Nevr again...

[Event "(CORR-92)devon vs. glos. (ward-higgs I)"]
[Site "devon vs. glos. (ward-higgs I"]
[Date "1993.??.??"]
[Round "?"]
[White "Regis, David"]
[Black "Dixon, Richard "]
[Result "1-0"]
[ECO "A36"]
[Annotator "Regis,Dave"]
[PlyCount "83"]
1. c4 g6 {( I knew a game of portisch\'s which I had in mind throughout: it
went ... )} (1... c5 2. Nf3 g6 3. e4 Nc6 4. d4 cxd4 5. Nxd4 Nf6 6. Nc3 Nxd4 7.
Qxd4 d6 8. Bg5 Bg7 9. Qd2 O-O 10. Bd3 a5 11. O-O a4 12. Rac1 Be6 13. Qc2 Nd7
14. f4 (14. Nxa4 Qa5 15. Bxe7 Rfe8) 14... Rc8 15. b3 axb3 16. axb3 Nf6 17. Kh1
Qa5 18. f5 Bd7 19. Nd5 Qd8 20. Qf2 Bc6 21. Qh4 Bxd5 22. exd5 Re8 23. Rf3 Nd7
24. Rcf1 Bf6 25. Rh3 Nf8 26. fxg6 fxg6 27. Bxg6 hxg6 28. Rxf6 {1-0 Portisch,
L-Reshevsky,S/Petropolis Interzonal 1973/IZT} (28. Rxf6 exf6 29. Qh8+ Kf7 30.
Rh7+ Nxh7 31. Qxh7+ Kf8 32. Bh6#)) 2. g3 Bg7 3. Bg2 Nf6 4. Nc3 O-O 5. e4 {(
new for me this season: I played the e3/Nge2/d4 plan for years but recently
fancied this, which looked more enterprising )} d6 6. Nge2 c5 {( black had a
lot of choice at this point: ...e5/...c6/etc. )} (6... c5 7. a3 a5 8. O-O Nc6
9. d3 Ne8 10. Be3 Nd4 11. Bxd4 cxd4 12. Nb5 Qb6 13. a4 Nc7 14. f4 Na6 15. h3 e5
16. f5 Bh6 {!} 17. h4 Bd7 18. Kh2 Nc5 19. Bh3 {was seirawan-vukic 1979 given
in povah\'s book: best play for both sides?}) 7. O-O Nc6 8. d3 Ne8 (8... Bg4 9.
f3 Bd7 10. Be3 a6 11. Qd2 Re8 12. g4 Bxg4 13. fxg4 Nxg4 14. Bg5 h6 15. Bf4 e5
16. Bg3 Nd4 17. Nd5 b5 18. Bh3 f5 19. exf5 gxf5 20. Nxd4 cxd4 21. Rxf5 Ne3 22.
Nxe3 dxe3 23. Qxe3) 9. Be3 (9. Be3 Nd4) 9... Nc7 {( ?! in the books )} (9...
Nd4 10. Rb1 b6 {?! ( 10...Nc7 is natural; 10...a5 invites 11 Bxd4, cxd4; 12
Nb5 )} 11. e5 Nxe2+ 12. Qxe2 Rb8 13. d4 cxd4 14. Bxd4 dxe5 15. Bxe5 Bg4 16. Qe3
Bxe5 17. Qxe5 Qd6 18. Qe3 Nf6 {(gurevich-sturua 1981)} 19. h3 Be6 20. Rbd1 Qc5
{( or ...Qc7,b3 with Nb5/Nd5 )} 21. Qxc5 bxc5 22. b3 {with weak pawns to go at
- povah}) 10. d4 {! ( in the books )} cxd4 11. Nxd4 Ne6 12. Nde2 Ne5 {! TN -
threatens c4 and e3 via g4} (12... Nc5 13. Rc1 Be6 14. b3 Qa5 15. Qd2 Rfc8 16.
Nf4 $18 Bd7 17. Rfd1 Ne5 18. b4 Qxb4 19. Nfd5 Qa5 20. Nxe7+ Kh8 21. Nxc8 Nxc4
22. Qe2 Nxe3 23. Qxe3 Rxc8 24. Rxd6 Be6 25. Rdd1 b6 26. e5 Nd7 27. f4 Bf8 28.
Ne4 Rxc1 29. Qxc1 Bb4 30. Ng5 Nc5 31. Nxe6 Nxe6 32. Qc4 b5 33. Qd5 Kg7 34. Kh1
h5 35. Rf1 {1-0 Barcza,G-Szilagyi,P/HUN-ch 1967/EXT 2000}) 13. b3 (13. b3 Ng4
14. Bc1 a6 15. h3 Nh6 16. Be3 b5 {( this I thought must be the logical plan
for black, to break up the bind )} 17. cxb5 axb5 18. Nxb5 Bxa1 19. Qxa1 g5 20.
e5) 13... b6 {( transposing to a hedgehog formation looks better than
szilagyi\'s setup )} 14. h3 {( stopping the ...Ng4 lines seemed sensible,
especially since last move they were possible. The knights now need to get out
of the way of the pawn roller )} Bb7 (14... Bb7 15. f4 {could be played: after}
Nc6 16. f5 Nc5 17. Rc1 {the hole on e5 is too glaring [} Ne5 {!} 18. Qc2) 15.
Rc1 {( looks slow and a bit aimless with hindsight )} Nc5 16. f4 {( now I
thought I was on the right lines: turn that space into attack, as in the
portisch game )} Ned7 17. g4 a5 {( I would have preferred ...a6 )} 18. Ng3 {(
I don\'t know where this is going yet, unless black can be obliged or tempted
to take on f5 eventually - if not the Ng3 and Bg2 are very passive )} Bc6 {( !
I now wanted the wN back on e2 )} 19. a4 {( this isn\'t really necessary [ 19...
a4; 20 b4 looked good ] but I thought was justified psychologically - what is
black\'s plan now ? the real challenge of course is to find white\'s plan! The
wQ is badly tied to b3 and d3 )} (19. Bd4 {( this was better I thought
afterwards )} e5 20. Be3 exf4 21. Bxf4 Ne5 22. Rc2 Ned3) 19... Rb8 20. Rf2 {(
Nicely ambiguous [ ideas of Qd2 or Rd2 ] but d3 is now more tender )} Qc8 21.
Nd5 {( I\'m obviously entertaining pushing f4-f5-f6 )} Re8 22. g5 {( not f5 yet
)} Qb7 {( adds to pressure but locks in Bc6 )} 23. f5 {( It\'s starting to get
hard for me to find a plan which doesn\'t involve f5 but Black can play ...e6
and build up nicely now. I had some fun with 23 f5 and 24 Nh5 in analysis
which was all bluff but looked worth starting )} (23. f5 {( doesn\'t threaten
f6 yet (...Bxd5; fxg7, Bxe4) but does stop ...e6 and I can start to build up
myself, I think )} Bf8 24. f6) 23... Bf8 {( I looked at ...Bxd5 and ...Be5 in
analysis but this looked most solid. If ...e6 white can afford ...exd5 because
exd5 wins the Bc6 )} 24. Bd4 {( 24 f6 simply takes a good square away from the
knight. Now I have more realistic-looking threats but I must say by this stage
I thought it was all starting to slip away )} Ne5 25. Rc3 Bg7 {( I now entered
a short sequence of moves which I thought were most solid and maintained a
space advantage despite exchanges )} 26. f6 Bf8 27. fxe7 Bxe7 28. h4 Bxd5 {(
the two bishops are no real advantage here as the Bg2 is inactive )} 29. exd5
Bf8 30. Rf4 {( else Qf1,Bg7 will threaten ...Nf3+ and ...Bxd4 )} Bg7 31. Qf1 {
( both sides have consolidated after some white initiative and I thought we\'re
now in for a phase of manoeuvring )} Re7 (31... Rf8 32. Bxc5 bxc5 33. Ne4) 32.
Bxc5 (32. Rf6 {( did I say manoeuvring? this sacrifice I don\'t think can be
accepted but after declining by eg. ...Rbe8 I don\'t know where White is going )
} Bxf6 33. gxf6 Ree8 34. Qf4 Kh8 35. Qh6 Rg8 36. Bxc5 bxc5 37. Ne4 Ng4 38. Qf4
Qd7 39. Rg3 Ne5 40. Bh3 Qc7 41. Qh6) 32... bxc5 {( obviously not 32...dxc5; 33
d6 )} 33. Ne4 {( This move cheered me up a little; I felt I had some chances
for the initiative )} (33. Rf6 {( again 33...Rbd8 looks better and now 34 Ne4
doesn\'t even threaten d6 )} Bxf6 34. gxf6 Ree8 35. Qf4 Kh8 36. Qh6 Rg8 37. Ne4
{( is the same as on move #32 )}) 33... Rd8 34. Nf6+ (34. Nxc5 {( idea d6
again )} Qb6) 34... Bxf6 (34... Bxf6 {( I wondered if ...Kh8 was better. I
would have played ...Bxf6 myself, though, since all the endgame trumps are
Black\'s: N vs. bad B, 3 vs 2 on the K-side while Black\'s 3 hold White\'s 4 on
the Q-side. Still, after ...Bxf6 I explored:} 35. gxf6 Ree8 36. Re4 Qc7 {(
better might be ...Nd7 or ...Qb4 or ...h5 as game: I favoured ...Nd7)} 37. Qf4
Kh8 38. Qh6 Rg8 39. Rh3 Rde8 40. h5 g5 41. Rg3 g4 42. Rf4 Nd7 43. Be4 Rxe4 44.
Rxe4 Ne5 {...which is plausible if optimistic)}) 35. gxf6 Ree8 36. Re4 h5 {( I
admit I analysed this only after sending off my last move - I concluded I had
some chances of a combinative breakthrough on the K-side but this couldn\'t be
forced, and I became slightly less alarmed about possible endgames )} 37. Rce3
Kh7 38. Qf4 {( I couldn\'t see a way of winning after what I thought was the
best line, ...Qb4, but I didn\'t have to lose either. White can try to contrive
a lever with Bh3, Qg5 and now Bf5, but ...Qb4 & ...Rg8 (forces Kh1) slows or
stops it)} (38. Qf4 Qb4 39. Bf3 {( this line I examined most closely; there
are several good Q+R endings for White after the exchange but the wK is
terribly exposed )} Qd2 40. Be2 {( this tack peters out because of the active
bQ and weak Pb3 )} Rb8 41. Qg3 {( threat Bxh5 )} Qc1+ 42. Kg2 Qc2 43. Kh3 Rxb3)
38... Qc8 {( I thought at the time, not as strong as ...Qb4, but realised it\'s
trickier )} (38... Qc8 39. Bh3 {( and now the crucial line, which I thought
would be played, is:- )} Ng4 {( I analysed a win after this. As 39...Nd7 fails
to 40 Re7 I think Black should admit the error by 39...Qb8 with fair chances
of holding the blockade )} 40. Re7 {( I spent a long time looking at 40 Bxg4
to get rid of the dud bishop and dangerous knight but 40...Rxe4 is an annoying
intermezzo; 40 Rxe8,Rxe8; 41 Re7 I decided wasn\'t as strong as the game: 41...
Rf8; 42 Qxd6 allows 42...Qf5 )} Rf8 {( Black could let the pawn go but ...Kg8
risks mate after Bxg4 and Qh6 )} 41. R3e6 Qb8 {( 41...Rd7; 42 Bxg4, hxg4; 43
h5, Rxe7; 44 fxe7 wins but 43...Qc7; 44 hxg6++ )} 42. Bxg4 hxg4 43. h5 {! ( 43.
..gxh5; 44 Qg5 mates and others allow 44 hxg6+ with mating attack; 43...Rg8
abandons the Pf7 )} Qxb3 44. hxg6+ Kxg6 45. Qxg4+ Kh6 46. Qg7+ Kh5 47. Re5+
dxe5 48. Rxe5+ Kh4 49. Re4+ Kh5 50. Qg4+ Kh6 51. Qh4+ Kg6 52. Rg4+ Kf5 53. Qg5#
) 39. Bh3 (39. Bh3 Ng4 40. Rxe8 Rxe8 41. Bxg4 {( white still can\'t do without
this, so it transposes )} hxg4 42. Re7 Rf8 43. Qxd6 Qf5 44. Re3 {!?} Qb1+ 45.
Kg2 Qc2+ 46. Kg3 Rb8 {( or ...Rc8 )}) 39... Qb8 {( if the knight moves don\'t
work this is probably best )} 40. Kh1 (40. Bf5 Rg8) 40... Qc7 (40... Rh8 {( I
was most interested in this line to stop Qh6 for once and for all: I analysed
optimistically... )} 41. Bf5 Kg8 42. Qg5 Qb4 43. Rg3 Rb8 {( not forced by any
means )} 44. Rxe5 dxe5 45. Bxg6) 41. Bf5 (41. Bf5 Rh8 42. Qg5 {( now ...Ng4
loses to Re7 )} Rde8 {( unfortunately there is now no mate or win of a rook
after 43 Rg3, Kg8; 44 Bxg6, fxg6; 45 Rxe5+= and in fact 43 Qxh5+ loses the
pawn straight back to 43 ...Kg8; 44 Qe2 (not Qg5,Rh5 -+) 44...Qd8; 45 Bh3,
Qxf6 )} 43. Rg3 Kg8 44. Bxg6 fxg6 45. Rxe5 Rxe5 {( keeping pawns intact for
the ending )} 46. Qxg6+ Kf8 47. Qg7+ Qxg7 48. fxg7+ Kg8 49. gxh8=B+ Kxh8 50.
Rg6 Re3 51. Rxd6 Rxb3 52. Rc6 Rh3+ 53. Kg2 Rxh4 54. Rxc5 Rd4 55. Rc6 h4 56. d6
Kg8 57. Rc8+ Kf7 58. Rc7+ Ke6 59. c5 Rxa4 60. Re7+ Kf5 61. d7 Ra2+ 62. Kh3 Rd2
63. c6 Rd3+ 64. Kxh4 Rd4+ 65. Kg3 Rd3+ 66. Kf2 Rd2+ 67. Ke3 Rd1 68. c7) 41...
Qb8 {( I immediately thought this was too resigned )} 42. Qg5 {( mates in six )
} (42. Qg5 {( I sent some analysis showing a win after ...)} Rh8 $1 {critical}
(42... Re6 43. dxe6 Rh8 44. exf7) (42... Qxb3 43. Qxh5+ (43. Rxb3 Rh8 44. Rxe5
dxe5 45. Bxg6+ fxg6 46. Rb7+ Rd7 47. Rxd7+ Kg8) 43... Kg8 44. Qh6 Qd1+ 45. Kg2
Qd2+ 46. Kh3 Qh2+ 47. Kxh2) 43. Rg3 $5 {( 43 Rxe5 is more forcing and does
mate in 5; after this Fritz threw in ...Qc8 which stops the mate! )} (43. Rxe5
$1 dxe5 (43... Kg8 44. Bxg6 Kf8 45. Bxf7 dxe5 46. Qg7#) 44. Rg3 Kg8 45. Bxg6
Kf8 46. Bxf7 Kxf7 47. Qg7+ Ke8 48. Qe7# $10) 43... Qc7 (43... Kg8 44. Rxe5 dxe5
) 44. Rxe5 dxe5 45. Bxg6+ Kg8 {( 45...fxg6; 46 Qxg6 # )} 46. Bf5+ {!} Kf8 47.
Qg7+ Ke8 48. Qxh8#) 1-0