Karpova: This game was annotated by Adolf Zinkl in the 'Neue Freie Presse' of March 26, 1910.Although not annotated, 13...Nbd7 appears to be an inaccuracy, leading to a much more weakened Black ♔side, than 10.g4 did for White (a move Zinkl criticised). Black's desire to complete his development is understandable, but he could have prepared the move by 13...Be7.
White gives away his advantage with 17.Qd2, a move also criticised by Zinkl. He suggests <17.f4> instead (followed by Ng3 and Nf5), but this doesn't look much better as it also allows 17...f5. 17.Ng3 looks like a better way for White to keep his advantage.
Zinkl criticises 27...Rxd6, even giving the move a <?>, but this seems exaggerated. It's true that 27...Rg1+ would probably have led to a draw soon, but the position is still about equal after the text move.
29...Rg1+ is called a pretty combination which cannot cancel out Black's difficulty according to Zinkl. But it looks like a slight inaccuracy. 29...Re6 looks like a safe alternative.
The endgame is winning for White, according to Zinkl, as he can force Black into zugzwang. The game would be drawn if the black c-♙ was still on c7. But White allows Black chances for a draw with 37.a3 instead of Zinkl's suggested <37.b4>. But then, Krejcik seems to have played for a win with 39...b5, when Zinkl suggests <39...c5> with a draw.
Source: 'Wiener Schachzeitung', Suppl 1910, pp. 272-276
Kibitzing
sac: 33.Kg4
}?