Rxg4 26. Nxe8 Bxe8 is actually fine for Black. Clearly the b-file is the right place for my rook, because it puts pressure on White’s weak pawn on b2. (This is, by the way, a weakness that Jesse never talked about, going all the way back to move eight. It’s a good question why Black gets criticized for his weak pawn on c7, but White gets a free pass on his weakie at b2.)

24. Bxb4 ab 25. Nc6 Re8

Setting a trap: If 26. Nxb4? Re4 wins a piece. (Loose pieces drop off!) This does cost me the d-pawn, but I thought that was more acceptable than losing the b-pawn.

26. Rxd5 Na4 27. Bd7 Nb6 28. Rd2 Nxd7 29. Rxd7 …

FEN: r3r1k1/3R1ppp/2N2b2/8/1p6/4P3/1P3PPP/1R4K1 b – – 0 29

I thought that Jesse’s comments on this position were especially interesting, because they answer the question, “What does a grandmaster know that I don’t know?”

In the game, I eagerly and gladly played

29. … Bxb2?!

thinking that this move would bring me much closer to a draw because the pawns would now be all on one wing. Jesse threw cold water on that idea.

JESSE: The bishop versus knight endgame, with all the pawns on the same side, is technically lost. The bishop’s strength is its ability to operate on both sides of the board. So Black should want to keep some queenside pawns on the board, in order to make his bishop stronger than the knight. To quote Smyslov, the knight is a local piece. But when all the pawns are on one side of the board, the bishop loses its strength and the knight gains in strength, because it can bounce back and forth to both colors. To learn more about how to win this sort of endgame, you should read Smyslov’s Endgame Virtuoso.

For this reason Black should try 29. … b3! He is probably still losing, but it at least gives him a chance.

MY RESPONSE: What a great explanation of the position! It’s fascinating that my intuition was 100 percent wrong: trading off all the queenside pawns (which I thought was good for me) was actually bad. To be honest, I really didn’t know. I haven’t studied B versus N endgames with pawns all on one side, ever. It’s a black hole in my education. Also, I did not know how rooks would affect the evaluation. I knew that R+4P versus R+3P is supposed to win for White, although it may take some good technique. I thought that N+4P versus B+3P might be a draw, but as Jesse explained, I was sadly mistaken. But what about R+N+4P versus R+B+3P? Jesse talked about that too. He felt that it was much better for the rook and knight, especially in this position, because they can gang up on the pawns on white squares, which Black’s bishop cannot defend. This is especially true if White’s rook is on the seventh rank, pressuring the f7 pawn, which is also vulnerable (eventually) to a pawn lever with e5-e6.

By the way, Fine’s Basic Chess Endings confirms that N+4P should beat B+3P. So this was known even before Smyslov. But still, if I really wanted to improve to 2300 or 2400 level, I should read Smyslov’s book. I haven’t done so yet, but I appreciate Jesse’s recommendation.

The game continued

30. Nxb4 Ra1?!

Again an inaccuracy. The b1 rook is not the one I should be trying to trade off. Instead I should trade off the much stronger rook on the seventh rank, with 30. … Red8. White doesn’t have a good way to decline this trade, and after 31.Rxd8 Rxd8 Black’s prospects are slightly better than they were in the game. “But White should still be winning” — Jesse.

31. Rxa1 Bxa1 32. g4 h6 33. Nd5 g6 34. Kg2 Kg7 35. f4 Bb7 36. Kf3 Re6 and White eventually won.

After this, unfortunately, I stopped keeping score. The tournament was played at what Jesse called the “fascist time control” of game in 90 minutes with a 30-second time delay. This has become standard in FIDE tournaments, but at the time it was my first experience ever with this time control. I didn’t know quite what to think when I got down to one or two minutes for the rest of the game. Am I in time trouble or not? And I didn’t even realize that one point of this time control is that you’re supposed to keep writing your moves down, no matter what. In USCF tournaments, if you get under 5 minutes and you don’t have a time delay (or have only a short one, like 5 seconds per move), then you don’t have to keep score. But in a FIDE tournament, you do. You always have at least 30 seconds to make every move, so it’s assumed that you have enough time to write your moves down.

After I got home (a few days later) I did try to reconstruct how the game ended. After an unknown number of moves, we got to a position something like this:

FEN: 2R5/1r3pk1/3b1Np1/6Pp/5P1P/4PK2/8/8 b – – 0 50

In my notebook, I wrote, “I was counting on 1. … Rb8, only to realize to my horror that White wins a piece with 2. Ne8+ Kh7 3. Nxd6 defending the rook! So I had to play 1. … Bf8, putting myself in near-zugzwang. Eventually, though maybe not immediately, White played 2. f5, and after 2. … gf 3. Nxh5+ I resigned.”

I learned a whole lot from this game about grandmaster thought patterns, my thought patterns, computer “thought” patterns, endgames and especially strategic imbalances. I’m grateful to Jesse for such a priceless lesson, given at no cost!