The Petroff Defence for beginners
An Exeter Junior Chess Club booklet
Edition 2.,May, 11
Karpov, The Open Game in Action
Kasparov/Keene, Batsford Chess Openings
Levy/Keene, An Opening Repertoire for the Attacking Club Player
walker, Chess Openings for Juniors
Various magazines and other books
a. Symmetry in chess 3
b. The isolated Queen's pawn 4
A. The Classical Variation, 3. Nxe5 5
3... d6 5
1. An alternative on move 4 for White: 4. Nxf7, The Cochrane Gambit 5
2. Main line with 4. Nf3 6
1.1 An alternative on move 5 for White: 5. Qe2, Lasker Variation 6
1.2 The main line with 5. d4 7
1.2 (a) 6...Bd6 Marshall Gambit 7
1.2 (b) 6...Be7 Main Line 8
1.2 (b1) Main line with 8. c4 8
(b1-i) 8... Bg4 9
(b1-ii) 8... Nb4 9
(b1-iii) 8... Nf6 10
1.2 (b2) Main line with 8. Re1 11
(b2-i) 8... Bf5 11
(b2-ii) 8...Bg4 12
(1) Modestly, 9. c3 12
(2) Ambitiously, 9. c4 13
B. The Steinitz Variation, 3. d4 14
(a) 3...exd4 14
(b) 3...Nxe4 15
(b1) 4. dxe5 15
(b2) 4. Bd3 15
(b2-i) 5...Nd7 15
(b2-ii) 5...Bd6 16
C. White tries to get into the Four Knight's Game, 3. Nc3 17
D. The Boden-Kieseritsky Gambit, 3. Bc4 18
Example games 19
Lasker - Pillsbury, St. Petersburg, 1895 19
Kupreichik - Mikhailchishin, Kujbyshev, 1986 20
Dolmatov - Makarichev, Palma, 1989 21
The Petroff is a quiet success story in chess - an old and often forgetten variation which can give Black a safe way of working for equal chances in an open game, while avoiding White's favourite lines. This booklet tries to give an overview of the Petroff from both sides. It has a reputation for being "dull and a draw", but can offer exciting attacking possibilities for either side. There are a couple of general themes I'd like to deal with immediately: symmetry, and the isolated Queen's pawn.
a. Symmetry in chess
Symmetry usually means equality, and likelihood of a draw - no good if you are trying to win. The symmetrical Petroff has reputation for drawishness, but that is not really justified. Look at the Illustrative Games if you still think the Petroff is dull! And what is drawn at Grandmaster level can still have a lot of meat left in it for amateur players (like the Colle system).
For example, after 1. e4 e6 2. d4 d5 3. exd5 exd5 (the Exchange Variation of the French Defence), we have a symmetrical pawn formation with the likelihood of exchanges of the major pieces down the e-file. In fact, Black wins lots of games in this line, perhaps because it is a line chosen by weak White players seeking a draw against strong Black players. (Perhaps the best way to play for a draw is to go for a win!)
This line of the Petroff (1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d5 d5) has perhaps the worst reputation for drawishness, but let's have a look. Black has an aggressive Knight on e4 - it even looks as if Black is a move ahead. White is not necessarily trying to draw in this variation, and it is White to move. White will try to undermine the Ne4 with moves like Bd3, Re1 and c4; Black can try to castle Queen's-side and belt down the other wing. All this can make for exciting and tense struggles in an open game, which might be a lot better than suffering White's favourite line of the Ruy Lopez.
b. The isolated Queen's pawn
Several lines of the Petroff can lead to an isolated Queen's pawn (IQP) position, if White plays to undermine the Black Knight on e4 with c4, and Black later plays ...dxc4. White will have the last remaining central pawn on d4, a situation which is known from many other openings. For example:
The isolated Queen's pawn in the Semi-Tarrasch Defence to the Queen's Gambit
1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Nf3
4... c5 5. cxd5 Nxd5 6. e3 Nc6 7. Bd3 Be7 8. O-O O-O 9. a3 cxd4 10.
The isolated Queen's pawn in the Caro-Kann Defence, Panov-Botvinnik Attack
1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4
4... Nf6 5. Nc3 e6 6. Nf3 Be7 7. cxd5 Nxd5 8. Bd3 Nc6 9. O-O O-O 10. a3
Exactly the same!
And in the Petroff:
So the IQP is important to know about, although in the Petroff Black doesn't have the Bc8 stuck behind the e6 pawn. White's ideas will still be the same: develop quickly, use the extra space to dodge about, avoid exchanges (to keep Black cramped), play Ne5, look for a chance to start a King's-side attack using the light-squared bishop, maybe throw in Rf1-e1-e3-h3.
So, let's have a look at some variations.
A. The Classical Variation, 3. Nxe5
Black must now play 3...d6. Why not 3... Nxe4? White answers 4. Qe2!
Now 4... Qe7...
[Not 4... Nf6?? 5. Nc6+!
winning the Q]
...and now 5. Qxe4 d6 6. f4 dxe5 7. fxe5 wins a pawn.
So, from the first diagram in the column:
Now, White almost always plays the main line with 4. Nf3, but there are two alternatives: 4. Nxf7, the Cochrane Gambit, and 4. Nc4. I have never seen this last move played, and I don't think it is dangerous, but it does deny Black the natural ...Bg4 pin. BCO recommends 4. Nc4 Nxe4 5. d4 d5 6. Ne3 Qf6 and 7...Be6 with an early ...O-O-O in mind.
So that leaves us with the Cochrane Gambit to examine before we go down the main line.
1. An alternative on move 4 for White: 4. Nxf7, The Cochrane Gambit
Looks crazy? Looks promising? Your chess personality may make you feel confident as White or Black in this position, but either way, you must know about it.
4... Kxf7 5. d4
Black has tried:
[5... Nxe4? 6. Qh5+ g6 7. Qd5+ Be6 8. Qxe4]
[5... d5?! 6. e5 Ne8 7. Bd3 g6 8. h4]
So Black should prefer the safer:
5...g6 6. Nc3
Now BCO recommends the odd-looking:
Other tries seem inadequate, e.g.
6... Bg7 7. Bc4+ Be6 8. Bxe6+ Kxe6 9. f4 Kf7 10. e5 Ne8
[or 10... Re8 11. O-O Nc6 12. exf6 Qxf6 13. Nb5 Re7 14. c3 o'neill-solomon 1985]
11. Qf3 c6 12. Ne4+-
... with better chances for White (Makropoulos-Toth 1981).
2. Main line with 4. Nf3
Black should take the chance to recover the pawn with
white has a choice now, between 5.Qe2, the Lasker Variation, leading to an early endgame, or 5. d4, the main line.
1.1 An alternative on move 5 for White: 5. Qe2, Lasker Variation
This pin is not dangerous here with the White Knight on f3, just awkward for the moment.
5... Qe7 6. d3 Nf6 7. Bg5 Nbd7 8. Nc3 Qxe2+ 9. Bxe2 Be7=
White cannot claim any advantage in this endgame, although Black still needs to get fully developed. This variation is why the Petroff is no good when you have to win as Black against a player who knows this line.
1.2 The main line with 5. d4
Now Black can get an equal stake in the centre with:
And White brings the Bf1 to its best square, so the King can castle.
Like White, Black should plan to get the K off the open e-file as soon as possible, and Black has a choice of two squares for the Bf8:
(a) 6...Bd6, a move of Frank Marshall - which looks good but is perhaps trying a bit too hard - or the safe (b) 6...Be7, sealing the e-file.
1.2 (a) 6...Bd6 Marshall Gambit
Let us follow the play after the natural moves:
7. O-O O-O 8. c4 Bg4!?
This is Marshall's idea.
9. cxd5 f5 10. Re1?
One natural move too many.
10... Bxh2+ 11. Kxh2 Nxf2 12. Qc2 Nxd3 13. Qxd3 Bxf3
winning, because if Qxf3, then ...Qh4+ will pick up the rook on e1. Instead of falling for this trap White should play:
10. Nc3 Nd7 11. h3 Bh5 12. Nxe4 fxe4 13. Bxe4 Nf6 14. Bf5 +-
...with advantage, as in the game Alexander- Mattinson from 1938.
So the rather boring truth about this line seems to be that back on move eight Black should play instead
when theory knows of
9. Nc3 Nxc3 10. bxc3 Bg4 11. cxd5 cxd5 12. Rb1 Nd7 13. h3 +=
Now, this position is solid but offers so few active possibilities for Black that we should prefer as Black 6...Be7.
1.2 (b) 6...Be7 Main Line
And now play often goes:
7. O-O Nc6
...when White has a choice:
Hit at the centre with 8.c4 (b1), or develop with 8.Re1 (b2). (White can also play 8.Re1 and 9.c4, but this allows 8...Bg4.)
1.2 (b1) Main line with 8. c4
What are the alternatives here? Black does not have time for (b1-i) 8...Bg4. The position of the knight is now so wobbly that Black should move it - both
(b1-ii) 8...Nb4, and
(b1-iii) 8...Nf6 have been tried.
(b1-i) 8... Bg4
Like I said, Black really doesn't have time for this.
9. cxd5 Qxd5 10. Nc3 Nxc3 11. bxc3 Bxf3 12. Qxf3 Qxf3 13. gxf3
The two bishops and strong centre give White the advantage in this
ending; White has natural ideas of Rb1 and Re1/Be3 which are difficult
(b1-ii) 8... Nb4
Now White usually preserves the bishop with:
[less promising seems 9. cxd5 Nxd3 10. Qxd3 Qxd5 11. Re1 Bf5 12. Ne5 g6= Belyavsky-Smyslov 1986 ]
After 9.Be2 Black has tried
(1) 9... Be6
A line quoted by Karpov.
10. c5 Nc6 11. Qa4 a5 12. Ne5+=
With an advantage for White.
(2) 9... O-O
10. Nc3 Be6
The Grandmasters are still having a think about this one. It's as difficult for White to get an advantage as it is for Black to equalise!
BCO quotes games with 11. Be3, 11. a3, and 11. Nxe4!?. Having just seen the Karpov line above, you might even fancy 11. c5!?
(3) 9... dxc4
This surrender of the centre doesn't seem to work out too well for Black.
10. Bxc4 O-O 11. Nc3 Nd6
...as in a couple of games from 1985, seems to leave White with a plus in this IQP position. White will plan to complete development, and then play Ne5 with hopes of an attack.
(b1-iii) 8... Nf6
This is a safe and simple move. Play may continue:
9. Nc3 O-O 10. h3 dxc4
11. Bxc4 Na5 12. Bd3 Be6
13. Re1 Nc6 +=
White is held to have a slight advantage in this well-explored position due to the extra central space, although in the ending the d-pawn may become exposed.
1.2 (b2) Main line with 8. Re1
Black has a choice of squares for the last minor piece:
(b2-i) 8... Bf5
Karpov has shown that this is inaccurate.
9. c4 Nb4 10. Bf1
[not 10...dxc4 11. Nc3 +- with clear advantage: Karpov-Portisch]
10... O-O 11. a3 Nc6 12. cxd5 Qxd5 13. Nc3 Nxc3 14. bxc3 Bg6+=
...with a small advantage for White: Karpov-Portisch 1982. Karpov is rather behind in development at the moment but he is threatening to unravel with c4, dominating the centre.
This very well-known position has a reputation for drawishness (which is not really true at amateur level). White can play modestly (1) 9. c3, or ambitiously (2) 9. c4.
(1) Modestly, 9. c3
This rather modest move can lead to an interesting gambit which Black should not be afraid to play
9 ... f5 10. Qb3 O-O 11. Nbd2
(See Lasker-Pillsbury below for 11.Bf4.)
11...Kh8 12. Qxb7 Rf6 13. Qb3 Rg6=
(2) Ambitiously, 9. c4
Black's knight is again too wobbly in the centre.
9... Nf6 10. cxd5
[10... Nxd5 is also known: 11. Nc3 O-O 12. h3 Be6 13. a3 Nxc3 14. bxc3 Bf6 15. Bf4 g6 16. Be4 Na5 17. Ne5 c6 18. Qf3+=
After this long sequence, White has a small advantage which will be long-lasting: Tseshkovsky-Barua 1986]
11. Qxf3 Qxd5 12. Qh3 Qxd4 13. Nc3 Rd8 14. Bf5 h5 15. Qg3 Kf8 16. Be3 Qb4 17. a3 Qa5 18. Qf3 g6 19. Bc2 h4
Another long variation, rather a messy one this time, which BCO assesses as unclear. I prefer White here: Velimirovic-Kurajica 1984]
B. The Steinitz Variation, 3. d4
This is the main alternative to 3. Nxe4 and in the late 70s was thought to give White good chances of advantage. That was because everybody then played (a) 3..exd4. More recently Black has been equalising with (b) 3...Nxe4.
Now White must try
4. e5 Ne4 5. Qxd4 +=
It has been thought for a long while that White has the edge here.
5... d5 6. exd6 Nxd6 7. Nc3 Nc6 8. Qf4 g6
[or 8... Bf5 when BCO gives 9. Bb5 Qe7+ 10. Be3 Nxb5 11. Nxb5 Qb4+ 12. Qxb4 Bxb4+ 13. c3 Bd6 14. Nxd6+ cxd6 15. O-O-O O-O-O 16. Nd4 Be6 17. b3 a6 18. Ne2+=
when Black's weak d-pawn gives White the better endgame (Gurgenidze-Beim, 1982) ]
Natural moves to follow are:
9. Bd3 Bg7 10. O-O O-O+=
All the central pawns have disappeared, but White is a little better placed (Short-Murei 1982-3)
Now White usually tries 4. Bd3, but we ought to have a quick look at 4.dxe5 before accepting this as best.
(b1) 4. dxe5
The game may now go:
4...d5 5. Nbd2 Nxd2 6. Bxd2 Be7 7. Bf4 c5 8. c3 Nc6 9. Bd3 Be6=
This position is about equal (Short-Seirawan 1986).
(b2) 4. Bd3
This is probably the best move here as on move 6 of the Classical Variation.
4... d5 5. Nxe5
This variation is both like, and unlike, the Classical. Black's main tries are (b2-i) 5...Nd7 and (b2-ii) 5...Bd6, although 5...Be7 and even other moves have been played.
White has tried all sorts of moves here:
(1) The cheeky 6. Nxf7
(2) The safe 6. Nxd7
(3) The logical 6. Qe2
I'll just give one quick example of each.
6. Nxf7 Kxf7 7. Qh5+ Ke7 8. Qxd5 Ndf6 9. Qb3 Be6 10. Qxb7 Bd5 11. Qa6 Kf7
Nenashev-Baikov 1985: "unclear" says BCO!
6. Nxd7 Bxd7 7. O-O Qh4 8. c4 O-O-O 9. c5 g5 10. Nc3 Bg7 11. Ne2 f5 12. f3 Rhf8 13. a4 Rde8°
Again, "unclear" says BCO (Sveshnikov-Makarichev, 1987). It does show how Black can play for a win if required.
6. Qe2 Nxe5 7. Bxe4 dxe4 8. Qxe4 Be6 9. Qxe5 Qd7 10. O-O O-O-O 11. Be3 Bb4 12. Nc3 f6 13. Qg3 Bxc3 14. bxc3 h5 15. h4 g5 16. f3 Rdg8-+
Not at all unclear: Black won in the famous upset of Karpov-Larsen, 1980
The balance looks a little tense, but Black seems to be able to hold.
6. O-O O-O 7. c4 Bxe5 8. dxe5 Nc6
This position offers equal chances.
C. White tries to get into the Four Knight's Game, 3. Nc3
Now, after 3... Nc6 we get into those genuinely dull, drawish variations of the Four Knights Game. I don't care if Short and Nunn have played it, it's a poor line for juniors (in my opinion). It tends to be stodgy and blocked, you probably won't enjoy it and it doesn't give you the sort of training in open games that you need.
Thankfully Janowski found that Black can play
BCO now gives
4. Nxe5 O-O 5. Be2 Re8 6. Nd3 Bxc3 7. dxc3 Nxe4 =
...with equal chances in a more open game. White's two bishops don't frighten me.
D. The Boden-Kieseritsky Gambit, 3. Bc4
Now Black can steer for the equally promising lines of the Two Knights Defence with 3...Nc6, but to look at that properly will take another booklet.
Two Knight's Defence
Black has no real reason to avoid:
white can try other moves, but the usual try is
4. Nc3 Nxc3 5. dxc3 Be7
Black has no weaknesses, and White will find it difficult to get through. Black will want to get the two bishops working, but must be careful to keep lines closed until development is complete. White has tried various plans in this position but they are all pretty unconvincing.
Lasker - Pillsbury, St. Petersburg, 1895
1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d4 d5 6. Bd3 Be7 7. O-O Nc6 8. Re1 Bg4 9. c3 f5 10. Qb3 O-O 11. Bf4
White allows the King's-side to be weakened.
11... Bxf3 12. gxf3 Ng5 13. Kg2 Qd7 14. Qc2 Ne6 15. Bc1 Bd6 16. Nd2 Rae8 17. Nf1
Black's pieces are much better placed than White's. If White was hoping to tempt Black into an unsound attack, he has only tempted Pillsbury into a sound one.
17... Nexd4 18. Qd1 Rxe1 19. Qxe1
Black now starts a king hunt, sacrificing a piece for a couple of pawns.
19... Nxf3 20. Kxf3 f4 21. Qd1 Ne5+ 22. Ke2 Qg4+ 23. Kd2 Qxd1+ 24. Kxd1 Nxd3
Black has recovered his piece, and still has his pawns. White is lost.
25. Ke2 Ne5 26. f3 Re8 27. b3 Ng4+ 28. Kd2 Ne3 29. Bb2 Ng2 30. h3 Bc5 31. Nh2 Bf2 32. c4 dxc4 33. bxc4 h5 0-1
If anything, White's position has gone from bad to worse.
Kupreichik - Mikhailchishin, Kujbyshev, 1986
1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. c4!? Nc6 6. d4?!
[Mikhailchishin gives 6. Be2 d5 7. Nc3 Be6]
6... d5 7. Nc3 Bb4 8. Qc2 Qe7 9. Be3 Bf5
Black has the advantage: White cannot easily complete development and his poor Queen is embarrassed.
[10. Bd3?! dxc4 11. Bxc4 Ng3-+]
10... Nxc3 11. bxc3 Ba3 12. Qd2 Nb4!!
White is busted. He tries to hang on with Ke2 but it doesn't take much longer with the K stuck in the middle.
13. Ke2 Nc2 14. Rd1 dxc4 15. Ne5 O-O 16. f3 b5 17. g4 Nxe3 18. Qxe3 Bd7 19. Kf2 Rae8 20. Qd2 Bc8 21. Re1 Qh4+ 22. Kg2 f6 0-1
Janowsky - Marshall, 1912
1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d4 d5 6. Bd3 Bd6
6...Be7 is honestly better
7. c4 O-O 8. cxd5
Castling was a lot better here
8... Bb4+ 9. Kf1 Qxd5 10.
Qc2 Re8 11. Nc3 Nxc3 12. bxc3
12... Qxf3!! 13. cxb4
[13. gxf3 Bh3+ 14. Kg1 Re1+ 15. Bf1 Rxf1#]
13... Nc6 14. Bb2 Nxb4 15. Bxh7+ Kh8 16. gxf3 Bh3+ 17. Kg1 Nxc2 18. Bxc2 Re2 19. Rc1 Rae8 20. Bc3
[brilliant, but 20... Rxc2 21. Rxc2 Re6 was easier]
[21. fxe3 Rg2+ 22. Kf1 Rxc2+ 23. Ke1 Rxc1+ 24. Kd2 Rxh1]
21... Rxf3 22. Bd1 Rf6 0-1
Dolmatov - Makarichev, Palma, 1989
1. e4 e5 2. Nf3 Nf6 3. d4 Nxe4 4. Bd3 d5 5. Nxe5 Nd7 6. Nxd7 Bxd7 7. O-O Qh4 8. c4 O-O-O 9. c5 g5 10. f3!? Nf6 11. Be3 Rg8 12. Nc3 g4
Black plays with great energy and has a fierce attack.
[13. Bf2 is better]
13... g3 14. hxg3 Rxg3
[not 15. Bf2, as 15....Rxg2+ 16. Kxg2 Qh3+ 17. Kg1 Bd6 18. cxd6 Rg8+ wins]
15... Bxc5 16. dxc5 Rdg8 17. Rad1 d4 18. c6 dxe3 19. cxd7+ Kd8 0-1
Black has various threats - ...Rh3, ...Rxg2, ...Qf2+ and so on - which cannot all be met.
"Dull and a draw", eh?