The Fabulous 60s: McKelvie upsets Benko

This just in from Dr. Neil McKelvie (Chemistry Professor at CCNY and Chess Master)

Mark…I noticed that (a) there have been no comments on my Denker submission; BUT (b) if you look up “Neil McKelvie” on GOOGLE, which I just did out of curiosity, I note that the first three entries – meaning most often accessed – are for me. (The next ones: I am not the principal of a religious English school in Yorkshire, and I do not play drums in a NZ rock band!) No 3 is for your BLOG. I have received no comments – have you?

MG Note: New Zealand (NZ) is a fantastic place, every chess player should visit it. The most recent NIC magazine has a story about the Queenstown, NZ Open organized by GM Chandler.  As Dr. McKelvie points out, in Auckland, NZ there happens to be MacKelvie Street but it’s listed as McKelvie Street.

McKelvie on Benko

Now: Pal Benko! I played him twice in MCC championships, and once in a US Open in Boston; but several times in Rapids (once coming in second to Bobby Fischer…7-0 I think was HIS score – ahead of Bisguier and Benko) This game is similar to the Denker game in that I played a highly speculative and probably unsound improvised gambit. *I* think that the most interesting Chess often comes from doubtful moves that no decent Computer would ever play! (Benko scored 7-0 the next year, ahead of Bisguier 5 1/2 – 1 1/2 and me 5-2)

McKelvie – GM Pal Benko Manhattan Club Championship – date 1966?

MG Note to readers: The Manhattan CC moved all over Manhattan, including a stint at the world famous Carnegie Hall at 57th and 7th Avenue.  This game was played before that venue.  Notes in the body of the game are by MG with Rybka kibitzing… see next section for McKelvie’s notes.

1. e4   c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3  e6 6. f4 a6 7. Be3 b5

As in many games, for example SM Bill Kelleher – M. Ginsburg, New England, 1980s (possibly early 90s).

Old Theory

8. e5!? Just as Kelleher played!  Theory presumes this to be premature but play gets very sharp.  It’s odd to see super sharp McKelvie openings because when I (MG) played him he reacted very passively in a QGD MCC Ch. 1985.  Maybe decaf that day?

In the 1970s, this type of structure was covered in a Scheveningen textbook.  Let’s see it:

What we had to work with in the 1970s

However this 8. e5!? lunge was little covered.  I was certainly shocked when Kelleher tried it against me.

8….dxe5 9. fxe5 Nd5 Just for completeness, 9….b4!? TN 10. exf6 bxc3 11. fxg7 Bxg7 12. bxc3 Qc7 is a small edge for white – thus playable.

10. Nd5 Qxd5 11. Be2 Still following the Kelleher game.  I don’t have that game score handy….(I won after insane complications).  The bizarre computer choice 11. Nf3 retains equality.

11…Qxe5 I  believe that I, too, accepted this pawn because it’s hard to see what else black can do.

12.Qd2    Bb7 A very important position for the theory of this line has been reached.  Interesting, Rybka judges white has almost equal chances.  Black has one narrow way (see next note) to get something.  As McK mentions in his notes below, 12…Bc5! is a good alternative here and Rybka agrees.

13.Bf4    Qd5(? – McK) The best, not easy to see at the board, is 13…Qc5! 14. O-O-O Be7 15. Nb3 Qc8! 16. Bd6 Qd8! 17. Nc5 Bd5! and black has a small plus.

14. O-O-O! A wild continuation hanging a2.  However in the end this turns out to be justified. Rybka mentions 14. Bf3 Qd7 15. Rf1!? with compensation.    It also gives an inhuman line 14. Bf3 Qd7 15. Qc3 Bxf3 16. gxf3!?, also with good compensation.

14…Qd7  (? – Rybka) Benko blinks first, makes a move that doesn’t contribute to development, and he lands in a lost game!  But starting here we have a fascinating battle of the chess engines.  It would be interesting to turn even more engines loose on this one.

Naturally Rybka 2.2 doesn’t like this game choice and recommends 14…Qxa2 15. Nb3 Be7 (forced) 16. Bd6 Bf6 17. Be5! O-O (17…Be7?! 18. Bxg7 is good for white after 18…Rg8 19. Qh6)  18. Bxf6 gxf6 19. Qf4 (or 19. Qh6) – thus far McK analysis- and now a truly amazing defense, 19…Nc6!! 20. Qxf6 Na5 and white has nothing better than a perpetual with queen checks on g5 and f6.  Incredible.   But hang on to your hats.  Rybka 3.1 has seen deeper!  19…Nc6 20. Qxf6 Na5 21. Nc5 Rac8 22. Rd4! and wins!  Thus we have to revise our opinion and say black should not grab on a2!

Rybka 3.1 indicates black should not grab on a2 just yet, develop with 14…Be7, but then 15. Bf3 Qxa2 16. Qc3! causes obvious problems.  Is there any defense at all?  Let’s take a look at this position; the resource it finds for black is truly amazing.

Position after 16. Qc3 (Analysis)

Readers:  A) What should black play from the diagram position above?  B) What’s the correct evaluation with best play for both sides?

15. Qc3! Now black has a horrible game in all lines.

15…Bd5? This makes it worse. 15…b4, while very lame, was the best chance.

16. Nf5! A real cruncher.  Black is dead lost.

16…Nc6 17. Rxd5! exd5  18. Bg4!  Kd8 What a depressing move to have to play. In fact, black could have resigned – see the note to white’s next move.

19. Nd4 (“!” – McK, “?”   – Rybka)

Rybka hates this move because of what’s out there.  Indeed, one of Rybka’s juicy moves, 19. Ne7!!, forces resignation after 19…Qxg4 20. Nxc6+.  Even worse, if that is possible, is 19…Qxe7 20. Qxc6 with utter destruction.   For the sadists in the audience, 19. Nh6!! is just as effective.  For example, 19. Nh6!! Nb4 20. Bg5+ Be7 21. Nxf7+ and it’s +13.95 in computer speak!

This just in from McKelvie:  “Just incidentally….I DID intend Ne7, which of course wins easily, but then picked Nd4, which wins a piece and ALSO wins easily. Why? After Ne7 Black can play B:e7 and then K:d7, with R+N for Q and dead lost, but at least developed and able to survive for a while. After Nd4 Black is still with a useless R and unmoved B. The way I played SHOULD have led to immediate resignation after Qe1/e3 instead of Re1…now THAT was careless of me, or perhaps I wanted to enjoy winning against Benko a bit longer!

I suspect Rybka cannot understand failing to win Q for two pieces instead of just winning a piece, unless I have missed some amazing defence after my Nd4.   Cheers – Neil McKelvie”

19…Nb4! Black doesn’t have to be asked twice to do this. He’s now at only -1.2; if white had done 19. Ne7 it would have been -5 in computer-speak.

20. Kb1 Qb7 20. Rc8 21. Qh3 also loses: 21…Qb7 22. Bg5+! Kc7 23. Qc3+ Kb8 24. Bf4+ and wins the rook.

21. a3 h5 22.  Bh3 a5 23. ab Ra6   24. Nxb5 axb5 25. Bc7+ Ke8  26. Re1+ Re6   27.B:e6   fe   28.Qh3    Rh6    29.R:e6+  Kf7  30.R:h6   gh   31.Qf5+   Kg8   32.Qe6+   Kh7   33.Qf7+   Bg7 34.Nd4    Qa7    35.Nf5    Qg1+ 36.Ka2    b3+  37.K:b3   Resigns

I will try to find the “counter-twin” Kelleher game.

Some notes by McKelvie

Some notes: 12….Bc5 looked good for Black, although after 13.O-O-O O-O (?! – Rybka)  (MG: Rybka likes 13….Bb7! first) 14.Bf3 Ra7 15.Bf4!? Qd4 16.Qd4 Bd4 17.Rd4 White has a little compensation with two Bishops…
13….Qc5 was much better than 13…Qd5. If 14…Qa2 15.Nb3  Be7 16.Bd6 Bf6 17.Be5 O-O(?)
18. Bf6 gf 19 Qh6 a5(?) 20. Bd3 f5 21.g4,,,,  (MG:  See game notes for a discussion of a preliminary computer try, 19…Nc6)
26. Qe3+ was quicker.

Cheers….Neil McKelvie

McKelvie Puzzle

One McK creation from MANY years ago…a Mate in Four (but the first move is fairly obvious).
White: Qh1; Kg2; Pg4; Nb4; Ne8 Black: Kd7 Pb7

9/21/09:  Neil sent in a correction, the above puzzle had a typo. Here is the right version.

White: Kg2; Qh1; N’s b5 and e8; P g4;

Black: Kd7; Pb7    White to play and Mate in 4.

Solution: 1.Ne8-c7
If 1….Ke7 2.Qh7+ If then 2…Kf6 3.Nd5+ and then mirror mates from 4.Qh5 or Qf5 Other moves are uninteresting. HOWEVER
If 1…..Kc6; some logic. Black’s possible second moves with the K are 2…Kb6; 2…Kc5 and 2…K back to d7.  For the Q to then mate in two more moves, it has to get to a3, d4, and e5 respectively. There is only one square from where all three can be reached: a1!
SO: 2.Qa1. But now; what if 2…Pb6. NOW, the K has three squares available: 3…c5 or d7 or b7. To mate then, the Q has to get to c3, e8…AND a8. There is only one square from which to reach all three:3.Qh8. Therefore: Z for Zugswang! Q from h1 -> a1; h8; and a8.

McKelvie on Celts, Irish, Scots

“Mc” and even “M’ ” are valid SCOTTISH (and Irish) abbreviations for “Mac”. For my family name, which comes from the whole area of northern Ireland, the islands to the north, and the Scottish land area to the east; south of Glasgow, “McKelvie is the Scottish spelling, and “McKelvey” is the Ulster spelling. We are supposedly all descendants of a chieftain named “Cielbach Mac Cielbach”, where the “C” turned sometimes into “K” and sometimes into “S” (the northern English name “Selby”) over 2000+ years.

Scots from the North ,”highlanders”, are invariably “Mac”. Lowland Scots, who originally came from Northern ireland anyway, are usually “Mc”. The ROMANS named the group from Northern Ireland the “Scotti”. They were in constant war with the O’Neill’s from the south of Ireland, and so pushed into the south of Scotland, then occupied by the Picts. The two groups united against the Roman invaders. Later a character called Kenneth MacAlpine had married a daughter of the Pictish King, and when he died he became the first king of a united Scotland, having had other claimants killed off. To this day the tall fair-haired Highlanders – descendants of the Picts? – look, think, and talk differently from the Lowlanders. The groups do not always get on well together.
So; the Northern Ireland conflict has a 2000+ year history.

MG Note:  Since I was/am a Philistine savage, previously I believed “Mc” was Irish and “Mac” was Scottish and that was that.  Clearly things are much “Highland mistier.”

Advertisement

Tags: Benko, Celts, Chandler, Denker, Irish, Manhattan Chess Club, McKelvie, McKelvie Street Auckland, New in Chess Magazine, New Zealand, Northern Ireland, Picts, Pritchett, Queensland, Queenstown NZ, Romans, Rybka 2.2, Rybka 3.1, Scots, Ulster