More Cute Stalemates
by Bruce Leverett

In the September issue of En Passant, Frank Cunliffe gave the final moves
of a game I played with Andrew Soltis, leading up to a stalemate trap.
Frank was quoting from Soltis' recent autobiography, Confessions of a
Grandmaster.  This game, played in the 1982 Marshall International in New York,
was as memorable for me as it was for Soltis, for it was the first game I ever
played against a GM that I didn't lose (except for a 1978 game against a much
older player).  Prompted by the publication of Soltis' book, I re-analyzed the
endgame, and I found that some of the cutest lines didn't even appear on the
board.  I hope En Passant readers will forgive me for rehashing this game,
as I will show the results of my analysis.

diagram:
  W Kf4 Ne4 Pa2,b3,c4,g4,h3
  B Ke6 Bc6 Pa5,b4,c5,d6,h6

The game continued 51 ... a4 52 h4 a3! 53 h5 Be8 54 g5 hxg5+ 55 Nxg5+ Ke7
56 h6 Bg6 57 Ne4 Kf8.  Now 58 Nd2 is an interesting defensive try, but my
subject is cute stalemates, so I'll have to leave the reader to work out for
himself whether it draws or not.  I played 58 Ke3 Kg8 and now, as Soltis
pointed out in the port mortem, 59 Nf6+! is a good idea.  If 59 ... Kh8,
Black's king is trapped in the corner, so White draws by 60 Kd2 Bb1 61 Kc1
(and now 61 ... Bxa2 even loses after 62 Kc2).  Also 59 ... Kf8 60 Ne4 only
repeats the position.  Black can triangulate with 59 ... Kf7 60 Ne4 Kf8; 
White can then try 61 Nd2 or 61 Nxd6.  The latter move enters the game
continuation with an extra tempo for White--probably unimportant as we shall
see.

I played 59 Nxd6 and now an apparently desperate situation was reached after
59 ... Bc2! 60 Ne4 Bxb3! 61 Nxc5.

diagram:
  W Ke3 Nc5 Pa2,c4,h6
  B Kg8 Bb3 Pa3,b4

Black has his choice of two tempting pawns.  It seems that either one of them
would be enough to win but it is at this point that stalemate comes into play.

Soltis rejected 61 ... Bxc4 because of the variation 62 Kd2 b3 63 Nxb3 Bxb3
64 Kc3! followed by 65 Kb4.  He suggests that 62 ... Bxa2 would have won, but
watch:  63 Nd3! b3 64 Kc1 Kh7 65 Nb2!

diagram:
  W Kc1 Nb2 Ph6
  B Kh7 Ba2 Pa3,b3

The threat of 66 Nc4 compels Black to capture, but 66 ... axb2+ 67 Kxb2 is a
draw.  If Black's king comes too close it's stalemate.  If you don't believe
me, try it yourself!

Soltis played 61 ... Bxa2 and I replied 62 Kd3.  This isn't fatal by itself,
but it shows that White hasn't found the correct drawing plan.  White should
play 62 Kd4.  Once again the threat of 63 Nd3 is strong.  If 62 ... Kh7
63 Nd3 b3 64 Kc3 Kxh6

diagram:
  W Kc3 Nd3 Pc4
  B Kh6 Ba2 Pa3,b3

This looks similar to the position with White's king at c1, but here 65 Nb2
is a waste of time:  65 ... Bb1! 66 Kxb3? a2 and Black wins.  In this variation
moves like 66 Nd1 may still draw, but the simplest is 65 c5.  Black will have
to sacrifice both pawns to free his bishop.

Black can also try 62 ... Bb1.  After 63 Nb3 Bc2 64 Na1 b3 65 Nxb3 Bxb3 66 Kc3,

diagram:
  W Kc3 Pc4,h6
  B Kg8 Bb3 Pa3

Black's choices are:  66 ... a2? 67 Kb2 draw; 66 ... Ba4 67 c5, luring the
bishop away so that the king can gobble the a-pawn; or 66 ... Bxc4!  Now there
is a curious repetition:  67 Kc2! Ba2 68 Kc3! Bc4 (or Bd5, or Be6, etc.)
69 Kc2 Ba2, and Black has made no progress.  One final trick is 67 ... Bd3+,
but 68 Kb3 (instead of 68 Kxd3) is sufficient.

Back to the game.  After 62 Kd3 Kh7 I still had time for 63 Kd4, but I had not
the slightest inkling of any of these variations.  Thinking I was in zugzwang,
I played what should have been the fatal blunder:  63 Kc2?  After 63 ... Bxc4
64 Kb1 Kxh6 65 Ka1 Soltis returned the favor with 65 ... Kg5? (65 ... Bd5!
wins).  The game ended 66 Ne4+ Kf4 67 Nd2 Be6 (on 67 ... Bd3 68 Nb3 White
can establish a blockade) 68 Nb1! a2 69 Nc3!

diagram:
  W Ka1 Nc3
  B Kf4 Be6 Pa2,b4

After playing the endgame like a fish, I redeemed myself by finding this
line.  A fitting conclusion would have been 69 ... bxc3 stalemate.
Anticlimactically the game ended with 69 ... Ke3 70 Nxa2 Bxa2 71 Kxa2 Kd4
and a draw was agreed.