Talk:Fried Liver Attack

Origins of the name
Does anyone know why this game is called the fried liver?

I found something that might come in handy. The original name of this opening is in Italian, and the website mentions Venice. Hans Ree is a grandmaster, so I suppose his explaination could make sense (if this is in fact Hans Ree's writing). http://www.chesscafe.com/text/hans40.txt

DaDoc540 20:23, 18 November 2006 (UTC)

Continuations?
What are white's next choices after Nc3? O-O is my next move.


 * Could you explain more? What moves are you taking before hand? I'm assuming you're doing 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 then what? 5.exd5 Nxd5 6.Nc6 ...? Black will probably do Qxg5. Assuming that, you don't want to castle, that invites Nf4 or Nxc3, which will both be painful. Instead look at Bxd5 which will probably be followed by black playing Bg4 and you with d2d3 to counter or f2f3 to block it. Note that with f2f3 he'll probably just move his bishop to e6, so even though it blocks it actually will probably come out bloodier than d2d3. 12.206.59.238 (talk) 16:19, 30 October 2008 (UTC)

Main line
What is the main line? —Preceding unsigned comment added by ChessCreator (talk • contribs) 10:34, 8 February 2008 (UTC)
 * The main line of the Two Knights Defense is given in that article: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Na5 6.Bb5+ c6 7.dxc6 bxc6 8.Be2 h6 9.Nf3. The Fried Liver deviates from the main line with 5...Nxd5, exposing Black to a strong attack.  It isn't a popular choice, as most players who choose the Two Knights prefer to be the aggressor (the main line gives Black some initiative at the cost of a pawn, and the Wilkes-Barre or Traxler gives Black even wilder tactical possibilities) rather than defend. The main line isn't a "refutation" of the Fried Liver, since it isn't the Fried Liver at all.  A refutation would come if Black can demonstrate a win after 6.Nxf7.  Quale (talk) 10:53, 8 February 2008 (UTC)
 * Okay, see what happened now. I meant the main line of the Fried Liver Attack, but it's not important as the previous edit by Smcb555 was talking about a refutation of the Two Knights Defense, and thus your reply within that context. ChessCreator (talk) 12:43, 8 February 2008 (UTC)
 * Got it. I don't think people normally talk about a main line in the Fried Liver.  There aren't that many options until Black's eighth move, and probably nothing stands out as a main line after that.  Quale (talk) 14:41, 8 February 2008 (UTC)
 * Depends on your level of detail I guess. Many years ago the main line of the fried liver was 6.Nxf7 Kxf7 7.Qf3+ Ke6 8.Nc3 Nb4 9. a3! (Somehow remember this moves was given an '!') that would have been in ECO. When Smcb555's edit was done saying that Nd4 was a refutation, my first thought was that the Fried Liver theory was rewritten and was now 8...Nd4, the move does attack the queen and the c2 pawn. But it seems Smcb555's meant 5...Nd4 (The Fritz) instead of the fried liver, although of course that is no refutation either. ChessCreator (talk) 15:08, 8 February 2008 (UTC)


 * I ran a very long analysis with Rybka 3 on the Fried liver and after a very long time she said draw (actually +.11 but I looked at at the main lines and they were draws) —Preceding unsigned comment added by 67.80.58.207 (talk) 21:43, 26 July 2009 (UTC)

Actually 9.a3?! is unsound and is met by 9...Nxc2+ 10.Kd1 Nxa1 11.Nxd5 c6! (mentioned in one of the earlier Kaissiber issues), White is thus better off with 9.Qe4, 9.0-0 or even 9.Bb3 (which I haven't seen mentioned in any source), all three lead to pretty unclear situations.Tws45 (talk) 10:15, 5 April 2011 (UTC)


 * I know Stockfish likes 10...Nxa1, but the analysis I have from a book (well, Kindle monograph) thinks 10...Nd4 is better, so while 10...Nxa1 may be better, I think a chess player (or engine) strong enough to effectively play 10...Nxa1 to refute 9.a3 would had instead played 5...Nxd5, avoiding Fried Liver attack. That said, if you can find that Kaissiber issue (or any other source) to argue for 10...Nxa1, I will restore that line, but right now, we can’t recommend lines based solely on unreferenced assertions or what Stockfish likes. Samboy (talk) 04:14, 4 April 2022 (UTC)
 * OK, as per the German language article, Shirov vs. Sulskis 2014 is probably notable enough (made Game of the Day at Chessgames) to add a note about 10...Nxa1 to the article. Samboy (talk) 04:36, 4 April 2022 (UTC)

7...Kg8??
I've never seen anyone play this allegedly "common" beginner blunder. Giving it a diagram definitely seems like undue weight. MaxBrowne (talk) 02:08, 19 April 2014 (UTC)

OK then. Aaronaraujo2013 (talk) 20:52, 19 April 2014 (UTC)

Stockfish vs. Humans
Let’s look at a common line in the Fried Liver:

1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 d5 5. exd5 Nxd5!? 6. Nxf7!? Kxf7 7. Qf3+ Ke6 8. Nc3 Ncb4 [1] 9. a3?! Nxc2+ 10. Kd1

and we get this:

Now, looking at this position, I want to point out why the WP:CHESSENGINE rule is important (Edit: Since the version of this article before I edited it marked 9. a3 with a “?” without reference, which indicates someone might have put chess engine analysis in this article). Running the position through Stockfish, and it will say the best move is 10... Nxa1!? but that was tried in a grandmaster level game and Black lost. So while Stockfish can navigate the tactical complexities after gaining a rook with Nxa1, I’m not sure a human player wants those complexities, and 10... Nd4 is probably the better move for a human defending as Black in this position (actually, 5... Na5 was the move to make as Black, but I digress). This position shows that, while a given move may be, with a powerful computer, better, the move, if played by a human, even a grandmaster human, is dubious. Hence WP:CHESSENGINE; I’m not becoming Stockfish any day soon, so I should play like a human, not a computer. Samboy (talk) 05:41, 9 April 2022 (UTC)

[1] Most notate this simply as Nb4, since the knight on d5 can not move because he is pinned, but I prefer using Ncb4 so computer programs don’t need to look for pins when evaluating which piece went to a given square.
 * WP:CHESSENGINE doesn't have anything whatsoever to do with what human players think of the quality of engine analysis, it's strictly a WP:RS and WP:NOR issue and that's the reason it's important. I'm not sure why you think Stockfish has anything to do with the lines in this article.  I didn't investigate in depth, but the only recent mentions I see of Stockfish are in your edit summaries.  Quale (talk) 06:01, 10 April 2022 (UTC)
 * The reason I bring it up in this article is because the version before I edited it said that 9. a3 was a bad move, without reference, and directly contradicting Yakov Estrin’s classic take on this line. So, I wonder how they decided a3 was weak. And, indeed, Stockfish 14 NNUE thinks 9. a3 is quite bad (-1.7 pawns or so for White)...but that move has been played in grandmaster level play with White winning, so while it’s a line the computers don’t like, it’s very difficult to refute as a human (Lichess history: Overall favors white 53% wins; favors White in 2000+ correspondence, 1 win; favors Black in 2200+ game/30, 1 win; favors Black in 2200+ 15 minutes + 15 seconds/move, 1 win).  Usually, I reasonably trust Stockfish to verify annotations (e.g. in Kasparov's Immortal, one reliable source points out that Kasparov would had won a piece after 22. ... Nfxd5?, but Stockfish uses a different line; the printed line is 23. exd5 Qd6 24. Qxf7+ Kb8 25. Re6 Qc7 26. Re7 Nd7 27. Nc6 Bxc6 28. dxc6 Qxc6 29. Bxd7 but Stockfish thinks Black loses less badly after 24... Nd7, so I pause the annotation after 24. Qxf7+ and point out 24... Kb8 is just one possible Black defense at that point), but with this line, putting in annotations with directly contradict Stockfish looks to be OK; so I put in 10...Nxa1 as !? even though Stockfish thinks it’s Black’s best move (-1.6 pawns for White at 22 ply; Black’s next-best move is 10... Nd4 which is even for both Black and White in Stockfish’s analysis). Samboy (talk) 10:47, 10 April 2022 (UTC)
 * You're right that people often insert computer analysis into chess articles, sometimes giving its source but also often not revealing the origin. Uncited material can be tagged as needed a reference (using cn and others), or it can simply be removed.  You are doing the third and best thing, replacing the uncited material with cited claims.  Thanks. Quale (talk) 05:03, 14 April 2022 (UTC)

10...Nxa1 is fine if you find the fearless 11.Nxd5 Qh4!! that Sulskis missed. According to this review, Erwin l'Ami mentions this move in his Fritz Trainer on the opening (don't have it myself, though). Double sharp (talk) 17:34, 11 January 2024 (UTC)

5... Nxd5 analysis paragraph in lead
Why is this paragraph here, and in particular, why does it awkwardly slice up the move list? I don't understand why an article about a chess opening is analyzing backward from the relevant point. It's not even like there was no better place on Wikipedia to cover this information; the article Two Knights Defense covers it just fine.

I suppose I could understand if this information is included to compare the Fried Liver Attack to related openings, but even still, this information should be placed in a less awkward spot. As it is, it initially gives the impression that 5... Nxd5 is the end of the move list, which is confusing. And actually, in the paragraph's current state, there is no discussion of the relation of the Fried Liver Attack to these other openings; it just states that all of these openings exist. ISaveNewspapers (talk) 06:05, 3 December 2023 (UTC)
 * Sorry, I should have responded to you here earlier but I hadn't noticed this section. I didn't perceive the lead as as awkward as you do, but your rearrangement of the text is fine.  I will say that I'm not at all convinced it's an improvement.  I have many books on chess openings that intersperse explanations and commentary with the moves so this presentation was not some avant garde experiment cooked up by a rogue wikipedian.  The argument that an article on a chess opening variation should be completely free of any context and not include any information on the key alternatives to the line is so bizarre it's hard for me to really engage with it.  This is an encyclopedia article intended for a general audience, not Informant where several paragraphs of bare moves and cryptic evaluation symbols would be acceptable.  I don't understand why we would expect or require a general reader desiring context to find key move alternatives on their own in a different article. I completely fail to see the problem you are trying to fix.  Quale (talk) 08:15, 29 December 2023 (UTC)