The Fried Liver Attack and the Lolli Attack are two venerable ideas for white that spring from the Two Knights Defense. They are aimed at refuting black's play after the sequence 1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 d5 5. ed Nxd5 ?!
This capture is generally considered tantamount to suicide given white's two most popular replies, 6. Nxf7 (Fried Liver Attack) and 6. d4 (Lolli Attack).
Personally, I've always been rather fond of 5...Na5 and have never had any great problems equalizing, although there has been quite a bit of theory to absorb along the way. However, because of all the positive chatter surrounding the Lolli and Fried Liver Attacks on Chess.com so, I decided to give 5...Nxd5 another look. My findings: contrary to their supports here, neither the Fried Liver or Lolli Attacks leave black with any particular difficulities (with just a little preparation) and in that sense, not only is 5...Nxd5 playable, it may even be considered preferable, given all the 5...Na5 theory the bold capture avoids.
Firstly, for those posters who have used the Perreaux link to defend their excitement over the Lolli (http://www.rci.rutgers.edu/~goeller/urusov/perreux/lolli_attack.html), please keep in mind the move order here is NOT that of the Lolli, although the position reached at move 7 could arise from the correct Lolli move order. This link espouses 1. e4 e5 2. Bc4 Nf6 3. d4 cxd4 4. Nf3 Nc6 (of course here, black can simply play 4...Nxe4 leading to a Bishop's Opening line known as the Urusov Gambit) 5. Ng5 (and this is now not a Lolli at all, but rather a Scotch Gambit) d5 6. exd5 Nxd5? (this is wrong -- 6...Qe7+ has been well known for many decades to be the best move for black) 7. O-O!.
The true Lolli move order is 1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 d5 5. exd5 Nxd5!?. It is at this point the road forks, where white chooses to enter the Fried Liver Attack (gotta love that name) with 6. Nxf7?! or the Lolli Attack with 6. d4. The difference between this position and the "Perreaux" position is very important.
To get to the Perreaux, white would have to play 6. d4 and black would have to reply 6...exd5. White can then follow with the Perreaux recommendation of 7. O-O!
Instead, after 6. d4 black here has 6...Bb4+! After 7. c3, black has taken the best square away for the white knight at b1. Since 7. B/Nd2 allow 7...Qxg5 and 7. Kf1 allows 7...O-O, white has little choice when it comes to answering this inconvenient check.
Following 7. c3 Be7 8. Nxf7 Kxf7 9. Qf3+ Ke6 10. Qe4 b5! (the key move both in the Lolli and the Fried Liver Attacks for black). Here, after 11. Bxb5 Bb7 12. f4 g6! Now 13. fxe4 (removing the e5 protection with 13. Bxc6 Bxc6 14. Qe5+ Kf7 leaves black with even material, development AND attack) is met with 13...Rf8!
Black's position, in spite of (because of?!) the position of his King, is now becoming quite active:
14. Qg4+ (obvious moves like 14. c4 fail because of the inherent activity of black's pieces, e.g. 14...Ncb4! 15. cxd5+ Bxd5 16. Qg4+ Rf5 17. Rf1 c6! [it seems to me black could try 17...Nc2+ 18. Kd1 Nxa1 but after 19. Bd3 white's pieces are becoming useful again] 18. Ba4 [18. Be2 Qf8 -- black wants to play ...Kd7-c7 -- 19. a3? Nc2+ 20. Kd1 -- 20. Kd2 Nxa1 and black threatens Nb3+ -- Nxa1 21. Bd3 Bb3+ 22. Kd2 Bg5+ 23. Kc3 Bxc1 24. Rxc1 c5+ and black is, for the moment at least, a rook up and white's king certainly doesn't seem any safer than black's!] 18...Nd3+ 19. Kd1 Nxe5 20. de Bf3+ and there goes the white queen) 14...Rf5 15. Bd3 Nxd4! 16. Rf1 (if 16. Qxd4 Rxe5+ and the black counterattack shifts into high gear, while on 16. cd Nb4 17. Bf5+ gf is a position that MCO regards as "unclear", but I can't imagine any white player -- even with the exchange and two pawns -- wanting to defend the white position and Fritz doesn't even list 17. Bf5+ among its candidates, it considers it so bad) 16....Bh4+ 17. g3 Ne3! 18. Bxe3 Nf3+ 19. Rxf3 Qxd3 20. Qd4 Qxd4 21. Bxd4 Rxf3 22. gxh4 Rh3 and black wins.
Now, let's turn our attention back to the Fried Liver Attack, which I was surprised to find was also an interesting and playable choice for black. The Fried Liver moves from the "choice" position above, where instead of 6. d4, white chooses the direct 6. Nxf7?!
Black proceeds directly with 6...Kxf7 7. Qf3+ Ke6 8. Nc3 (notice how the knight contributes to the attack, which black disallows in the Lolli continuations above with a timely ...Bb4+) Nb4 9. a3! (on 9. Qe4, black again has the surprising idea of 9...b5! 10. Bb3! [10. Bxb5 c6 11. Bc4 Kd6! 12. O-O -- 12. Bxd5 exd5 13. Nb5+ Kc6 with obvious advantage -- 12...Nf6! followed by ...Nxc2 and ...Nd4, while 10. Nxb5 meets with 10...c6 11. Nd4+ Kd6 12. Nf3 Qf6 followed by ...Bf5] 10...c6! 11. O-O Bb7 12. Re1 Kd7! 13. a3 [13. Qxe4 Qf6 and black is developing smoothly with ...Bd6 and Rae8 following] 13...Nxc3 14. dc Nd5 15. Bxd5 cd 16. Qxe5 Qf6 --again that key move! -- 17. Qe2 Bd6 18. Qxb5+ Bc6 19. Qd3 and both 20...Rhe8 and 20...Rhf8 provide black an advantage -- according to both Fritz and Crafty -- despite white having gained his 3rd pawn for the piece -- Black's two bishops are awesome) 9....Nc2+ 10. Kd1 Nd4 11. Bxd5+ Kd7! 12. Qg3 (if 12. Qf7+ Qe7 followed up with ...c6 as after 12. Qe3 Bd6) Qf6 13. Ne4 (13. Re1 c6 are rated by the the two engines listed above as just slightly better for black while 13. d3 Nf5 14. Qg5 Be7 are rated even better with =+) 13...Qf5 14. d3 c6 15. Ba2 Kc7 16. Be3 Qg4+ 17. Qxg4 Bxg4+ 18. f3 Bf5 and, according to silicon genuises, this position is flat equal.
With reasonable play, it seems that black has virtually no losing chances with either line. Can we now say that 5...Nxd5 is rehabilitated? Given all the lines that black has to play against (and learn) with 5...Na5 perhaps the oldest trap in the book is now the smart play!
I for one definitely plan to add 5...Nxd5 in my repertoire, because I doubt few white players are ready for its intricacies. There is something audacious and liberating about marching your king right into the thick of the battle. Any brave souls out there care to join me?