I have picked up the Latvian gambit once again lately and started studying the critical lines. Now there are enough critical lines to study in the Latvian, believe me, but the poisoned pawn variation is surely one of them.
1.e4 e5 2.Nf3 f5 3.Bc4 fxe4 4.Nxe5 Qg5 5.d4 Qxg2 6.Qh5+ g6 7.Bf7+ ( a critical move in white's attack ! ) Kd8 8.Bxg6 Qxh1+ 9.Ke2 (diagram).
Most authors indicate that black is completely lost in this line, but is seems to me that black is alive and kicking !
9...Qxc1 !!! condemned by many. Obviously, white can force a draw now with 10.Nf7+ Ke8 , but can white win ???
Let look first at 10.Nc3 ( diagram)
The other winning candidate is 10.Nd2 (diagram), but also here, I see that 10...Qxb2 brings white nothing more than a draw.
I have checked the far majority of the lines that previously were supposed to win for white, but it just seems black can escape with a draw !!
On the downside, it seems that a line that was believed to win for black is actually winning for white :1.e4 e5 2.Nf3 f5 3.Bc4 fxe4 4.Nxe5 d5 5.Qh5+ g6 6.Nxg6 hxg6 7.Qxh8 Kf7 8.Bb3 Bg7 9.Qh7 (diagram)
Does this restore the Latvian to full equality ?? Unfortunately not, as there are still other refutations to refute ! It just shows that there is so much to discover in gambit play.