The diagrammed position is taken from game 19 of the match, following Black's 23rd move.
The opening moves were 1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 Bg4 5.Be2 e6 6.O-O Be7 7.h3 Bh5 8.c4 Nb6 9.Nc3 O-O 10.Be3 d5 11.c5 Bxf3 12.Bxf3 Nc4 13.b3 Nxe3 14.fxe3 b6 15.e4 c6 16.b4 bxc5 17.bxc5 Qa5 18.Nxd5! Bg5! 19.Bh5 cxd5 20.Bxf7+ Rxf7 21.Rxf7 Qd2! 22.Qxd2 Bxd2 23.Raf1 Nc6.
Now Spassky played 24.exd5.
Soltis cites the recent correspondence game Van der Gracht-Prins in which White played 24.Rc7! Nxd4 (In Both Sides of the Chessboard, Robert Byrne and Ivo Nei give 24...Nd8 as better) 25.Rff7 Be3+ 26.Kh1 Bh6 27.g4! "and demonstrated a forced win." Soltis analyzes 27...dxe4 28.g5 e3 29.Kg2! e2 30.Kf2 Bxg5 31.Rxg7+ Kf8 32.Rxh7 Kg8 33.Rcg7+ Kf8 34.Rxg5 which wins for White.
An improvement on 32...Kg8?, however, is 32...Bh4+!. Now 33.Rxh4? runs into 33...e1=Q+ 34.Kxe1 Nf3+ 35.Kf2 Nxh4 36.Rh7 Ng6. White is fortunate that, after 32...Bh4+!, he still has a draw with 33.Kg2 Kg8 34.Rcg7+ Kf8 35.Rd7 Kg8 36.Rdg7+ Kf8 37.Rf7+ Kg8, etc.
{This article originally appeared in Atlantic Chess News in 1988. My analysis was used by grandmaster Larry Evans in What's the Best Move? in the February 1989 issue of Chess Life except that my 35...Kg8 was unaccountably changed to 35...Ke8 with the unlikely continuation 36.Rdg7 Kf8 because 35...Ke8?? loses to 36.Rb7.}
