Keres - Alexander 









Paul Keres (2675) - C.H.O.D. Alexander (2500) 
[E21]
Margate,  Great Britain.  (ENG)   1937  

[A.J. Goldsby I]


Chernev writes: 
"Keres, like Alekhine, can pack more combinative action into one game than 
some Masters do in a lifetime of play. Watch the sparks fly in this 20 move sizzler!" 

[ See the book, {The} "1000 Best Short Games of Chess,"  by the late, 
great Irving Chernev.  Game # 738, pages 385-386. ] 

A very, very brilliant game of chess.  Keres definitely packs some big tactical 
punches into a very short span of time. A real tactical melee. 

Brilliance by Keres of the very First Order.


1. d4 Nf62. c4 e63. Nc3 Bb4; {Diagram?} 
"The 'Nimzo-Indian' Defense."
  

4. Nf3!?
,  {Diagram?} 
Simple logical development. Kasparov would later champion this line.

[ White could have also played: 4.Qc2, which is: "The Classical Variation." 
See the game Capablanca - Mattison; for the move, 4. Qc2. 
(And a complete analysis of that line in the Nimzo-Indian.) 
Or  4.e3,  which is, "The Rubinstein Variation." (The most popular.) ].  

4...b6!?
, {Diagram?}
Fianchettoing the QB is a common idea in this variation, but it is 
not clear which line is the best for Black.

  [ The book line is: 4...d55.Bg5 Nbd76.cxd5 exd57.e3 c5
     8.Bd3
Qa59.Qc2 c410.Bf5 0-011.0-0, ("=") Etc. 

(See the diagram just below.)

  Analysis position. The book line after 11. 0-0.

The position is close to equal, or White may have a very small edge. ("+/=") ].  

5. g3
,  (Maybe - '!?")  {Diagram?}
A standard way of meeting Black's fianchetto of his QB - White 
fianchetto's his own KB. 

{ MCO-14 gives the line:  5.Bg5!? h6;   6.Bh4 Bb77.e3 g5
 (7...Bxc3+!?~; is col. # 62.) 

8.Bg3
Ne49.Qc2 Bxc3+; 10.bxc3 d6; 11.Bd3 Nxg3;  The end of column # 61, 
page # 553.  12.fxg3(12.hxg3!? Nd7; ("=") This looks "+/=" to me. {A.J.G.})  
12...Nd713.0-0 Qe714.Rf2 0-0-0!?15.Be4 Rhf8!16.Ne1 Bxe4
17.Qxe4
Kb818.Qh7!?,   (18.Qc6!, ("+/=") - {A.J.G.}  Or 18.Rb1!?;  
or 18.Nc2!?~)
  18...Rh819.Qc2 h520.e4 h4; ("=") 

(See the diagram just below.) 
  Analysis position, of the MCO line, after 20...h4.

"The chances are even." - GM N. DeFirmian in MCO.  

Zvjaginsev - J. Timman; Biel, 1995. 
[ See MCO-14;  pages # 553-554, columns # 61-66; 
(Mainly column # 61 here.) and note (d.). ] }

5...Bb7
6. Bg2 Qc8!?;  {Diagram?}
Black defends his Bishop on b7 and prepares the nice break, 
Pawn@c7 to c5. 

But ... I do not like this move. It is artificial looking and sets Black up for 
a fork on d6. It also weakens the already critical dark squares, especially d6. 

[ Probably best is: 6...0-0!7.0-0 c5!;  with close to equality for Black. ]. 

7. 0-0
,  (Maybe - '!')  {Diagram?} 
This is the best, as it frees the Knight on c3 from the pin. 

[ Interesting is: 7.Bf4!? ].  

7...c5?!
;  (Maybe - '?')   (This move has definite drawbacks.) 
This is dead wrong. Black further exacerbates his dark-square problem. 

[Much better was: 7...Bxc38.bxc3 0-09.Bf4, ("=" or "+/=") ].  

8. Nb5! cxd4[]
;  {Diagram?} 
This is 100% forced now.  
(Black needs to [un-cover] the protection to d6.) 

[ Not 8...Qc6?9.d5! exd510.Ne5!,  ("+/-") ].  

9. Bf4!,  {Diagram?} 
Very thematic. 

Chernev writes: "Now threatening a check at B7, (c7)." 

"Spearing the dark squares, and threatening Nc7+ - winning an exchange." 
 - LM A.J. Goldsby I. 

[ 9.a3!? a6!~] .

9...Na6
;  {Diagram?} 
This is forced to prevent Nc7. 

[ 9...a6??10.Nc7+ Kf811.Nxa8, ("+/-") ] 

10. Bd6!
,  {Diagram?} 
A very good move.  

Chernev writes: 
"Intending 11. BxB/b4, NxB/b4; 12. N-Q6ch (Nd6+), 
winning the (Black) Queen." 

[ Also good for White is: 10.Qxd4!?, ("+/=") when White is clearly better. ].  

10...Qxc4
;  {Diagram?} 
Seems to be the only defense for Black.

[ Inferior are:  10...Ne4?!11.Bxb4! Nxb412.Nd2!, ("+/-") is winning for White. 
Also bad for Black is: 10...h5??11.Bxb4 Nxb4?!12.Nd6+, ("+/-") 
which wins the Black Queen. 
A mistake is:  10...Bxd6??11.Nxd6+,  which also wins the Black Queen. ("+/-") 
Interesting is:  10...Bc5!?11.e3!, ("+/=") 
And White is better. (This line is CLEARLY better for White, yet it may 
not be much worse than what happened in the game.) ].  

11. Qa4!
,  {Diagram?} 
Clearly the best. (It takes the computer a few seconds to find this move.) 

Chernev writes: 
"Now he plans: 
a.) 12. P-QR3 (a3) attacking the pinned Bishop; 
b.) 12. R-B1 (Rc1) driving the Queen away from the Bishop; and 
c.) 12. Kt-K5 (Ne5) {attacking the Queen and the other Bishop} 12...Q-B1; (Qc8) 
13. BxB/Kt4 (Bxb4), winning." 
Chernev continues:
"Keres, as we see, is a triple-threat man." 

11. Qa4! is easily the best move. White has multiple threats. 
The end is probably near for Black. 

[ Also interesting is 11.a4!?  ("+/-") ].  

11...Bc6!?
;  (Maybe - '?!')  {Diagram?} 
Chernev makes no comment on this move, 
but it looks like a possible error. 
(The computer evaluations jump quite a bit after this move.) 

[Maybe Black had to try: 11...Qc6!?;   but White can respond 
with 12. BxB/b4, winning a piece. ("+/-") ].  

12. Nfxd4 Bxg2
13. Qxa6!,  (Maybe only - '!?')  {Diagram?} 
Definitely interesting. 

It is hard to be sure, but this may not be the best move. 

Chernev writes:  "Menacing the Queen by discovered attack." 

[ Not 13.Kxg2? Qd5+14.Nf3 Bxd615.Rad1 Nc5!; and the game is close to equal. 
( 16.Qa3!?, "+/=" );  

The computer gives as best:  13.Rfc1!?, (!) 13...Qxd4; What else can Black do? 
(13...Nc514.Bxc5 Qxc515.Rxc5, "+/-")  14.Nxd4 Bxd615.Qxa6, "+/-" 
and Black can resign. (This seems to be easier than the actual game continuation.) ].  

13...Bxf1
;  {Diagram?} 
This looks completely forced. 

[ No better for Black was: 13...Qc8!?; ('?!')  14.Qxc8+! Rxc815.Kxg2 Bc5
16.Nxa7!
Ra817.Ndb5! Bxd6  18.Nxd6+ Ke719.Nab5, "+/-"  and White - 
a piece up - is easily winning. ].  

14. Rxf1!
,  {Diagram?} 
A nice, cool, calm recapture. 

[Not 14.Nc7+?! Qxc715.Bxc7 Bh3;  which allows Black counterplay. ].  

14...Qxd4
; {Diagram?} 
Equal to capitulation. (To me, anyway.) 

Chernev writes:  "Not desperation, but there is no plausible defense. 
If 14...Qc8; 15.QxQ/c8+ wins a piece. Or if 14...Kd8; 15. b2-b3 does likewise, 
while 14...Rc8; permits 15. BxB/b4." 

[ If 14...Qc8!?; 15.Qxc8+ Rxc816.Bxb4 Ne417.f3 Rc418.Ba3!, "+/-" 
White is winning. If 14...Nd515.e4!,  ("+/") White is clearly better, 
if not winning. (Maybe - "+/-"); 

Or 14...Kd8 ; 15.Bxb4!, This must be best.  
 (Chernev recommends (the vastly inferior) instead: 15.b3!? Qc816.Qxc8+ Rxc8;
17.Bxb4, "+/-"
  15...Qxb416.Qb7 Rc817.Nxa7 Qc418.Nxc8 Qxc8
19.Qxb6+
,  White is clearly much better. ("+/-" or "+/-"); 
If 14...Rc8?; 15. b3!, "+/-"  White is winning easily. (The Black Queen has no 
viable squares to protect the Black Rook on c8. I.e., 15...Qc1!?; 16. BxB/b4! ) 
 Chernev instead recommends (after 14...Rc8?;) that Black play the move, Bishop captures Bishop/b4, i.e. 15.Bxb4 Kd8;   (15...Qxb4??; 16.Qxc8+)  16.Qa3, "+/-"  ].

15. Nxd4 Bxd6
16. Nb5!,  {Diagram?} 
The sharpest and best. 

[ Not  16.Qb7? 0-017.Nb5, (+/=")  which allows most of White's advantage 
to evaporate.  Maybe 16.Rc1!?, "+/-" ]. 

16...Ke7
;  {Diagram?} 
Not much choice here. 

Chernev writes: 
"What else? If 16...Bb8; (to guard the c7-square) then 17. Qb7, 
  catches the Rook, or if 16...Be5;  then 17. f4,  drives the Bishop 
  off the diagonal."  

17. Rd1 Bc5;  {Diagram?} 
The best?  

"The alternatives allow White to win by the tactic of exchanging, 
then pinning." - Chernev. (See the 2 variations below.) 

[Chernev offers the variations: 
Var. # 1.)  17...Ne8?18.Nxd6 Nxd619.Qa3, "+/-" or ... 
Var. # 2.)
  17...Nd5?18.Nxd6 Kxd619.e4, "+/-".]  

18. a3!?,  (Maybe - '?!')  {Diagram?} 
White prepares to kick the Bishop. 

[Maybe the best is: 18.e3! Rhc8; 19.a3, "+/" White is better. (Maybe - "+/-") ].  

18...Ne4!?;  (Maybe - '?!')  {Diagram?} 
No comment from Chernev, but this move has got to be a major 
oversight or error. 
(At the very least, Black should have tried the BxPch move.) 
This move looks strong, with its double attack on f2, but misses 
White's response. 

[ Black had to try: 18...Bxf2+!!;  a real shot by Black.  19.Kg2, (!) Looks forced. 
 (NOT 19.Kxf2?, ('??')  19...Ne4+20.Ke3 Nc5; White's Queen is now trapped. 
21.Qxa7[]
Rxa7
22.Nxa7 Ra8;  "=/+" 
  19...Rhd8!20.Kf3!,  "+/" or "+/-" 
White ... after navigating an incredible minefield ... is clearly better, if not winning. ].

19. Rxd7+!
,   {Diagram?} 
A very nice move, and a nice example of a "pseudo-sacrifice." 

[ Also good was: 19.e3!?, "+/-" ].

19...Kf6
; {Diagram?} 
This is forced. 

[ Chernev writes: "If 19...Kxd7?!; ('?')  20.Qb7+ Ke821.Qxa8+,
followed by removing the other Rook. ("+/-") ].  

20. Qb7
,  (Maybe - '!')   Black Resigns.  1-0 

[ Chernev gives the following variation:  ( 20.Qb7, and now)  and says: 
"If 20...Nxf221.b4 Be322.Qf3+, ("+/-")   makes further resistance futile." ] 

***

An incredible game of chess, but Alexander did not exploit all of his 
opportunities to the maximum extent. 

(Keres may have missed the best move,  and  Chernev was  NOT  in 
his usual flawless form.) 

Easily in the - 
<< "Top 100" Most Brilliant Short {Miniature} Games Ever Played. >> 
(But perhaps not quite accurate enough to be in the Top 10.) 

But a very pretty game of chess, none-the-less!!

1 - 0

(I was still a probably a pre-teen-ager when I borrowed a book from a friend.
It was a book of Keres' games.
Many of his games impressed me mightily. This game was one of them.
I am truly happy to finally be able to bring you this game.)


This game is the full length version of the game as it exists in my database.
 (I have not shortened it for publication.) 
 If you would like a copy of that game (in ChessBase format) to study, please contact me. 


Click  HERE  to return to my GeoCities  "Home Page." 

 Click  HERE  to return to my (GeoCities)   "Best Short Games Page."  


 Copyright A.J. Goldsby I. © A.J. Goldsby, 1998 - 2005.
 © A.J. Goldsby, 2006. (
All rights reserved.)