50 Years of Chess: Year 41

Recently I wrote a post in which I listed my best and worst tournaments by rating improvement or rating loss. But one of the highlights of my chess career was a tournament where I gained only 2 rating points! I also didn’t win any prize money or beat any USCF masters. (I did, by a technicality, beat a FIDE Master, but more about that below.)

Can you figure out what kind of tournament would make such an indelible impression in spite of the lack of individual accomplishments? If you answered “a team tournament,” go to the head of the class! 2012 was the year that I finally got a chance to play on a winning team in the U.S. Amateur Team championships (specifically, in the West region).

I wrote all about this during a series of posts in February 2012 (click on that month in the right margin if you want to see them all). So I’ll just focus on the high points here.

The team I played on was called “Forfeit by Disconnection,” and we had Robin Cunningham (now Robin Joseph) on first board, Julian Chan on second, Todd Rumph on third, me on fourth and Steven Gaffagan as alternate. Robin was my friend from the North Carolina days, in the mid-1980s, when he was a college student and I was a professor at Duke. We kept in touch over the years, and finally in the mid-2000s he moved to northern California. He was on the far northern side of the Bay area and I was way on the southern side, so we still rarely saw each other, but back in 2008 I had formed a Santa Cruz team for the short-lived East Bay Chess League and I had invited him to play with us. We won that season and then the league folded.

One thing I have to say about Robin is that he is like a good-luck charm for team chess. Besides our victory in the East Bay Chess League, he has been on four winning teams in the U.S. Amateur Team championship, in three different regions: the East twice (with “Walk Your Dog,” a UNC-Duke team in the 1980s), the South once, and finally the West in 2012. So he probably holds or at least shares the record for winning this tournament in the most different regions. Has anyone else won three?

There are two kinds of teams in the U.S. Amateur: “stacked” teams, with two or three highly rated players and a weak fourth board, and “balanced” teams, in which all the players are roughly the same strength. No matter which kind of team you have, the average rating of your players has to be below 2200. We were a balanced team. Robin at 2261 was a little higher, and Steven at 2044 was a little lower, and the rest of us were all in the 2100-2199 range.

We had two ingredients for success in this tournament. One was a perfect 6-0 mark on board four: Steven Gaffagan won all of his games and I won in the two rounds where he didn’t play. As a balanced team, it’s important to win all of your “must-win” games on board four, because you know you’re going to take your lumps on boards one and two. Second, we were unbelievably lucky. After winning 3.5-0.5 in round one and drawing 2-2 in round two, we won four consecutive matches by the narrowest margin, 2.5-1.5. And we did it again in the first round of the playoffs. You can’t win five consecutive matches by one point without some timely breaks. Steven won from an exchange-down position in round four. Julian pulled out an amazing endgame save in round five. Todd built a fortress in an exchange-down endgame in round six. In the first round of the playoffs, Steven swindled a draw by perpetual check when he was three pawns behind. Every one of these was a critical half-point that came like a gift from the chess gods.

To me, the most important match came against NorCal House of Chess in round five. Part of its significance became apparent only in retrospect. They won the next three years in a row, not only the West region but the whole country, becoming the first team ever to do that. Although we didn’t know it at the time, we stopped them from four-peating! They had IM Ricardo de Guzman on board one, FM Ronald Cusi on board two, expert Ted Castro on board three, and FM Tanuj Vasudeva on board four.

You mighty wonder how you could have a FIDE Master on board 4 and still have an average rating less than 2200. Well, Vasudeva was actually a USCF expert, but he had won the Pan-American Under 8 championship a couple years earlier, and the winner of that tournament earns an automatic FIDE Master title. So he had the title, but he didn’t have the rating.

NorCal House of Chess was a formidable opponent. They had practically the maximum possible rating: a team average of 2199.5, in addition to a kid on board four whose strength might be a couple hundred points above his rating. Todd and I did our jobs, winning on boards three and four, but Robin lost to de Guzman on board one. It all came down to Julian’s game against Cusi on board two. When we left for lunch, Robin thought that Cusi’s Q+N would be too much for Julian’s Q+B. I didn’t agree: I couldn’t see any clear path for Cusi to win. When we came back, it turned out that I was right! Cusi had over-pressed in his efforts to win and allowed Julian to get a perpetual check. (That game, by the way, was the only blemish on an otherwise perfect tournament for Cusi, who went 5.5-0.5.)

Julian’s game may have changed the course of history. If he had lost, the match would have been tied 2-2, both teams would have ended the tournament with a 5-1 mark, NorCal would have won on tiebreaks, and maybe they would have won four titles in a row instead of three.

In April 2012, we had a playoff against the North, East, and South champions online. (Actually, the East champions couldn’t play that weekend and the runner-up team participated instead.) Our run of luck continued against the North champion in the semifinals. Steven Gaffagan pulled out a ridiculous draw in a position where he was three pawns behind and his opponent had four connected passed pawns. (Click here to read how he did it.) And Julian pulled out a win in a game that earlier had appeared to be drawn at best. Just like that, a match that looked like a possible 1.5-2.5 defeat turned into yet another 2.5-1.5 victory, our fifth in a row.

In the finals, our luck finally ran out against the East team. We actually drew first blood, as I won on board four against an expert named Hibiki Sakai. If you’re keeping track, that means that our board-four players had a record of 7.5-0.5 in the six rounds of the West tournament plus the two rounds of the playoffs.

Unfortunately, Julian lost his game on board two about a minute after mine finished, so our 1-0 lead did not last long. And the other two games, Todd’s and Robin’s, went into a tailspin. Both of them had inferior positions, but if it were an individual tournament they might have held draws. But because it was a team tournament, they both felt compelled to play risky moves to salvage the match, and they both ended up losing. So we lost the match, 3-1.

Still, I think that Robin had the best take on it: we finished second out of more than 400 teams who played in the tournament nationwide. And we had so much fun, and so many amazing moments. So there is no need to feel sorry for us. For me it’s right up there with my two North Carolina championships as the best tournaments of my life.

The game I’d like to show you from 2012 was my win against Hibiki Sakai in the playoffs. Like the game in my previous post, I got absolutely no advantage out of the opening, but gradually amassed little positional advantages during the middlegame. When the position finally opened up, Sakai almost immediately walked into a tactical trap, but even if he had not done so, he was in a lot of trouble. It’s kind of a funny game, where it’s hard to put a finger on any major errors by Black (aside from the one at the very end), yet his game just went gradually downhill.

Dana Mackenzie — Hibiki Sakai, 4/20/2012

1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. cd cd

At the time, one of the young players in Santa Cruz, Cailen Melville (a class-B player), was into the Exchange Slav, so I had started playing it too.

5. Nc3 Nc6 6. Qb3 …

This was Cailen’s idea to break the symmetry. In principle I think that Black can probably hold after 6. … Qb6 7. Qxb6 ab, but it’s hard to imagine very many players of the Black pieces wanting to go into this position. The downside of Qb3 is that b3 is a somewhat awkward square for White’s queen. This can be turned into an upside, though: Black may feel tempted to play … a6, … Na5, … b5, and … Nc4, and then White will try to prove that Black has overextended his position.

6. … e6 7. Bg5 …

Here Cailen and I diverged. He preferred 7. Bf4, but I was attracted to the idea of playing it just like a Queen’s Gambit Declined and arguing that Black’s game is even more lifeless than usual. To be honest, either approach is fine. One of the difficulties of annotating games in this opening is that every move “equalizes,” but they all have slightly different flavors.

7. … Be7 8. e3 O-O 9. Bd3 h6 10. Bh4 b6

All super-standard stuff, but for the reader’s sake I will put a diagram here. The one thing about the position that is not really standard is the position of White’s queen. Where will it go?

FEN: r1bq1rk1/p3bpp1/1pn1pn1p/3p4/3P3B/1QNBPN2/PP3PPP/R3K2R w KQ – 0 11

11. Rc1 Bb7 12. Bb1 Rc8 13. O-O Ba6 14. Rfd1 …

This move may be inaccurate because it takes away the flight square d1 from my queen. Black could take advantage of this fact with 14. … Na5 and I have two not entirely satisfactory choices: 15. Qc2, which allows 15. … Ne4 followed by an exchange on c3, giving White a backward pawn; or 15. Qa4, after which Black can play 15. … Qe8, either forcing a queen trade or forcing 15. Qc2, which we already said was not so hot.

The above is computer analysis, of course. In human play, Black would have to think hard before playing … Na5. Is it okay to allow White’s knight into e5? Will Black’s knight be poorly placed on a5 after White moves his queen away? And in that line 14. … Na5 15. Qc2 Ne4 we just mentioned, is Black really threatening … Nxc3? Not really, since it would be answered by Qh7 mate.

I’m not saying the computer evaluation of 14. … Na5 is wrong, but there are lots of things to consider. Anyway, my opponent chose to move a different knight to a different rim. It’s one of several places in this game where he played more passively than the computer recommendation. Individually, every case is debatable, but on the whole they form a pattern of unenterprising play. He kept playing it safe, and eventually he was sorry.

14. … Ne8?! 15. Bxe7 Qxe7 16. Ne5 …

Over the last couple moves there has been a subtle shift in the position. Black now has to worry about how he is going to defend the kingside. If White’s knight stays at e5, the b1-h7 diagonal becomes more tender because Black can’t block it with … g6. If Black trades on e5, White’s pawn on e5 is strong and White also has a nice outpost on d4 for a knight.

Instead of describing the Exchange Slav as “boring and drawish,” perhaps we should call it “subtle.” The positional imbalances are smaller than they are in other openings, but they are still there.

16. … Nxe5 17. de f5!?

FEN: 2r1nrk1/p3q1p1/bp2p2p/3pPp2/8/1QN1P3/PP3PPP/1BRR2K1 w – f6 0 18

Black’s move spices things up considerably! White faces an important decision: to take or not to take en passant?

The computer thinks that I should not take, because I have a better move: 18. Nb5! Notice first that 18. … Bxb5? 19. Rxc8 loses the exchange. Black could avoid that by trading rooks first with 18. … Rxc1 19. Rxc1 Bxb5 20. Qxb5, but this position looks very good for White. Black’s knight has no moves, and even after he runs away to g7, White’s bishop will be a better minor piece. After 18. Nb5 the computer thinks the best move for Black is 18. … Rc5, but now White plays 19. Nd4 and gets his knight to its ideal outpost.

But there are also good arguments for taking en passant. After 18. ef Black will get three pawn islands — definitely a weaker pawn formation, especially if I can blockade the center pawns. A second argument was that I already had an answer to the Mike Splane Question: How am I going to win this game? I felt certain that the way to win was with a kingside attack, focused on the b1-h7 diagonal. So I was determined to keep that diagonal open. If I played the computer’s move, 18. Nb5, I would have to come up with a completely different strategy for winning the game. After 18. Nb5 Rc5 19. Nd4 Nc7 I don’t see what White’s winning plan is. Does anybody else have any ideas?

Because 18. ef contributes to my master plan, that’s the move I played.

18. ef Nxf6 19. Qa4 …

Playing on both sides of the board. Note that White has to keep a careful eye out for Black’s move … Ng4, followed by … Qh4, which would give Black a blitzkrieg attack on the kingside. Perhaps this is why the computer did not approve of my move 18. ef.

19. … Bb7 20. Ne2! …

White correctly declines the poisoned pawn on a7, because 20. Qxa7? Ng4! is just what Black wants. Instead of tilting at windmills on the queenside, White’s idea is to bring the knight to f4 and g6, a beautiful outpost.

20. … a6?!

Continuing down a path toward passivity. Black worries too much about his a-pawn and hastens to move it to a safe square. But meanwhile he does nothing about my knight’s march. The most courageous and most principled move would be 20. … e5, to prevent Nf4. Now we would have a new question: is Black’s d5-e5 pawn duo a strong center or is it overextended? Again, not an easy question. This game is full of questions that aren’t easy to answer.

21. Nf4 Qd6 22. Ng6 Rfe8

On the principle that exchanges relieve a cramped position, Black might want to consider trading rooks with 22. … Rxc1 23. Rxc1 Rc8.

23. Rxc8 Rxc8 24. Qd4 Nd7

Black belatedly turns his attention toward forcing the break … e6-e5, but the time for that was earlier. White is now in a position to prevent it and obtain a blockade on Black’s center pawns.

FEN: 2r3k1/1b1n2p1/pp1qp1Np/3p4/3Q4/4P3/PP3PPP/1B1R2K1 w – – 0 25

25. f4 …

An obvious move, but nevertheless I like it. White is willing to accept a slight weakening of his pawn structure in order to keep firm control over the center and in order to keep the b1-h7 diagonal open.

25. … Re8

Pursuing his impossible dream of playing … e5. It’s hard for Black to come up with any play here. Maybe he should just play 25. … Qc5, not so much with the intent of trading queens (which would be bad) but just to sit and ask White what he plans to do.

26. Ne5! …

Again, I’m willing to weaken my pawn structure to maintain my bind.

26. … Nxe5 27. fe Qc7

Another micro-inaccuracy? 27. … Qc5 brings up the issue of a queen trade and, very importantly, it protects the f8 square. You’ll see why this is important very soon.

28. Rf1! …

Of course, 28. Qd3 is also tempting, but as a general rule it’s good to close the mating net before you start chasing the king.

28. … Rf8?

FEN: 5rk1/1bq3p1/pp2p2p/3pP3/3Q4/4P3/PP4PP/1B3RK1 w – – 0 29

Black’s only clear blunder of the game. But it would be wrong to say that Black lost because of this blunder. It was all the other micro-inaccuracies, all of the chances he had to put a little bit of pressure on me but didn’t, that eventually led to a position that was already almost impossible to defend.

29. Bh7+! …

Sweeping from one end of the board to the other, this deflection sacrifice is the grand culmination of White’s strategy of controlling the b1-h7 diagonal.

29. …Kxh7 30. Rxf8 Black resigns

The sudden ending of this game does bring up a question: Was White’s position actually winning even before Black played 28. … Rf8? In my previous post about this game, I wrote, “he has to play 28. … g5 in order to defend h7.” Well, 28. … g5 is sheer desperation, so if Black is forced to play that move he must be in a world of trouble. As I wrote in my earlier post, 29. Qd3 Qg7 30. Rf6! is very strong, and after 30. … Kh8 the most accurate move is not 31. Qf1 as I wrote before, but 31. Rg6 Qh7 32. Qf1, when the threat of Qf6+ is decisive. Or 31. Rg6 Qf8 32. Rxe6!, when Black can’t take back because of the mate threat on h7.

I think that the most testing defense is 28. … Re7, a move that doesn’t create any weaknesses and doesn’t walk into any immediate tactical refutations. What would you do next as White?

FEN: 6k1/1bq1r1p1/pp2p2p/3pP3/3Q4/4P3/PP4PP/1B3RK1 w – – 0 29

I suspect that my intention here was to play 29. Bg6. That’s why I wrote that Black “has to play g5.” And in fact, 29. Bg6 is a very good move. Black has to worry about back-rank mates as well as several other problems: the check on f7, the pawn on e6, etc.

For those who like to live on the wild side, though, the computer finds something even better, and much more satisfying. It recommends 29. Qg4!!, offering two pawns in order to checkmate Black’s king. The main point is that after 29. … Qxe5 30. Qg6, the only way for Black to stop mate is to give up his queen. The threat of Qg6 is so strong, in fact, that Black probably will have to play the desperate move 29. … g5. Then White answers 30. Qh5!, continuing to offer the two-pawn sacrifice. This is a position very much like the one in my last post, where White is free to forget about anything happening outside the critical zone, roughly defined as the square from f6 to f8 to h8 to h6. Inside the critical zone, White has three attackers to Black’s two defenders, and that is all that matters. For example, if 29. Qg4 g5 30. Qh5 Qxe5 31. Qxh6 (threatening mate on f8) Qg7 32. Bh7+! crashes through. Or if 29. Qg4 g5 30. Qh5 Qxe5 31. Qxh6 Qxe3+ 32. Kh1 Rf7, again 33. Bh7+! is the winning shot. This is starting to get repetitive! Black plays anything, White sacrifices the bishop on h7 and wins.

Although some might say that 29. Qg4 is a computer move, and others might say that it is unnecessary brilliance because 29. Bg6 is also very effective and much safer, I think it is worth seeing these lines for two reasons: they show us the power of the “critical zone” concept and they show the usefulness of the Mike Splane question. As far back as move 18 I thought that I was going to win by checkmate on h7, and that is almost exactly what happened. The lines after 29. Qg4 are the most fitting culmination of that strategy.

Lessons:

1. You can win (and lose) in “boring and drawish” openings too. They are just a little bit more subtle than tactical openings.
2. “Better safe than sorry” doesn’t work in chess. If you always play it safe, eventually you’ll be sorry.
3. In quiet positions, try to accumulate incremental positional advantages: good outposts, open lines for your pieces, weaknesses in your opponent’s position, etc. Have faith that these things will be useful to you in the end.
4. By asking the Mike Splane Question, “How will I win this game?” you can give direction to your play.
5. When you have a preponderance of material in the critical zone, don’t hesitate to sacrifice material that is either irrelevant to the critical zone, or which might lure your opponent’s pieces away from the critical zone.
6. First close the mating net, then start chasing the king.