Lev Polugaevsky was one of the greatest exponents of the Sicilian Defense, which he almost always played against 1.e4. After he was diagnosed with brain cancer, the Buenos Aires Sicilian (1994) tournament was hurriedly organized in his honor. The Open Sicilian [1.e4 c5 2.♘f3 (d6, ♘c6, or e6) 3.d4 cxd4 4.♘xd4] was obligatory. But no one played his beloved Polugaevsky Variation (1.e4 c5 2.♘f3 d6 3.d4 cxd4 4.♘xd4 ♘f6 5.♘c3 a6 6.♗g5 e6 7.f4 b5!?) in the tournament.

This had always saddened me. Why was the Polugaevsky so rarely played, then and now? I looked at it briefly with Stockfish 17, which assessed the variation as around +0.6. A significant advantage, but well within the drawing margin, at least if you have an engine by your side. I decided to take it upon myself to prove that the Polugaevsky was playable. I had previously done something similar with the Damiano Variation of Petroff's Defense (1.e4 e5 2.♘f3 ♘f6 3.♘xe5 ♘xe4!?), demonstrating that it was playable. https://www.chessgames.com/perl/che... That inspired both a book and a Chess Informant survey on that variation.

I started play in two sections of the German Open correspondence tournament on October 25, 2024. In my games as Black, I responded to both 1.e4 and 1.♘f3 with 1...c5!, and told my opponents that if allowed, I would play the Polugaevsky Variation. A good deal for them, since they would get a larger than normal advantage. CCM/FM Peter Roth accepted the challenge, as did CCM Michael Preussner - twice!

The theoretically critical response to the Polugaevsky is 8.e5 dxe5 9.dxe5 ♕c7! 10.exf6 ♕e5+ 11.Be2 ♕xg5 12.O-O ♖a7! 13.♕d3! ♖d7! 14.♘e4! ♕e5 15.♘f3! Opening Explorer. Not surprisingly, my opponents played this against me in all three games.

After 15.♘f3!, Black has essentially three possible ways to play:

(a) Most people play 15...♕xb2, including GM Jeffery Xiong. E.g., T Kantans vs Xiong, 2017. Unfortunately, Stockfish 17 says it's refuted by 16.♕e3 ♗b7 17.♖ab1 ♕xa2 18.♔h1 ♗xe4 19.♕xe4 ♕d5 20.♕f4, making it an untenable choice in correspondence.

(b) 15...♖xd3 16.♘xe5 ♖d4! 17.fxg7 ♗xg7 18.♘xf7 O-O 19.♘fg5 ♖xf1+ 20.♗xf1 h6 21.c3 ♖d8 22.♘h3 ♗e5 23.♘hf2 ♗c7 24.♗e2 ♗b7 25.♖d1 ♖xd1+ 26.♗xd1 ♘d7 27.g3 ♔f7. Stockfish 17, depth 70, gives this as +0.73.

I really wanted to play line b, because I love responding to ♘xf7 with O-O à la the immortal game F A Hoffmann vs A Petrov, 1844. I can recall answering ♘xf7 or ...♘xf2 with O-O! only four times in my life, including recently in F Rhine vs NN, 2024. However, I really didn't like the resultant position with its evaluation of +0.73. I might actually lose, and perhaps even in three games! So I played line c in all of the games.

The games mirrored each other for some time, not surprisingly since play is very forcing. Roth played 24.a3, as did Preussner in this game. I drew both of these games. P Roth vs F Rhine, 2024 ; M Preussner vs F Rhine, 2024. Preussner deviated with 24.♖fd1 in our other game. I drew that one, as well. M Preussner vs F Rhine, 2024.