Master of Improvisation

by admin on August 11, 2018

Master of Business Administration, Master of Chess, … All of these titles describe Mike Splane, but I would like to add a new one to them: Master of Improvisation.

Mike sent around an e-mail yesterday with a game he had just played at the Kolty Club on Thursday night. It obviously had to be something special, because he doesn’t usually do that. It is, indeed, a fascinating performance that shows perfectly a certain Jekyll-and-Hyde characteristic of Mike’s chess: He has a remarkable ability to save and win positions that are just plain bad.

In this game Mike, practically sleepwalks through the opening. Yet just when it looks as if he is about to walk over the edge of the cliff, he comes up with an amazing series of sacrifices that throw the game into chaos. As he wrote, “I had to play the wildest moves I could find to keep everything imbalanced. I attribute the win to luck far more than to skill, unless you call ‘keeping it complicated and setting problems for the opponent to solve’ a skill.”

Mike Splane — Frisco del Rosario

Mike mentioned that this opponent has been very difficult for him. After winning his first three tournament games against Rosario, Mike had not beaten him for twelve years. According to the USCF website, his record is +2 =8 -3.

1. Nf3 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nc3 d6 6. d3 Nc6 7. Rb1 e5 8. b4 Re8 9. e4 h6 10. h3 Be6 11. Be3 Qd7 12. Qd2 Kh7 13. Nd5 Ng8 14. Nh4 Nce7 15. g4 c6 16. Nc3 d5

Position after 16. … d5. White to move.

FEN: r3r1n1/pp1qnpbk/2p1b1pp/3pp3/1PP1P1PN/2NPB2P/P2Q1PB1/1R2K2R w K – 0 17

Mike’s laconic annotation: “I miscounted when I played Nc3 and for a moment I thought this was a pawn sacrifice. Then I saw I was in a bit of trouble.” I think we should add Master of Self-Deprecating Humor to Mike’s titles!

I’d like to say a little bit about where I think Mike went wrong in this opening. One of Mike’s favorite comments is that you shouldn’t castle if there is something better to do. Even though I concur in general, I think he sometimes takes this advice to the extreme. Here he has undertaken all sorts of middlegame adventures (Rb1, b4, Nh4, Nd5 come to mind) when he has still not resolved the basic issue of king safety. I feel as if he has been asking himself, “Besides castling, what else can I do in this position?” The trouble with this approach is that if something unexpected happens, the king could get caught in the center.

Here “something unexpected” is that Black threatens … d4 winning a piece, and the only way to stop it is to trade off the c- and e-pawns, giving Black a very mobile center and giving White a weak pawn at d3.

17. cd cd 18. ed Nxd5 19. Ne4 b6 20. O-O Rac8 21. a4? …

“Whatever the reason, I completely missed the point of his move, thinking it was just routine development.” — Mike.

21. … f5!

Position after 21. … f5. White to move.

FEN: 2r1r1n1/p2q2bk/1p2b1pp/3npp2/PP2N1PN/3PB2P/3Q1PB1/1R3RK1 w – – 0 22

It’s a little hard for me to imagine overlooking Black’s 16. .. d5 and 21. … f5. They are the first moves that I would be looking at! Maybe Mike was just a little bit too caught up in his own plans and devices, and forgot to ask one of the most basic questions: “What does my opponent want to do?”

At this point White’s position already seems hopeless. Black threatens … f4 winning a piece, and White has no way to stop it. Or does he? In his moment of peril, Mike wakes up and starts playing amazing moves.

22. gf gf 23. f4! …

Position after 23. f4. Black to move.

FEN: 2r1r1n1/p2q2bk/1p2b2p/3npp2/PP2NP1N/3PB2P/3Q2B1/1R3RK1 b – – 0 23.

This move is the turning point of the whole game. “Frisco looked shocked when I played 23. f4, like a cat who discovers the mouse he thought he as toying with has suddenly grown claws and teeth,” Mike writes.

Yes, indeed! If Black grabs the piece with 23. … fe, then the mouse rears up and eats the queen with 24. Be4+ Kh8 25. Ng6+ Kh7 26. Nf8+ Kh8 27. Nxd7.

Whenever something unexpected and unsettling happens at the board, I think it’s important to take a pause to regroup. Re-focus on the position as if you were seeing it for the first time. You especially need to throw away any thoughts of how totally won your position was 1 move, 3 moves, 7 moves ago. Those thoughts will not help you now.

I don’t think that Rosario did this. He played a sort of knee-jerk response that tells me he is not really thinking about the position.

23. … Nxe3?

If I were Black, I would ask, “What are the pieces that White needed to make his sacrifice work?” The answer is the bishop on g2 and the knight on h4. There’s nothing I can do about the bishop — but there is certainly something I can do about the knight! It’s sitting there completely undefended. So if I attack that knight, he has two choices: either move the h4 knight and lose the e4 knight, or vice versa. Therefore the move I would look at is 23. … Qe7! (Also, 23. … Qd8 is equally good.)

Admittedly it’s not quite as simple as that. First, Black has to see that 24. Qf2? fails because the queen is overloaded, so Black can just win a piece with 24. … Nxe3. (That’s why it would have been a good idea to “save” this threat instead of playing it immediately.) Second, Black has to consider the piece sac 24. Ng5. But in my opinion, Black is doing all right after 24. … hg 25. fg Nxe3 26. Qxe3 Rf8. The computer (Rybka) backs me up on this and evaluates the position at +2 for Black. White just doesn’t have a good way of turning up the pressure; the move g6+ only helps Black because it makes h6 available for Black’s knight.

When I asked Mike about this, he wrote back: “I don’t think it is a clear cut win for Black after [25. … Qe7] 26. fe Qxh4 27. Nd6. White’s pieces are all coordinated and he is winning back the exchange, if he wants it.” Alas, I cannot share this rosy assessment. The thing he’s missing, I think, is 27. … Bxe5!, which not only threatens the knight on d6 but also plans to answer 28. Nxc8 with 28. … Nxe3! The knight is untouchable because of 29. Qxe3? Bd4. Instead White could play 28. Bf2, hoping to gain a tempo with the attack on Black’s queen, but 28. … Qf4 looks very effective to me, threatening mate. After 29. Qxf4 Nxf4 30. Nxe8 Rxe8 it’s hard for White to deal with the three threats of … Nxd3 and … Nxe2+ and the more prosaic … Nxg2. Black’s armada of minor pieces is overwhelming. I think that this is a clear-cut win.

Although the tactics are admittedly quite involved, the principle behind 23. … Qe7 is simple: LPDO. Loose pieces drop off. Especially when White’s attack depends so much on one particular loose piece.

After Black misses this opportunity, his game goes rapidly downhill… with some ingenious assistance from White.

24. Qxe3 Ne7 

Mike points out the amusing variation 24. … ef 25. Rxf4!! (seemingly walking into a trap) Bd4 25. Ng5+ hg 26. Rxd4. Rosario now threatens again to take on e4 … but for a second time, Mike ignores the threat.

25. fe! fe

If 25. … Bxe5 26. Kh1! Bd4 27. Qf4, White just keeps offering the knight, and it doesn’t appear as if it will ever be safe for Black to take it.

26. Be4+ Kh8?

The only hope was 26. … Kg8, but my money would still be on White.

Position after 26. … Kh8. White to move.

FEN: 2r1r2k/p2qn1b1/1p2b2p/4P3/PP2B2N/3PQ2P/8/1R3RK1 w – – 0 27

Now Mike adds one more log to the fire, and the blaze engulfs Black’s king.

27. Rf6! Nd5

In his annotations Mike gives several other variations, but they boil down to the same thing: White captures on h6 with check. A sample line is 27. … Bxf6 28. Qxh6+ Kg8 29. Bh7+ Kh8 30. ef Qd4+ 31. Kh1 Bd5+ 32. Be4. CHECK.

28. Rxh6+ Bxh6 29. Qh6+ Kg8 30. Kh1! Re7 31. Rg1+ Rg7 32. Rxg7+ Qxg7 33. Qxe6+ resigns

So, was it luck or skill? If it were one game, I might say luck. But Mike does this sort of thing so often that I think it’s skill. Maybe it even goes back to not taking anything that happens on the chessboard too seriously — neither the bad nor the good. If your position is so bad that it looks ridiculous, it frees you up to look at ridiculous ideas. In the end, maybe self-deprecating humor is the best weapon!

Print Friendly, PDF & Email

Leave a Comment

Previous post:

Next post: