Stefan Bucker has posted an article on the main line of the Kieseritzy on Chess Cafe. An expanded version of this article will appear in the next issue of Kaissiber. In my opinion this patches up a line of the King's Gambit where white was struggling. 

I have played the King's Gambit from both sides for many years both over the board and in correspondence. I am currently happy playing against every line from black. Black can equalize in some lines if he is very careful. However many of these positions are even but not drawish. Some of the positions that arise are very difficult and quite different to other openings.      

Here we go again, this time bashing the Allgaier.

1.e4 e5 2.f4 exf4 3.Nf3 g5 4.h4 g4 5.Ng5 h6 6.Nxf7 Kxf7
-7.Nc3 d5 8 d4 f3 9 Nxd5 Nf6 10 Nxf6 Qxf6 11 gxf3 Qxf3 12 Bc4+ and I think Ke8 is more than =+, eg 13.Qxf3 gxf3 14.Kf2 Bg4 15.h5 Nc6 16.c3 Na5 17.Bd3 c5.
-According to an analysis from Chabelsky 7.Bc4+ d5 8.Bxd5+
a)8...Ke8 9.d4 (but then why not 7.d4 immediately?) Nf6 10.Nc3 Nh5 11.0-0 c6 12.Bb3 Bg7 3.e5 Tf8 14.Ne4 Qxh4+ 15.Nd6+ Kd7 and Black has the extra piece plus the winning attack ...
b)8...Kg7 9.Bxb7 Bxb7 10.Qxg4+ Kf7 11.Qh5+ Ke7 12.Qe5+ and 100 years ago one knew better than current sicilon power: Kd7 13.Qxh8 (13.Qf5+ Kc6 14.Rh3 a5! and the black king escapes) Nf6 14.e5 Bxg2 (Fritz6 thinks White is better, but we should rather rely on the old masters) 15.Rg1 (15.Rh2 f3 16.Qxf6 Qxf6 17.exf6 Bc5, Gunsberg in 1895; Fritz6 gives 18.d4 Bxd4 19.Nd2 Nc6 20.f7 Ne5 but has changed its evaluation drastically) f3 16.Qxf6 Qxf6 17.exf6 Bc5 (Lewenfisj 1910). All this shows that Eric Schiller's books sometimes are useful; it's to be found in his "Who is afraid of the KG". Heartwarming that these old masters understood stuff like this better than 21st century computers.
-So I tried to revive 7.d4, similar to the HAG.
a)7.d4 f3 (this thematic move is probably even stronger than d5 8.Bxf4 Nf6 9.Nc3 Nxe4 or Be6) 8.Bc4+ d5 9.Bxd5+
a1)9...Ke8 10.gxf3 Nf6 11.Bb3 gxf3 12.Be3 Nc6 and I doubt White's compensation.
a2)9...Kg7 10.gxf3 Nf6 11.Nc3 Bb4 12.Bc4 c5! -+Le Goff-Bauget, corr 1993 is an improvement on the old 12...gxf3.

Sidenote, possibly not very relevant: C2 (see previous page). After 7...Nf6, Harding's main line is 8 d4 d5 9 Bxf4 Bb4 10 Bd3 and now Bxc3+ 11.bxc3 dxe4 12 Bc4+ Be6 and in no single line Fritz will drop his evaluation, as far as I can see

On the topic of how to answer 2 f4 exf4 3 Nf3, I'm not convinced that Black is better with best play after 3...g5.  Black can certainly equalise, but it's difficult to prove more than that.  The Hanstein Gambit, while unpromising for White, probably equalises, while in the Kieseritzky, see Stefan Bucker's analysis of 4 h4 g4 5 Ne5 d6 Nxg4 Nf6, and 5...Nf6 6 Bc4! (rather than the more ambitious, but insufficient, 6 d4 d6) d5 7 exd5.
http://www.chesscafe.com/text/kaiss34.pdf
http://www.chesscafe.com/text/kaiss35.pdf

Also, I wonder if Fritz's 7 Bc4+ d5 8 Bxd5+ Kg7 9 Bxb7 mentioned above is wholly sound in the Allgaier; if it is, even the Allgaier may be good enough for equality.

As Black (I sometimes meet 1 e4 with 1...e5) I quite like the Cunningham in response to the KG, and I also like the "nameless 3...Nc6 defence" followed by 4...g5- it prevents the Kieseritzky and in my experience often confuses White players who are expecting 3...g5.

MNb wrote on 07/29/08 at 01:57:37:



1.  After 8 d4 Bb4 I reckon White can try 9 Bc4+ d5 10 Bxd5+ Kg7 11 Bxf4, unclear, as Bb4 isn't usually one of Black's best moves in the dodgy 8 Bc4+ line.

2.  Having another run-through of 8 d4 d5 9 Bxf4 Nf6, I'm not sure that that stretch of the article is among Harding's best, as he glosses over Glazkov's 10 Bb5+ Bxc3+ 11 bxc3 Nf6 12 0-0 Kg7, when 13 exd5 is fine for White, and Rodricht's 14 Bxg4 in the 10 Be2 line, which also looks OK (though I feel Black has improvements in the 13...dxe4 line, I don't trust 10 Be2 and 13 c4 at all).

As well as 10 Bb5 which is about equal, 10 Bd3 is also better IMO than 10 Be2.  So maybe even that line may be playable for White.

3.  8 d4 d5 9 Bxf4 Be6 (the immediate 8...Be6 is illegal!) can be met by 10 Be2, undermining the g4 and d5-pawns, while 10 Bd3 isn't too bad either, e.g. 10...Nf6 11 exd5 Nxd5 12 0-0.

Anders wrote on 07/29/08 at 13:13:52:


Well, I usually play the Modern but I certainly would also consider the Cunningham.  For whatever it's worth.

I apologize that it's not in the spirit of this thread, but personally I would never consider 3...g5.  I understand that Black is better, but there are much simpler defenses that utterly frustrate White's quest for (a) the advantage and (b) the thrill of playing his heavily prepared, antique system ca. 1865.

Here's Harding's articles:
http://www.chesscafe.com/text/kibitz78.pdf
http://www.chesscafe.com/text/kibitz79.pdf

I've checked over Gallagher's NCO line with the assistance of Fritz 10, and think there may be improvements for White over it, but nothing particularly inspiring.  My analysis of the Allgaier from tonight's survey (which probably has many holes) runs:

A.  7 d4 f3 (7...d5!?) 8 Bc4+ d5 9 Bxd5+ Kg7 10 gxf3 Nf6 11 Bc4 (11 Nc3 Bb4 -/+ Gallagher, e.g. 12 Bc4 c5!) 11...c5 12 e5 Nc6 =+.

B.  7 Bc4+ d5 8 Bxd5+ Kg7 (8...Ke7 9 d4 gives White sufficient compensation) 9 Bxb7 is Fritz's suggestion (9 d4 f3! =+, transposes to line A) 9...Bxb7 (9...f3 10 Bxc8 f2+ 11 Kxf2 Qxc8 12 d4 with compensation) 10 Qxg4+ Kh7 (10...Kf7 11 Qh5+ Ke6 12 Qf5+ Kd6 13 d4 with compensation) 11 Qf5+ Kg7 12 Qg4+, with a repetition draw.

C.  7 Nc3

C1.  7...Bb4 8 Bc4+ d5 9 Nxd5 Be6 10 Qe2 with compensation.

C2.  After 7...Nf6, Harding's main line is 8 d4 d5 9 Bxf4 Bb4 10 Bd3 dxe4 11 Bc4+ Kg7 12 0-0 Nc6 13 d5, which is unclear.  This is one of those lines where Fritz initially thinks Black is much better, but push it down its suggested lines, and the evaluation drops substantially.

C3.  7...d5 is the one move that might be better than 7...Nc6.  8 d4 (8 Nxd5 Nf6 9 Nxf4 Qd4 10 d3 Bc5 -/+) 8...f3 9 Nxd5 Nf6 10 Nxf6 Qxf6 11 gxf3 Qxf3 12 Bc4+ Kg7 13 Qxf3 gxf3 =+.

C4.  7...Nc6 and now:

C4a.  8 Bc4+ d5 9 Bxd5+ Kg7 10 d4 Bd6 (10...f3 11 gxf3 Be7 12 Be3 Bxh4+ 13 Kd2 is unclear, 10...Nf6 11 Bxf4 with compensation) 11 Bxc6+ (11 e5 Nxe5 12 dxe5 Bxe5 leaves Black a pawn up and destroys White's centre) 11...bxc6 and I think Black may be slightly better, though the position is still murky.

C4b.  8 d4 d5 (8...f3 is probably best met by 9 Be3 with another unclear position, rather than 9 Bc4+ which tends to transpose to line C4a)   

C4b1.  9 Bxf4 Bb4 is good for Black, but I think White's two move-9 alternatives are worth another look:   
C4b2.  9 exd5 Qe7+ (9...Nce7 10 Bxf4 Ng6 11 Be5 Bg7 12 Qe2 unclear) 10 Kf2 Ne5 =+
C4b3.  9 Nxd5 Nf6 (9...f3 10 gxf3 Nge7 11 Bc4 Nxd5 is considered good for Black according to Harding, but White seems fine to me: 12 Bxd5+ Kg7 13 Be3 with compensation) 10 Bxf4 is extremely messy.

In conclusion, unless I've missed something (which I might well have done), White's best move in the Allgaier seems to be 7 Bc4+ with sufficient compensation, though Black can force a quick draw in that line.  7 Nc3 is best met by 7...d5 with an edge for Black, while the Hampe-Allgaier with 7...Nc6 looks unclear.  It also appears that MNb is right about 8 d4 with the intention of queenside castling, rather than 8 Bc4+, being White's best approach in the Hampe-Allgaier lines.

Tim Harding wrote an article on the Allgaier a while ago, in which he concluded that 7 Nc3 was the best move, but it was still doubtful as to whether White got enough against accurate defence.  Black's best may be to transpose to the Hamppe-Allgaier with 7...Nc6, when Harding suggests 8 Bc4+ as a largely unexplored possibility.

7 d4 and 7 Bc4+ are inaccurate in view of the line given in NCO.

One last addition: a transposition table for all the first 6 moves or so of the ...g5 systems. No evaluations on here, just move orders. Anything in bold represents the primary move order to achieve individual systems within the ...g5 strategy, italicized lines are transpositions. This is important because much of the future viability of the KG might lie in move orders avoiding unpleasant positions.




I also found these notes, which talk about some of the naming conventions and general strategy ideas:

General Guidelines (Black plays ...g4)

• The Kieseritzky is probably the most logical continuation
• Lines with 4 Bc4/d4/Nc3 either:
  
  • require White to sacrifice a piece
  • allow Black to play …Qh4+
• lines that prepare …g4 with …d6/Nc6 prohibit Nf3-e5
• White’s light-squared bishop usually belongs on c4, hitting f7 (though only after …g4 has been played because of 4 Bc4 Bg7!)

General Guidelines (Black sets up with Bg7/d6/h6)

• If White plays Bc4, it’s a Hanstein Gambit
• If White plays Bc4 and h4, it’s a subset of the Hanstein called the Philidor Gambit (or Greco-Philidor Attack)
• If White plays 4 Bc4 Bg7, 5 d4/Nc3 still allows Black the …g4 option
• If White doesn’t play Bc4
  
  • and Black plays …h6, it’s a Becker Defense
  • and Black plays …d6, it’s a Fischer Defense
  • and Black plays …Nc6 without …d6, it’s a Pierce Gambit Declined
• White’s light-squared bishop usually belongs on f1 (guarding h3 in some lines) or b5 (after …Nc6)