His nickname of Iron Tigran gives a hint as to the type of repertoire that can be constructed from his games in ChessBase's 2020 Mega database.
White
Open 1.d4 aiming for a standard Queen's Gambit set-up with 1...d5 2.c4 e6 3.Nc3 Nf6, at which point Petrosian scored 67% with his most-popular continuation, 4.Nf3, but a massive 86% with the Exchange Variation: 4.cxd5 exd5 5.Bg5. In the mainline position reached by various move-orders, eg 5...Be7 6.e3 0-0 7.Bd3 c6 8.Qc2 Nbd7, he scored 88% with 9.Nge2.
Against the Alatorsev Variation: 3...Be7 he usually played 4.Nf3 Nf6, scoring 70% when he continued 5.Bf4.
Against the Queen's Gambit Accepted: 2...dxc4 he scored 69% with 3.Nf3.
Against the Slav: 2...c6 he usually played 3.Nf3 Nf6 4.Nc3 (Petrosian often reversed White's last two moves), scoring 77% when he met 4...e6 with 5.e3. Against 4...dxc4 he scored 88% with the direct 5.e4.
After 1...Nf6 2.c4 e6 Petrosian, by a very small margin, preferred 3.Nf3 to 3.Nc3, but scored six percentage points higher with the latter, which fits in with the repertoire choice of the Exchange Variation.
Against the Nimzo-Indian: 3...Bb4 he scored 74% with 4.e3. He usually met 4...0-0 and 4...c5 with 5.Bd3, but against 4...d5 he always played 5.a3. Against 4...b6 he usually played 5.Nf3 but did much better (75% as opposed to 50%) with 5.Nge2.
Against the Grünfeld: 2...g6 3.Nc3 d5 Petrosian scored a very respectable 78% with 4.Nf3 but an amazing 90% with the somewhat unusual 4.Bf4!? The main line runs 4...Bg7 5.e3 0-0, at which point Petrosian liked another slightly unusual continuation, 5.Qb3!? (see below for how he used this system to beat Korchnoi at the 1961 USSR Championship).
Against the King's Indian: 3...Bg7 4.e4 d6 Petrosian scored 86% with the Sämisch: 5.f3, rising to 92% when he met 5...0-0 with 6.Be3 and to 93% when he met the main continuation 6...e5 with 7.d5.
Against the Benoni: 2...c5 he scored 89% with 3.d5, rising to 100% the five times he reached the mainline position after 3...e6 4.Nc3 exd5 5.cxd5 d6 6.e4.
Petrosian only faced the Benko Gambit: 3...b5 three times, drawing with 4.cxb5 but scoring a win and a draw with 4.Nf3.
He varied his answers the few times he faced the Dutch: 1...f5 but won both games when he played the modern favourite 2.Bg5.
Against 1...e6 Petrosian scored 83% with 2.d4 d5 3.Nc3 but 2.c4, which was his usual reply, is a simpler repertoire fit. If Black then went for a delayed Dutch: 2...f5 - this only happened twice to Petrosian in Mega20 - he played 2.g3, winning both games.
Black v 1.e4
Petrosian played the French and the Siclian: 1...c5 with almost equal frequency, but scored six percentage points higher with the latter. Against 2.Nf3 he usually played 2...d6, varying his replies after 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 but most often using the Najdorf: 5...a6, with which he scored 72%.
Against 6.Bg5 he came to prefer the slightly unusual 6...Nbd7. Only one opponent continued with today's main line, 7.f4, which Petrosian beat by grabbing a 'posioned pawn' with 7...Qb6 8.Qd2 Qxb2. Much more popular among Petrosian's opponents was 7.Bc4, which he always met with 7...Qa5, scoring 63% with 8.Qd2 e6.
Against 6.Be2 he scored 77% with 6...e6.
Petrosian only faced other continuations a handful of times, but for it is worth he beat 6.f4 with 6...Qc7 (twice) and 6...e5 (once), scored a win and a draw with 6.Bc4 e6 7.Bb3 Be7 and beat 6.g3 with 6...e5 7.Nde2 Be7.
He only faced 4.Qxd4 twice, winning both times by playing 4...Bd7 and 5...Nc6.
Against the Moscow: 3.Bb5+ he scored 60% with 3...Bd7.
Against the Closed: 2.Nc3 Nc6 3.g3 Petrosian counter-fianchettoed with 3...g6 4.Bg2 Bg7, scoring 62% with 4.d3 d6.
He only faced 2.c3 once(!) in Mega20, replying 2...Nf6.
Black v 1.d4 and Others
Against 1.d4 Petrosian usually aimed for a Nimzo-Indian: 1...Nf6 2.c4 e6 3.Nc3 Bb4, normally meeting 4.e3 with 4...0-0. Against both 5.Bd3 and 5.Nge2 he did best with 5...d5 but against 5.Nf3 he played 5...c5.
Against 4.Qc2 he again scored best by first castling, meeting the main reply 5.a3 with 5...Bxc3+ 6.Qxc3 b6.
Against 4.Nf3, the only other move he faced more than two times, he usually replied 4...c5.
When White avoided the Nimzo-Indian with 3.Nf3 Petrosian scored 57% with the Queen's Indian: 3...b6, rising to 62% when he met 4.g3 - the only move he faced more than twice - with 4...Ba6.
Against 2.Nf3 he usually played 2...d5, meeting 3.c4 with a delayed Queen's Gambit Accepted: 3...dxc4. After 4.e3 e6 5.Bxc4 he nearly always played 5...c5. After 4.Nc3 he transposed into a Slav set-up with 4...c6, scoring 70% when he met the main line 5.a4 Bf5 6.e3 e6 7.Bxc4 Bb4 8.0-0 with 8...Nbd7. Petrosian only face 5.e4 once in Mega20, meeting the main line 5...b5 6.e5 Nd5 7.a4 with 7...e6.
Against the English: 1.c4 he usually played 1...Nf6, scoring 59% when he met 2.Nc3 with 2...e5. Against 2.Nf3 he usually replied 2...e6, varying in whether he met 3.Nc3 and 3.g3 with 3...d5 or 3...b6. If White transposed into a queen's pawn opening with 3.d4, the repertoire choice is the Queen's Indian: 3...b6. He only reached the position after 2.g3 three times, once playing 2...g6 and twice playing 2...c6.
Against 1.Nf3 Petrosian usually replied 1...Nf6, when transpositions are likely to lines covered above. If White continued 2.g3 he scored 75% with 2...d5 3.Bg2 c6, meeting 4.0-0 with 4...Bg4.
He usually met 1.g3 with 1...d5, playing the same move the only time he faced 1.b3.
Tigran V Petrosian - Victor Korchnoi
USSR Championship (Moscow) 1961
Grünfeld Defence
1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bf4!? Bg7 5.e3 0-0 6.Qb3!?
The main move is 6.Rc1. Also more popular than the text are 6.Nf3 and 6.cxd5.
6...dxc4
The main line runs 6...c6 7.Nf3 Qa5 8.Nd2 Nbd7 9.Be2 Nh5!? 10.Bxh5 dxc4, when the analysis engines Stockfish11 and Komodo11.01 reckon that after 11.Nxc4 Qxh5 12.0-0 White's lead in development is worth slightly more than Black's bishop-pair.
7.Bxc4 Nbd7 8.Nf3 Nb6 9.0-0!?
At the 1952 USSR Championship (also held in Moscow) the game Oleg Moiseev - Korchnoi saw 9.Be2 Be6 10.Qc2 Rc8, after which the engines reckon White could have got an advantage with 11.e4. However Petrosian and Korchnoi may both have been aware of a 1953 Argentine game in which 10...Nbd5!? seemed to improve Black's play.
The text prioritises development over the bishop-pair, which belies any impression people might have that Petrosian was adverse to dynamic play.
9...Nxc4 10.Qxc4 c6 11.Be5!?
Ensuring that if Black wants to keep the bishop-pair he will have to move his dark-square bishop before he moves his knight.
11...Bh6!?
'Unpinning' the knight. Grandmaster Ratmir Kholmov in Mega20 gives this an exclamation mark, but the engines reckon Black should get on with development.
12.Rfd1
Kholmov points out that 12.Bxf6 exf6 13.e4 can be met by 13...f5. The point is that an exchange of pawns opens the position for the bishops, while 14.e5 Be6 is very pleasant for Black.
12...Be6 13.Qb4 b6 14.Qa4 Qe8 15.Rac1 Nd5 16.Nd2 Qd7 17.Qa3 Rac8!?
Wrong rook? Kholmov suggests so, but Korchnoi seems to have thought the king's rook was best reserved for the d file.
18.Ne2?!
This may be taking manoeuvring too far. The engines suggest trying to the fix the c pawn with 18.b4.
18...a5
Kholmov points out the strength of 18...Bg4! The obvious 19.Re1 allows 19...c5! as 20.dxc5 runs into 20...Nxe3!, eg 21.fxe3 Qxd2 22.Nf4 Rxc5. Better may be 19...Nf3!? but White's king is clearly at risk after 19...Bxf3.
19.Nc4 Qa7 20.b3 Rfd8 21.h3 Qa6 22.Qb2 f6 23.Bg3 Bf7 24.Kh2 Rd7 25.Nc3 Nxc3 26.Qxc3 Rcd8 27.Rc2!?
After mutual manoeuvring Petrosian tempts his opponent into playing the move he has clearly been lining up.
27...e5?
The game would be roughly equal, according to the engines, after a less-committal move such as 27...a4.
28.Rcd2 e4
Or 28...exd4 29.Rxd4 Rxd4 30.Rxd4 Rxd4 31.Qxd4, when both b6 and f6 are hanging. Black is therefore obliged to play 31...Bxc4, but the engines' 32.Qd8+! (Kholmov's 32.bxc4 is also strong) is decisive, for example 32...Kf7 33.Qd7+ Kg8 34.Qe8+ Kg7 35.Bd6! Bg8 (only move) 36.Bf8+ Kh8 37.Bxh6 Qb7 38.Qd8 Qf7 (only move) 39.Qxb6 and the black queenside collapses.
29.d5! Bxd5 30.Qxf6
Sacrificing the exchange with 30.Rxd5!? may be even better.
30...Bg7 31.Qg5 Rf8
Curiously Black does not seem to have a reasonable move, for example the engines reckon best is 31...b5 32.Ne5 Bxe5 33.Bxe5 Rf8 but have White as much better after 34.f4 or 34.f3.
32.Qg4 Rdf7 33.Nd6 Re7
 |
| How should White proceed? |
*****
*****
*****
*****
*****
34.Rxd5!?
Petrosian was famous for his positional exchange sacrifices, and he may well have been itching to make this capture for several moves.
34...cxd5 35.Qg5 Bf6 36.Qxd5+ Kh8 37.Nxe4 Bg7 38.Nd6 Qa8
Trying to get queens off, after which Black would have drawing chances.
39.Qb5 Qa7 40.Nc4 1-0
Black lost on time, but the engines reckon White's advantage is decisive anyway.
No comments:
Post a Comment