Understanding Grandmaster play – Fabiano annotates

12/18/2007 – Recently a 15-year-old won the Italian Championship. Fabiano Caruana took the title in this twelve-player round robin with a resounding three-point victory and a 2740 performance, which exceeded a theoretical GM norm by a point and a half (Fabiano already has the GM title). Lase week we featured this achievement in a special report. Today we bring you a key game annotated by the champ.

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!

Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!


The result of the Championship can be seen in the table below. Note that we have corrected the flags of the first and third players, after half of Italy wrote to us protesting the American flag next to Fabiano Caruana and the Argentinian next to Carlos Garcia Palermo. Both were registered as Italian players (Fabiano has dual American/Italian citizenship), although the database we used to generate the cross table had not heard about it yet. Now we have Mega 2008 and the problem should not occur again.

In our interview with Fabiano the young GM (one of the youngest in the world today) mentioned that he had annotated a key game for ChessBase Magazine, where there is of course much more. Here is the game for you to read and replay: Note that on our JavaScript board you can click anywhere in the notation or analysis, and the board on the left will display the position. We urge you to play through these notes as they provide a unique insight into the thought process of a very profound young player.

Godena,Michele (2535) - Caruana,Fabiano (2594) [B51]
Italian Championship Absolute '07 Martina Franca (8), 01.12.2007
[Annotations by Fabiano Caruana]

1.e4 c5 2.Nf3 d6 3.Bb5+ Nc6 4.0-0 Bd7 5.Re1 Nf6 6.c3 a6 7.Ba4 b5 8.Bc2 Bg4 9.d3 e6 10.Nbd2 Be7 11.h3 Bh5 12.Nf1 0-0 13.Ng3 Bg6 14.Nh4 d5. I had prepared this variation before the tournament, specifically for Godena. I don't believe White has any advantage here, although with almost no practical tests it is hard to give a definite verdict now.

15.exd5. The safest choice, leading to an absolutely equal position. Indeed, in most games after this move the players would shortly agree to a draw. However, I know from experience that even the most equal positions can contain a lot of play and chances to outplay a weaker opponent. In this game I achieved exactly that after numerous small mistakes from my opponent. 15.e5 Nd7 16.Nxg6 hxg6 17.d4 has been the main choice recently, with satisfactory results for White. Of course this is the only way to fight for an advantage, and Black must play carefully, but I feel Black has reasonable chances after 17...b4 18.Ne2 Rb8.

15.f4!? leads to highly tactical and unclear play: 15...d4 (15...Nxe4 16.Nxg6 Nxg3 17.Nxf8 Bxf8; 15...Ne8 16.Nxg6 hxg6 17.Qf3 f5+/=) 16.f5 Bd6

  • A) 17.e5 Bxe5 18.Rxe5 Nxe5 19.fxg6 Nxg6=/+;
  • B) 17.Qf3 Nd7 (17...Qc7 18.Ne2 exf5 19.Nxf5) 18.Nxg6 hxg6 19.fxe6 Nde5-/+;
  • C) 17.Nf1! exf5 18.exf5 Bh5 (18...Nh5? 19.fxg6 Qxh4 20.Re4+-) 19.g4 dxc3 20.bxc3 (20.gxh5 Be5; 20.Nf3? Nxg4 21.hxg4 Bxg4 22.bxc3 Qf6)
    • C1) 20...Nd5 21.Nf3 Nxc3 22.Qd2 Be5!? (22...Qf6? 23.Bb2 Bxg4 24.Bxc3 Qxf5 25.hxg4 Qxf3 26.Bd1; 22...Bxg4 23.hxg4 Qf6 24.Bb2 Nd4! 25.Nxd4 cxd4) 23.Nxe5! (23.gxh5!? Ne2+ 24.Rxe2 Bxa1 25.h6) 23...Qd4+ 24.Qf2 Nxe5 25.Be3 Qd5 26.Nh2!+/-;
    • C2) 20...Ne5!? 21.gxh5 Nd5 also must be considered and is perhaps the best option at Black's disposal.;
    • C3) 20...Be5
      • C3a) 21.gxh5? Bxc3;
      • C3b) 21.Bd2 Nd5 22.Qf3 (22.Nf3 Nxc3 23.Qc1 Nd4-/+; 22.d4!? cxd4 23.Nf3) 22...Bd4+!! 23.Kh1 (23.cxd4 Nxd4 24.Qd1 Qxh4) 23...Nxc3 24.Ng2 (24.Qxc6 Qxh4 25.gxh5 Qxh3+ 26.Nh2 Qxh5; 24.Ng6!?) 24...Nb4 25.Bb3 a5!;
      • C3c) 21.Bb2 Nd5 22.Nf3 (22.Rxe5? Nxe5 23.Nf3 Nxg4 24.hxg4 Bxg4-/+) 22...Nxc3 23.Qd2 Ne2+ 24.Rxe2 Bxb2 25.Rae1 Nd4-/+;
      • C3d) 21.d4! cxd4 22.Nf3 (22.gxh5 d3! 23.Qxd3 Qxd3 24.Bxd3 Bxc3) 22...d3 23.Nxe5 Nxe5 24.Rxe5 Qc7 25.Re3 Qxc3 26.Bd2 Qxc2 27.gxh5+/- However, White's position is very suspicious and I'm sure Black is fine after 15.f4. Perhaps best is 20...Ne5!?.

15...Qxd5 Movsesian recaptured with the knight, but I prefer Tiviakov's move. The knight will remain on f6 to restrict White's on g3. 16.Nxg6 hxg6 17.Qe2 Rfd8

18.a3?! This is actually not helpful for White's cause at all. I would probably not want to play ...b4 anyway, as it increases the scope of White's light-squared bishop. It is interesting that Godena makes a series of small errors, which don't hurt his position tremendously but make it increasingly difficult to defend; eventually when he must play very accurately to hold the position and with limited time, he quite naturally collapses and gets a losing position. 18.Bf4 Nh5 (18...Rac8=) 19.Nxh5 Qxh5 20.Qe4 Qd5 21.Qxd5 exd5 22.Bd1 g5 23.Bh2 Bd6= (Smirin-Tiviakov, Gothenburg 2005). 18...Rac8 19.Be3. 19.Bf4 Bd6 shows the difference between 18.Bf4 and 18.a3: now there is no Bb3 for White and Black can achieve this favourable exchange. 19...Qe5!

20.Qf3. Again this is unnecessary. White is absolutely not worse in this position and therefore doesn't have to force equality! After a natural move like 20.Rad1 I will be hard-pressed to find any fruitful ideas. 20.Ne4 Nxe4 21.dxe4 Bg5!? 22.Rad1 Bxe3 23.Qxe3 g5=/+; 20.Rad1 Qc7 (20...Nd5 21.Bd2 Qxe2 22.Nxe2=) 21.Bd2 Bd6 22.Ne4 Nxe4 23.Qxe4. 20...Qc7 21.Bf4. Due to one tactical trick White cannot equalise immediately, but there are still no problems after the calm 21.Rac1. [21.Ne4? Nd4! 22.Nxf6+ (22.cxd4 cxd4) 22...Bxf6 23.Qd1 b4-/+; 21.Rac1 Ne5 (21...Nd5 22.Bd2; 21...Bd6 22.Ne4 Nxe4 23.Qxe4=) 22.Qe2 Nd5 23.Bd2=]. 21...Bd6 22.Bg5. After this the position becomes very unbalanced and Black certainly has more chances for an advantage than before. Of course White should have played it safe with 22.Bxd6. After 22.Bxd6 Qxd6 I still can't claim that Black is better. After the trade of bishops White cannot 'force' equality by Ne4 since that would lead to a clear good knight vs bad bishop position. Nevertheless, if White does nothing it is difficult to suggest a plan for Black. So although Black is now calling the shots, it shouldn't be enough for anything real. 22...Bxg3

23.Qxg3? Now this mistake is very real! After 23.fxg3 I would prefer to be Black, but perhaps this is compensated by the machine's endorsement of White's position. In any case it is roughly equal. 23.fxg3 Rd5 24.Bf4 (24.Bxf6? Rf5 25.Qe3 gxf6=/+) 24...e5 (24...Qd7 25.g4) 25.Be3 c4!? 26.dxc4 bxc4. 23...Qxg3 24.fxg3 Rd5 25.Be3 A tough choice for Godena, who was already down to a few minutes. The alternative 25.Bf4 is also unpleasant. 25.Bf4 Nh5 26.Rad1 a5!? (preventing b4) (26...Rcd8 27.b4) 27.Kf2 Nxf4 28.gxf4 g5! 29.fxg5 Rxg5=/+. 25...Ne5

26.Red1! A surprise for me, after which things are again far from simple. Godena's tenacious play in the following time trouble phase (only for him, as I had plenty of time) deserves much praise. In the end, however, the position proves too difficult to defend and he commits an error. 26.Rad1 Rcd8 27.Bxc5 (27.d4 Nc4) 27...Nxd3 28.Bxd3 Rxd3 29.Rxd3 Rxd3 30.Be3 Nd5 31.Kf2 f6=/+; 26.Bxc5 Nxd3 27.Bxd3 Rxd3 28.Bd4 (28.Be3 Rcd8 29.a4 Nd5) 28...Rxg3 29.Bxf6 gxf6 30.Rad1 Rg5 31.Rd6 a5-/+.

26...c4. 26...Rcd8 27.Bxc5 Nxd3 28.Bb6 (28.Be7 R8d7 29.Bxf6 gxf6) 28...R8d6 29.Bc7 Rd7 30.Bxd3 Rxd3 31.Rxd3 Rxd3 32.a4! Rd2 33.axb5 axb5 34.Ra8+ Kh7 35.Rf8 Rxb2 36.Rxf7 Kg8 (36...Rb1+ 37.Kh2 Rf1 38.Re7 Nd5 39.Rd7 Nxc3 40.Be5 Nd5 41.g4=) 37.Re7 Nd5 38.Re8+ Kf7 (38...Kh7 39.Be5 Ne3 40.Re7=) 39.Rb8!! (I looked for a while for something in this line, but couldn't find anything and in the end played 26...c4; the compter confirms my evaluation). 27.d4. The alternative looks very articifical but should also be taken seriously. However, Black must be much better. 27.dxc4 Nxc4 28.Bc1 (28.Rxd5 exd5! 29.Bc1 a5-/+) 28...a5 (28...Ne4!?) 29.a4 b4 30.cxb4 (30.b3? Ne3! 31.Bxe3 Rxc3-/+) 30...axb4 31.Bb3 Ne4. 27...Nc6 28.g4 Rd7

29.Re1! An incredibly powerful move. For a while I couldn't understand its value until I accurately calculated the consequences of 32.Re2!. In the end I spent 20 minutes and realised that even if White is fine there I have no other ideas and just have to go for it. The d5-square looks nice but by itself it won't amount to anything; it is difficult to find any chinks in White's armour and all pawn breaks will just open lines for White's bishops. White on the other hand has active counterplay with h4, g5, g4 and h5. – 29.g5 Nd5 30.Bd2 Na5 31.Kf2 Nb3 32.Rab1 a5 33.Be1 shows why the rooks must be placed on e2 and d1 respectively: 33...Re8 , and ...e5 comes next with an initiative for Black. 29.Kf2 Nd5 30.Bd2 Na5 31.Ke2 Nb3 32.Rab1 e5! 33.dxe5 Re8. 29...Nd5 30.Bd2 Na5. The knight goes to b3 to prepare the powerful ...a5 and ...b4. White has great difficulty defending against this and will struggle to make use of his clumsy rooks. However, with the regrouping manouver Re2 and Be1 White could sucsessfully combat White's initiative. This is indeed why I awarded 29.Re1 an exclamation mark. 31.Rad1 Nb3

32.Be3?! A mistake in very serious time trouble. White could have completed his great idea with 32.Re2, when Black has retains some pressure following 32...Nxd2, but not much more. 32.Kf2 is a more clumsy version of the following note, and indeed Black can take advantage of it: 32...a5 33.Ke2 b4 34.h4

  • A) 34...Rb7 35.g5 (35.h5? g5! 36.Rh1 Nxd2 37.Kxd2 Nf6-/+) 35...bxc3 36.bxc3 a4 37.g4;
  • B) 34...Kf8 35.g5 Ke7 36.g4 Rh8 37.Rh1 bxa3 38.bxa3 e5!?;
  • C) 34...e5!
    • C1) 35.Bxb3? cxb3 36.axb4 axb4 37.dxe5 (37.cxb4 Rc2) 37...bxc3 38.bxc3 Rxc3!-/+;
    • C2) 35.dxe5 Nxd2 36.Rxd2 bxc3 (36...Nf4+ 37.Ke3 Rxd2 38.Kxd2 Nxg2 39.Re4 bxa3 40.bxa3 Nxh4 41.Bd1) 37.bxc3 Nxc3+ 38.Ke3 Rxd2 39.Kxd2 Nb5 40.a4 c3+ 41.Kd3 Rd8+ 42.Kc4 Na3+ 43.Kxc3 Rc8+ 44.Kb2 Nxc2 45.Rc1 Rc4 46.Rxc2 Rxa4 47.Rc8+ Kh7 48.Rc7 g5!-/+; 32.Re2! a5 (32...Nxd2! 33.Rdxd2 a5 34.Be4 b4 35.Bxd5 Rxd5 36.axb4 axb4 37.Kf2 Ra8=/+) 33.Be1 (clearly an ideal setup for White) 33...Kf8 34.Kf2 Ke7 35.Kf3=.


33.Be4 White decides there is no alternative but to remove some pressure from c3, even at the cost of his valuable light-squared bishop. However, there was another possibility which shouldn't be worse than the game continuation. 33.Bf2 b4 34.axb4 axb4 35.cxb4

  • A) 35...Rb8 36.Bxb3 cxb3 37.Rd3 Rxb4 38.Ra1 Nf4 (38...g5 39.Ra3 Rdb7 40.Bg3 Kh7 41.Kf2 Kg6 42.Kf3=/+) 39.Ra8+ Kh7 40.Rf3 (40.Rd2? Rc7 41.Bg3 g5-/+) 40...g5 41.Ra3 Rdb7 42.Kf1 Kg6 43.Be1=/+;
  • B) 35...Nxb4? 36.Bxb3 cxb3 37.Re3;
  • C) 35...Rb7! 36.Be4 Rxb4 37.Bxd5 (37.Bg3 Ra8) 37...exd5 38.Re5 Rb5=/+.

33...b4 34.axb4 axb4 35.Bxd5 Rxd5 36.Bf2 Ra8 37.Re2 [Also very bad for White is 37.Kf1 Raa5 38.Bg3 g5 39.Kf2 Kh7 40.Kf3 Kg6-/+] 37...bxc3 38.bxc3

38...Rda5. I naturally didn't want to allow White's rooks activity after 38...Nc5 39.Rb1 Nd3 40.Rb7 Ra1+ (40...Ra3 41.Rc2) 41.Kh2 Rc1 42.Ra2 Rxc3? (42...Nxf2 43.Rxf2 f6 44.Rf3=; 42...Kh7!?) 43.Ra8+ Kh7 44.g5! Rxg5 45.Rbb8+/-. 39.Be1? This is another error, even more serious than the previous one. It is amazing how resilient White's position has proven to be; although I've made enormous progress from a roughly equal position, it still isn't easy to bring the full point home. White must play the computer's recommendation, when Black's advantage should be manageable: 39.Be3! g5 40.Kf2 f6 (40...Kh7 41.g3) 41.g3 Kf7 42.h4. 39...Ra1 40.Rxa1 Rxa1. White has reached the time control but the damage to his position is done. Black has a strategically winning game due to the passivity of White's pieces. 41.Kh2

41...Kh7? Incredible! I can say I've played well so far, but it isn't time to relax yet! Work remains and only one move consolidates the advantage. I thought there was no difference between this and 41...g5, since 42.g5 is met by 42...Nc1 43.Rb2 Nd3, when there is no check on b8. This error stemmed from overestimating the ...Nc1-d3 idea. 41...Nc1 42.Rb2 Nd3 43.Rb8+ Kh7 44.Bg3; 41...g5! 42.Re5 (42.Bg3 Rc1 43.Be1 Kh7) 42...Kh7 43.Bg3

  • A) 43...Rc1 44.Rxg5 Rxc3 45.d5
    • A1) 45...Rxg3? 46.Kxg3 exd5 47.Rh5+! Kg6 48.Rxd5 c3 49.Rd6++-;
    • A2) 45...exd5? 46.Be5! (46.Rxd5? Rxg3 47.Kxg3 c3-+) 46...d4 47.Rxg7+ Kh6 48.g5+ Kh5 49.Bf6->;
    • A3) 45...f6!? 46.Rh5+ Kg6 47.dxe6 Re3 48.Rh8 Rxe6 49.Rc8 Re4-/+;
    • A4) 45...Re3 46.d6 Rd3 47.Rb5 f6-/+;
  • B) 43...Kg6 44.d5 Nd2 45.h4 (45.Be1 Kf6; 45.Bf2 Nf1+ 46.Kg1 Ne3+-+) 45...gxh4 46.Bf4 (46.Bxh4 exd5 47.Rxd5 Nf1+ 48.Kh3 Rc1-+) 46...Nf1+ 47.Kh3 exd5 48.Rxd5 Ne3! 49.Rg5+ Kh7! (49...Kf6 50.Kxh4 Nxg2+ 51.Kg3 Nxf4 52.Rf5+ Ke6 53.Rxf4 Ra4 54.g5) 50.Rh5+ Kg8 51.Kxh4 Nxg2+ 52.Kg3 Ne1!?-/+ This is a truly amazing variation, which I naturally couldn't work out over the board. My calculation wasn't too far off but I missed a number of details such as 48...Ne3 (which is why I rejected 44...Nd2) and, even more importantly, 46.Be5.

42.g5! Now I realised what I had done. My advantage is of course still there but it is much more difficult to win. 42...Nc1 43.Rc2 Nd3 44.Bg3 Kg8 45.Bd6. 45.h4 Kf8 (45...f6) 46.Bd6+ Ke8 47.Be5 Nxe5 48.dxe5 Re1 49.Rb2 Rxe5 50.Rb8+ Ke7 51.Rb7+= shows why the king cannot escape from its prison and why 45.Bd6 has little point. 45...Ra6

46.Be7!? Tricky. He obviously wanted to tempt me into 46...Nf4, when he sucseeds in activating his rook due to 47.Rb2 Nd5 48.Rb8 Kh7 49.Bb4! 46...Kh7. 46...Nf4? 47.Rb2 Nd5 48.Rb8+ Kh7 49.Bb4!<=>; 46...Ra7 47.Bd6 f6 48.gxf6 gxf6 is very similar to the game, but I wanted to manouver for a while and run him down on time. 47.Kg3 Ra5 48.h4 Kg8 49.Bd6 f6. Now I achieve this when he has only five minutes left again. With so little time the defensive task is undoubtedly very difficult. 50.gxf6 gxf6 51.Kh3 Ra1 52.g4 Kf7 53.h5!? 53.g5 f5 54.Kg3 Ke8; 53.Kh2 e5 54.dxe5 fxe5 White's situation is unenviable in both lines, as he has no counterplay and I can torture him for a long time. 53...g5. There are other good possibilities but I was attracted to this.

54.h6? Collapsing in what is essentially a losing position. After this White has no more chances. 54.Kh2 f5

  • A) 55.gxf5? exf5 56.Rg2 g4;
  • B) 55.Rg2 f4 56.Be5 (56.Rg1!? Ra6 57.Be5 Ra2+ 58.Rg2 Ra3 59.Rc2 Ra1; 56.Rc2 f3 57.Bg3 Kf6-+) 56...Re1 57.Ra2 f3 58.Ra7+ Ke8 59.h6 Nxe5 60.dxe5 (60.h7 Ng6 61.Kg3 Rh1 62.Kxf3 Nf8-+) 60...f2 61.h7 Rh1+ 62.Kxh1 f1Q+ 63.Kh2 Qf2+ 64.Kh3 Qh4+ 65.Kg2 Qxg4+ 66.Kf2 Qh4+ 67.Kf3 Qh3+ 68.Kf2 Kf8!-+;
  • C) 55.Re2!? Nc1 (55...f4 56.Kg2) 56.Re5 (56.Rf2 Ra2 57.Rxa2 Nxa2-+; 56.Re1 Ra2+ 57.Kh1 Nd3 58.Rb1 Nf2+ 59.Kg1 Nxg4 60.Rb7+ Kg8 61.Be7 f4 62.Bxg5 f3-+) 56...Ra2+ 57.Kh3 Nd3 58.gxf5 Nxe5 59.fxe6+ Kxe6 60.Bxe5 Kf5 61.h6 Kg6 62.Kg4 Ra1-+.

Of course these variations cannot be properly analysed, as both sides have an endless plethora of possibilities, but my general feeling is that Black should win.

54...Kg6. I took some time to calculate everything to the end and it was worth it. 55.Bf8 Rc1 56.Ra2. 56.Rxc1 Nxc1 57.Be7 Ne2 58.h7 Kxh7 59.Bxf6 Kg6 60.Be5 Nxc3-+. 56...Rxc3 57.Ra7 Nf4+ 58.Kh2 Rh3+ 59.Kg1 Rg3+! The following variations demonstrate how not to finish a chess game: 59...c3? 60.Rg7+ Kxh6 61.Rd7+=; 59...Rxh6? 60.Rg7#. 60.Kf2 Rxg4 61.h7 Rh4 62.Rg7+ Kf5 63.Rc7 Nd5 0-1. [Click to replay]

Previous articles

Fabiano Caruana, Italian Champion 2007
13.12.2007 – Our recent article which called 15-year-old Fabiano Caruana "Italy's top grandmaster" drew some criticism. While the youngster was definitely very talented, we were told, he was not the Italian champion or even clearly the strongest player in the country. Last week Fabiano put an end to the debate with a resounding three-point victory in the 2007 Italian Championship. Interview.

Who was the future GM? Fabiano Caruana, Italy's top grandmaster!
18.10.2007 – He was born in Miami, Florida, in 1992, went to school in Brooklyn, played in Queens in the Susan Polgar Club, trained with Pal Benko, defeated a GM at the age of ten, moved to Spain, then to Hungary, and made his GM norms at the age of 14 years. Today, at 15, he is the strongest player in Italy and the world's youngest grandmaster. Pictorial answer to our quiz.

Fabiano Caruana – youngest US and Italian GM in history
20.07.2007 – At the "First Saturday" GM tournament in Budapest a 14-year-old lad took first place, scoring 7.0/9, a point and a half ahead of the rest. The 2631 performance gave Fabiano Caruana his final GM norm, making him the youngest US or Italian grandmaster in history (he has dual citizenship). We take you on a tour of Budapest, the tournament and end with an indepth chat with Fabiano.
Fabulous Fabiano
19.05.2003 – He's just ten years old, has trophies that are taller than himself, and travels to Europe to play in chess tournaments. Recently Fabiano Caruana of Brooklyn jetted to Nashville to win his section of the National Elementary Championship. Yet that is not his greatest triumph. Last September Fabiano beat a full-grown, certified grandmaster. More...

Discussion and Feedback Join the public discussion or submit your feedback to the editors


Rules for reader comments


Not registered yet? Register