To Teach Openings, or Not?

When I started teaching the St. Pius kids, I only devoted one class to openings, and then only to basic principles. I explained that the three goals of the opening were to control the center, develop your pieces, and protect the king. I explained the difference between classical and modern development, showing them the Ruy Lopez and King’s Indian Defense main lines as respective examples. But I didn’t teach them any more specific lines, or insist that they play any particular openings. I didn’t want to force them to play particular lines, especially when I know my biases would leak through (I play a more aggressive, tactical style than most, and currently favor the Ruy Lopez or Evans Gambit for white, and the Sicilian Najdorf, Marshall Gambit, and Grunfeld for black).

And then I sent them off into the tournament world to get beaten up with weak opening play.

Looking at my students’ games without specific opening guidance, I’m seeing a lot of either overly cautious play or tactically sloppy play. My star student, lacking an idea of how to defend against e4, stumbled into the Damiano trap as Black in his first game at the Chicago championships: 1. e4 e5 2. Nf3 f6? 3. Nxe5 fxe5 4. Qh5+ g6? 5. Qxe5+ Qe7 6. Qxh8. Other students like to play passive moves; many of my games against Kris as White have started: 1. e4 d6 2. d4 e6? 3. c4 c6?.

So, no more friendly “do what you want” opening teaching. This week I start teaching them how to develop a real opening repertoire. They get some choice in which openings to pick, but only a limited choice.

1. As White, I want them all playing 1. e4. 3 reasons of varying quality for not letting them play 1. d4. First, the good reason: as many other chess teachers have noted, king pawn openings tend to be more open and tactical, and that’s what beginning students need to practice. Second, the practical reason: queen pawn openings tend to lead to complicated, subtle, positional play in closed positions, and these kids just aren’t ready to learn that style of play. And third, the less good reason: I know king pawn openings about 50 times better than I know queen pawn openings for White; my Queen’s Gambit, Slav, and Nimzo-Indian knowledge is practically nonexistent, while I’ve probably played as White against the Sicilian about a thousand times by now. Some day, I will have a student advanced enough who wants to learn the Queen’s Gambit, and I’ll have to learn the theory to teach it. But not today.

2. Having played 1. e4, they get their choice of responses to double king pawn openings limited to the Ruy Lopez main line, Giuoco Piano, or Evans Gambit. The King’s Gambit, Scotch Game, and Three Knights/Four Knights Game are all playable, but the first requires tricky theoretical knowledge and the latter two are pretty boring and passive. The Ruy Lopez and Giuoco Piano are very sound and easy to understand, and the Evans Gambit, while less sound, is an easy gambit to play and emphasizes attacking and fast development.

3. In response to 1. e4, they can choose between the Ruy Lopez, Sicilian Dragon or Najdorf, French, or Caro-Kann. All sound, playable defenses that aren’t hard to understand. I don’t want them trying to play the Pirc or Modern, both defenses that don’t directly challenge the center–they’re too willing to play passively in such lines.

4. In response to 1. d4, they can choose between the QGD, Slav, or King’s Indian. Either classical development or aggressive modern development are OK. The Grunfeld is a little too theory and trick based for these kids, and I just don’t have a good enough grasp of Nimzo- or Queen’s Indian Defense theory to teach it.

Opening principles are more imporant than lots of memorization, but a little memorization is necessary, and will help the kids get more wins.

Back, now as a coach!

So, after the NY International in June, my life got taken over by a wedding and move, and then I kind of forgot about this blog. But in November of last year, I began coaching the St. Pius V chess team in Chicago. These kids started knowing little more than the rules, but in a few months of not too structured lectures, we have 5-6 kids in grades 4-8 who have tournament experience, and one 5th grade girl won the 1st place unrated trophy at the “Greater Chicago Scholastic Chess Championships”, K-6 U600 section last Sunday.

My only tournament since then was a 4 round rapid tournament in December, one of the Knight’s Quest series in Northbrook, IL in which I scored 2/4 with draws against 1500 and 1900 players, a win against a 2000, and a loss to a 2150.

So, more coaching and less playing (and no directing, though I’ll be running my own scholastic tournament in May).

Winning!

I’ve been busy with work and wedding planning for the last few weeks, and my New Yorker Open games were a kind of sad and mediocre lot. I went 3/6, losing a full 2 points on obvious blunders, and getting bogged down in a few long drawn out draws and losses.

The New York International, on the other hand, went much better! Playing in the U1900 section, I went 4/5 (3 wins, 2 draws) to tie for 2nd and clearly win the 1st U1800 prize. I came very close to winning round 5 as well, but the late night and tiring game had me missing the square coverage that allowed an easy win for me in the final endgame position.

The best game of the tournament, though, was Game 3, an explosive Sicilian Kan/Maroczy game in which I calculated out an 8 move mating net over the board:

Brian Beck (1790) vs. Alexandra Wiener (~1830)
New York International, U1900 Section, Round 3

1. e4 c5 2. Nf3 e6 3. d4 cxd4 4. Nxd4 a6 5. c4

Not a line for White where I know much theory. The basic idea is that without Black’s playing 4. … Nf6, white can gain extra control of the center with 5. c4, the Maroczy Bind position. This appears to result in positions that are more of a fight for the center than the more typical alternate side attacks of, say, the Be3 lines of the Najdorf or Dragon.

5. … Qc7 6. Bd3 Nf6 7. Nc3 Bc5 8. Nb3 Bd6 9. h3 Nc6 10. Be3 Ne5 11. c5!

A key move for White. At this point, White remains strong in the center and has completed development, while Black’s light-squared bishop remains entombed and her king is still stuck in the center. 11. c5! presses White’s advantage, driving Black’s decent bishop back to the ineffective e7 square.

11. … Be7 12. O-O?! (With a space advantage, White should avoid trading pieces and giving Black more room to move. Fortunately, Black thinks that the major issue in the position is her good knight on e5, and refuses to trade) d6 13. cxd6 Bxd6 14. Be2 b5?

Black tries to relieve the cramping, but tactics dominate strategy. I actually missed the strongest tactic, 15. Bxb5!!, after which 15. … axb5 16. Nxb5 Qc6 17. Nxd6+ wins 2 pawns. But this line is still ugly for Black:

15. Rc1 Qd7 16. f4! Nc4 17. Bxc4 bxc4 18. Nd2 Bb4?

A big mistake. Now Black pays for her uncastled king:

19. Nxc4! Bxc3 20. Nd6+!! Ke7? (Kf8 is better, because Black can get away from White’s rampaging pieces)

21. Rxc3! (Black can’t take on d6 because Bc5 wins the queen) Bb7 22. Bc5 Ke8 23. Nxb7+ Kc7

Now, I could just sit back, trade queens and win the piece up endgame. But “when you see a good move, look for a better one!” I had 20 minutes left on the first time control, so why not take a look, and I calculate out my mate in 8:

24. Bd6+!! Kxb7 25. Qb3+ Qb5 26. Rc7+ Kb6 27. Qe3+ Ka5 28. b4+ Qxb4 29. Bxb4+ Kxb4 30. Qb3+ Ka5 31. Rc5#

Sometimes you just have to calculate

An exciting endgame from Sunday’s deciding game in the Championship section is instructive:

White to play, Azeez Alade (2040) vs. Matthew Zafra (~1850)

I’m fairly sure Matthew had a win earlier, but he just needed a draw to win the tournament. A win for Azeez would have won the tournament for him as well. So, Azeez takes the chance he needs to to win:

1. Kb6 Kd7?

Kd7 is already losing! Black can easily hold the draw with 1. … Bb8!, and because White can never cover the promotion square with his king, White can’t make any progress as the kings block each other forever. White should have a win after Kd7, but…

2. Ka7 Bc7 3. Draw offer?? 1/2-1/2

Azeez just didn’t calculate far enough here, probably confusing the actual position after promotion and trade:

With this one:

In the real position, while Black’s king is way ahead of the game, but white’s extra pawn on f3 proves surprisingly useful:

3. b8=Q Bxb8 4. Kxb8 Kd6 5. Kc8 Kxd5 6. Kd7 Ke5 7. Ke7 Kf4 8. Kf6 Kxf3 9. Kxg5, after which white easily promotes. Without that f3 pawn, Black wins the race with 8. … Kxg4.

Unbalanced endgames are frequently decided by these races that require calculating 7-10 moves ahead. When you get to one, that’s the time to buckle down and think–trying to get a result on principle might just throw away a perfectly good half point.

Never Give Up! (in blitz)

There was a small turnout at yesterday’s NYChessKids tournament, where I directed an 8 player Open section (1300-1600) and a 6 player Championship section (1600+). We ended with an exciting blitz playoff in the Open section. Oliver Neubauer of Dalton and Ben Drevitch of M.S. 54 tied for first with 2.5/3, and in the first game of the playoff, Oliver won after Ben had a brain fart and hung his queen. In the second game, Ben did better, and got to a winning endgame, but something went wrong:

White to Play

With a minute left on each side, Ben (playing White) overlooked that his bishop covered g8, so he missed the quick 1. g7+ Kh7 2. g8=Q+ Kh6 3. Qg6#, and instead let Oliver swindle him out of the half point he needed to keep the playoff alive:

1. h6? b3! 2. Bxb3? a2! 3. Bxa2?? 1/2-1/2 (stalemate).

In blitz, it’s really never over until it’s over.

Another opening change

After a few King’s Indian Defense rounds in a Chess.com correspondence tournament, I’m realizing that opening just doesn’t fit my tactical style. It’s an attacking opening, but one that’s heavy on alternate side pawn attacks in a closed center, and I don’t have a good feel for when defensive moves or attacking moves are necessary in those positions.

So, to finish my transition from a hodgepodge of opening choices in my scholastic years to a repertoire that fits my calculating, tactical style, my new response to 1. d4 will be the Grunfeld. It’s certainly become a lot more popular lately among GMs, and had a very good showing in round 1 if the World Championships, with Anand having to salvage a draw from a weaker position as White.

FIDE Rating established!

Last night’s game was a pretty cool win against a 1950, and this time it was my opponent who blew a draw in one move. But it’s a lot shorter.

Brian Beck (USCF 1790, FIDE UNR), vs. Mulazim Muwwakkil (USCF 1910, FIDE 1943)
Marshall FIDE Mondays, Rd. 4, G/120 d5
French Defense, Rubinstein Variation

1. e4 e6 2. d4 d5 3. Nd2 dxe4 4. Nxe4 Bd7 5. Nf3 Bc6 6. Bd3 Nd7 7. c3?!

A little passive–apparently the correct line is 7. O-O Ngf6 8. Ng3 Be7 9. Re1 O-O, where White holds on to more pieces to pressure Black’s cramped position. 7. c3 solidifies the center, but the line I play lets Black trade off a few pieces to gain space.

7. … Ngf6 8. Qe2 Be7 9. O-O O-O 10. Re1 Nxe4 11. Bxe4 Bxe4 12. Qxe4 c6

White has a slight, solid advantage; control of the center (black’s not moving that e pawn for a long time), a little lead in development, and more space both in the center and the kingside. With a fairly open center, piece play is the order of the day. I decide to go with a kingside piece attack; this isn’t as effective as I’d like, and with ideal play, Black would equalize at some point with c5, breaking up my center and/or opening the c file for black’s rooks. The better plan is to secure the center first and finish development, starting with Bf4. Instead, wanting to keep my bishop flexible, I play:

13. Qg4 Nf6 14. Qh3 Qc7 15. Re5 Rad8 16. Be3 Rd5 17. Re1 Rxe5 18. Nxe5 Qa5 19. a3 Qa4

An immediate 20. Bg5! is the best way to take advantage of Black’s cramped position, but I keep shoving at Black’s king:

20. Qg3 Ne4 21. Qf4 f5

An awful blunder would be 22. f3??, after which 22. … g5!! traps White’s queen.

22. Nd7 Bg5 23. Qe5

And here’s the line where something exciting will happen. Black’s f-pawn push drastically weakened the e-pawn, but Black has back rank tricks to stay alive.

23. … Bxe3 24. Qxe6+ Kh8 25. fxe3 Qc2 26. Rf1 Qe2??

White to play and win…

27. Qxe4!!

Black can’t recapture on e4 because 28. Rxf8 is checkmate. Black can’t recover the piece. 26. … Qd3! would save the draw for Black, after 27. h3 Qxe3+ 28. Kh2 Ng3 29. Re1! (29. Qxe3 Nxf1+ 30. Kg1 Nxe3 31. Nxf8 leads to a slightly better endgame for Black, though probably not winning) … Nf1+ 30. Kh1 Ng3+ draws by repetition. In the game, however, Black’s position is resignable in a few moves:

27. … Qxf1+ 28. Kxf1 fxe4+ 29. Nxf8 Resigns.

A solid game from me, no serious errors but not the best moves (but if I always made the best moves, I’d be aiming for GM, not expert, right?)

Well, by my FIDE rating, I’d already be an expert now! 3 games against FIDE players, with at least 1 point, is enough to establish a rating, and if I didn’t play the remaining tournament games, that rating would be 2022. Let’s see if I can keep up this pace in the last two games.

The Four Tiers of Chess Skill

I was thinking about my recent games, and the scholastic games I’ve watched, and I would suggest that their are roughly 4 levels of chess ability, defined by how the player comes up with his moves.

Level 1: Random chess (roughly 0-1000 by rating)

These players know the rules, and maybe have an idea about really basic tactical ideas (e.g., what a fork is), really basic positional ideas (e.g., put rooks on open files), and what pieces are worth. But these players are defined by making the first move that pops into their heads, maybe stopping to consider if their opponent has an obvious reply. Within this level, better players will have a better understanding of tactics, and may have a better idea of what kind of move is good, but without developing a more practical thought process, they won’t break into class E.

Level 2: Hope chess (roughly 1000-1700 by rating)

These players have started taking their opponent’s moves into account, but not in a methodical way, and they still make the first move they like. The rule for these players is, “if you see a good move, make it.” As they go up the ladder from E to A, they improve their tactical vision, develop a better understanding of imbalances, and learn how to win simple but nontrivial endgames, and their instincts improve, but they’re still not using an efficient thought process. They’re playing what Dan Heisman calls “Hope Chess,” where they are only looking at the first possible move their opponent might respond with rather than really looking to see if their move fails against any opponsing response.

Level 3: Real chess (roughly 1700-2300 by rating)

The top amateur players reside here, where having developed impressive tactical skills and an effective thought process, they are now troubled by the difficulty of positional evaluation. Whild players here are much less likely to make obvious tactical errors, their errors are likely to be either in insufficient analysis (such as stopping short in a position that requires 6 or more ply of analysis to really figure out) or incorrect evaluation (such as incorrectly determining whether closing the position with a pawn push or exchanging is better for the position). Players above this are Masters, and are thinking about the game in a very different way from amateurs.

Level 4: Master chess (roughly 2300+ by rating)

Players at this level have developed a positional vocabulary; they know from memory who’s better and why, and what to do, in a wide variety of position types, and so when they analyze positions not in their vocabulary, they can rely on this vocabulary rather than on heuristic positional concepts. A 2000 player analyzes a Najdorf Sicilian middlegame and looks to see how each side keep control of critical squares and activates their pieces. A 2400 player looks to see if the line of moves leads to a position he knows is winning. Studies show that as you move up from expert to master to grandmaster, players have fewer and fewer positions each game where they have to figure out what to do from heuristic principles.

There’s a reason that only the most dedicated players can start playing Master chess. It requires a level of dedication and study that the average weekend player can’t match. Which is why my goal, for now, is just to reach first Expert (and the 61 point bump from the New York International will help with that!), and then maybe at some point NM. IM is the realm of professionals with a lot more time on their hands, and of child prodigies (who also have a lot more time on their hands).

Chess is Frustrating

Monday night’s game ended my non-losing streak, though only at the last minute. 45 moves of perfectly solid play leading to a drawn position against an expert, only to throw it away on one move with 13 minutes left on the clock.

Brian Beck (FIDE UNR, USCF 1790) vs. Edward Kopiecki (FIDE 2006, USCF 2041)
MCC FIDE Mondays Apr-May 2012, Rd 3
Ruy Lopez, Worrall Attack (C77)

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba5 Nf6 5. Qe2 b5 6. Bb3 Be7 7. O-O O-O 8. c3 d5 9. d3 d4 10. cxd4 Nxd4 11. Nxd4 Qxd4 12. Be3 Qd6

13. f3?!

Book moves are Nc3 or Rc1; h3 is also popular. 13. f3?! isn’t awful, and white is perfectly safe afterwards, but it’s very quiet and drawish.

13. … Rd8 14. Rd1 Be6 15. Bxe6 Qxe6 16. Nd2 Nd7 17. Nb3 c5

Computers evaluate the game as no more than 0.1 in either direction for nearly the entire game. The play proceeds naturally from here, as White concentrates on the c-pawn, trying to force Black to move to c4 and trade it for his backwards d-pawn, preferably leaving a weak isolated c pawn for black.

18. Rdc1 Rac8 19. Rc2 Rc7 20. Rac1 Rdc8 21. Qf2 c4 22. dxc4 Rxc4 23. Qe2 Rxc7 24. Rxc7 Rxc7 25. Qxc7

As planned; White’s pawn weakness is traded off and White’s pieces are slightly more active.

25. … h6 26. a3 Bg5 27. Bxg5 hxg5 28. Qc3 g4 29. Nc5 Qb6 30. b4 a5?

Black’s first blunder; it turns out that White can safely capture on g4 here with a substantial advantage. I see ghosts, however, and play instead:

31. Kf1? axb4 32. axb4 Nxc5 33. bxc5 Qa6 34. Qxe5 gxf3 35. gxf3 b4+ 36. Kf2 Qb5 37. Qd5 b3

Now at this point I’m going to need to start checking to hold a draw, which I do:

38. Qd8+ Kh7 39. Qh4+ Kg6 40. Qg4+ Kf6 41. Qf5+ Ke7 42. Qe5+ Kd7 43. Qd6+ Kc8 44. Qf8+ Kb7 45. Qxf7+ Ka6

And from here, White draws in lots of different ways, such as 46. Qf8! threatening Qa8#, or 46. Qe6+ Ka5 47. Qd5!, after which an immediate b3 loses to Qa2+, winning the pawn and trading queens. Instead, miscalculating and missing lines with Qc5+ covering the promotion square, I throw the game away with

46. c6?? b2 47. c7 Kb7 48. Qf8 Kxc7 49. Qxg7+ Kb6 50. Qh6+ Qc6 51. Qf4 b1=Q 1-0

All that solid chess thrown away in one move!

GM Nigel Davies on Rust Removal

GM Nigel Davies: Maybe Long Play is Better. Davies noticed that he had a significantly harder time in rapid chess tournaments than in long play tournaments, and I’m noticing the same problem. When you come back, tactics, positional play, and endgames all come back easily enough. Those are concepts that are relatively easy to remember, just hard to execute.

No, the tough rust to remove is forgotten opening knowledge, and that’s where an extra few hours really comes in handy. When you need to start thinking up opening moves from scratch on move 6, you need a lot more time than your opponent who’s got the book lines memorized through move 12, and then has a good general knowledge of the strategies that come out of that particular opening.

I wonder if, for the resuming player, it makes sense to focus on more obscure but still sound lines of major openings. Like the Ruy Lopez Worrall Attack (1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. Qe2), my new White opening of choice. Force your opponent to fumble around as much as you do. Of course, that doesn’t quite fit with my choice of black opening (Najdorf Sicilian and KID)…