In a recent post of PCG specialist American NM Jim West (
http://jimwestonchess.blogspot.com/ ) write after 5..fxe4 6.Nxe5 dxe5 7.Qh5+ Kd7 8.dxe5 second player should to play
8..Kc6 (!?) - instead of my previous 8..c6 - 9.Qe2 Bc5 10.Nc3 Kb7 11.Nxe4 Bb6 in the spirit of some lines on Latvian with the same variation where Black King has a possible bolt-hole on b7. Maybe it's true, but
11.a4!? seems stronger, but also I prefer
6.dxe5! or
6.Ng5!? with a powerful initiative as anybody can see it playing some moves with Black pieces ...
Paradoxically in West book "The Dynamic Philidor Countergambit" ( 2nd. ed., 1996 ) he doesn't analyze 1.e4 e5 2.Nf3 d6 3.d4 f5 4.Bc4
c6!? ( only 4..exd4; 4..fxe4; and 4..Nc6 ) 5.dxe5 fxe4 6.Ng5 d5 ( I know 6 games ) trasposing to above noted
Schild-Bullockus, cr ICCF, 1979 via LG: 1.e4 e5 2.Nf3 f5 3.Bc4 b5?! 4.Bb3! d6 5.d4 fxe4 6.Ng5 ( 6.dxe5! is even better ) 6..d5 7.dxe5 c6 ( see The Latvian gambit lives!, Kosten, 2001, page 167 ) and now contradicting English GM I think best move is
8.Nxe4 ( instead of 8.0-0 in four Khulmann-Criel games, or "logical" 8.e6 played by Schild ):
- 8..dxe4?? 9.Bf7+ Ke7 10.Bg5+ Nf6 11.Qxd8+ Kxd8 12.exf6
- 8..Qc7 9.Nd6+ Bxd6 10.exd6 Qxd6 "which is not too bad" according Kosten, but I personally prefer White after 11.0-0
- 8..Nd7 - relatively the best - 9.e6 Ndf6 ( 9..Ne5 10.f4 Ng6 11.Ng3 ) 10.Ng5 Nh6 11.0-0
Definetevely Lev Zilbermintz gambit ..b5?! seems refuted with the declined move
5.Bb3!