The Petrov Defense (Russian Game, sometimes spelled Petroff) is the reply 1. e4 e5 2. Nf3 Nf6. Black does not guard the e5-pawn at all. The knight simply attacks e4 in return, so the position stays symmetrical and both sides have a pawn hanging. Among lichess players rated 1000 to 1600, roughly one player in nine answers 2. Nf3 this way, which makes it a minority choice that most club opponents have never studied from White's side.
This course plays the direct version. When White takes on e5, Black takes back on e4 at once with 3...Nxe4, and the game turns on one position: 4. Qe2, which White chooses in more than half of these games. The queen attacks the knight on e4 and stands behind White's own knight on the e-file, aimed at the black king. The right answer is 4...Qe7, mirroring the queen. After 5. Qxe4 d6 the white knight on e5 cannot move, because the moment it steps aside Black captures the white queen on e4 with check. Black wins the piece back and plays for more. From there the course teaches two middlegame plans: a pawn storm with ...g5, ...f5 and ...f4 that traps a bishop on g3, and a queenless ending where ...Bf5, long castling and a knight jump to b4 pick up White's extra pawn.
Play 1...e5 against 1. e4 and, when the king's knight comes to f3, bring your own knight to f6. What happens next is White's decision. At the 1000 to 1600 level about one White player in three takes on e5, and about the same number play 3. Nc3, which invites a Four Knights Game. Nearly one in five plays 3. Bc4, and 3. d3 and 3. d4 turn up in roughly one game in fourteen and one in twenty. The course answers every one of these moves, because in our own users' Petrov games White left the original course's lines in nearly two games out of three, mostly right there on move three.
After 3. Nxe5 Nxe4 the options narrow again. 4. Qe2 is the main test. 4. Qf3, 4. Bc4 and 4. d3 each appear in somewhere between one game in nine and one in fourteen, and 4. Nxf7, the greedy knight grab, turns up about once in thirty-five games. If you arrive through 2. Nc3 Nf6 3. Nf3 or 2. d3 Nf6 3. Nf3, the same chapters apply. White's other second moves (2. Bc4, 2. d4, 2. f4, 2. Qh5) usually take the game somewhere else, and the Open Game course covers them. One exception: 2. Bc4 Nf6 3. Nf3 lands straight in this course's 3. Bc4 lines.
Strengths
Drawbacks
4. Qe2 Qe7 main line: trapping a bishop on g3
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nxe4 4. Qe2 Qe7 5. Qxe4 d6 6. d4 dxe5 7. dxe5 Nc6 8. Bf4 g5 9. Bg3 f5 10. Qe3 f4 11. Bxf4 gxf4
The course opens with this game. White plays one of the two most common moves at every turn, yet the bishop is lost. After 6. d4 dxe5 7. dxe5 Nc6, Black's knight and queen gang up on e5. Once 8. Bf4 defends it, ...g5 chases the bishop to g3, ...f5 threatens ...f4 and there is no way out, so on move 11 Black wins the bishop for two pawns. The same pawn storm works against 10. Qe2, 10. Qd5 and 10. Qa4, and against 8. Bb5 the course castles long before storming.
The 6th-move checks: 6. Bb5+ and 6. Qa4+
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nxe4 4. Qe2 Qe7 5. Qxe4 d6 6. Bb5+ c6 7. Ba4 Qxe5 8. Qxe5+ dxe5
Roughly two White players in five check on move six instead of playing 6. d4. Block 6. Bb5+ with ...c6. If the bishop retreats, take the knight on e5 and trade queens with a level game. If it captures on c6, recapture with the knight and ...dxe5 leaves White a piece down for a pawn. Against 6. Qa4+ the bishop blocks on d7, the knight comes to c6 and Black castles long with a very active position.
Queens off: 7. Qxe5 and the knight raid
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nxe4 4. Qe2 Qe7 5. Qxe4 d6 6. d4 dxe5 7. Qxe5 Qxe5+ 8. dxe5 Nc6 9. Bf4 Bf5 10. Bd3 Bxd3 11. cxd3 Nb4
About four White players in ten trade queens to keep the e5-pawn in an ending. The recipe repeats across the chapter's seven lines: ...Nc6 hits e5, ...Bf5 develops, long castling brings a rook to the open d-file while the white king still sits in the center, and ...Nb4 attacks d3 and c2. Against 10. c3 and 11. Bb5, ...Rd5 forks both bishops, and against a king that wants to castle short, ...Bc5 keeps it on e1.
4. Bc4, answered by ...d5 and ...Qg5
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nxe4 4. Bc4 d5 5. Bb3 Qg5 6. Nf3 Qxg2 7. Rf1 Bg4
Pointing a bishop at f7 looks dangerous, but ...d5 blocks it and gains time. After 5. Bb3 the queen jump to g5 attacks both the knight on e5 and the g2-pawn. If White saves the knight, the queen takes g2 and the pin with ...Bg4 leaves the white kingside in ruins. If White castles instead, the queen takes the knight and ...Nxf2 with ...Bc5 pins the rook. The same queen jump, with ...Be6 first, wins a knight after 3. Bc4 Nxe4 4. Nxe5 d5 5. Qf3.
The e-file ambush against 4. Nxf7
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nxe4 4. Nxf7 Qe7 5. Nxh8 Nc3+ 6. Be2 Nxd1
Taking back on f7 with the king is what most club players do, and in our users' games it lost more often than not. Instead, put the queen on e7. When the knight grabs the rook, ...Nc3+ is a discovered check along the open e-file and the knight takes the white queen on the next move. The identical idea wins against 3. c3 Nxe4 4. Nxe5 Qe7 5. d4 d6 6. Nf3 Nxc3+.
Four Knights with 3. Nc3: the fork trick
1. e4 e5 2. Nf3 Nf6 3. Nc3 Nc6 4. Bc4 Nxe4 5. Nxe4 d5 6. Bxd5 Qxd5 7. Nc3 Qd8
3. Nc3 is White's most popular third move at the band, so it gets its own lines. After 3...Nc6 4. Bc4 the knight takes e4 anyway. Giving it up for a pawn is only temporary, because ...d5 forks the knight and the bishop and wins the piece back. Black ends up with the bishop pair and a comfortable center. Against 4. d4, Black captures, pins with ...Bb4 and follows with ...d5.
3. d4, taking on e4
1. e4 e5 2. Nf3 Nf6 3. d4 Nxe4 4. dxe5 d5 5. Bd3 Nc5 6. O-O Nxd3 7. Qxd3 Be7
When White opens the center with 3. d4, Black takes the e4-pawn and props the knight up with ...d5. Against 4. dxe5 the knight drops back to c5 and swaps off White's most dangerous bishop. Against 4. Nxe5 the plan is ...Nd7 to trade the strong knight on e5, followed by quick castling. Both lines reach calm, equal positions with no weaknesses.
Stafford Gambit, the optional weapon
1. e4 e5 2. Nf3 Nf6 3. Nxe5 Nc6 4. Nxc6 dxc6 5. Nc3 Bc5 6. Bc4 Ng4 7. O-O Qh4 8. h3 Nxf2 9. Qf3 Nxh3+ 10. Kh2 Nf2+ 11. Kg1 Qh1#
Instead of 3...Nxe4, Black can offer a pawn with 3...Nc6, and at club level about a third of Black players already do. After 4. Nxc6 dxc6 the open lines and quick development create threats on f2 and h2, and the course shows how normal-looking castling runs into a mating attack. It is objectively dubious: 6. h3 takes away the g4-square and leaves White better, and if White declines with 4. Nf3 the course takes on e4 and trades queens. Treat it as an extra weapon, not as your main answer to 3. Nxe5.
If you are White, the soundest way to meet the Petrov is 3. Nxe5 d6 4. Nf3 Nxe4 5. d4, which keeps a small, safe edge for White. Against the immediate 3...Nxe4, 4. Qe2 is the right test. If Black retreats with 4...Nf6, 5. Nc6+ wins the queen. If Black finds 4...Qe7, take on e4 and give the pinned knight up with 5...d6 6. d4. After 6...dxe5 7. dxe5 Nc6, be careful with 8. Bf4, because the bishop can be trapped on g3 by ...g5, ...f5 and ...f4. At the 1000 to 1600 level, 8. Nc3 is the move Black handles worst. Against 3...Nc6, the Stafford Gambit, take the knight with 4. Nxc6 dxc6, then play 5. Nc3 Bc5 6. h3 to keep a knight off g4. Accepting the pawn this way leaves you clearly better.
For White
Put the Black knight on e4 under pressure with 4. Qe2, and if Black defends correctly, return the piece with 6. d4 and keep the e5-pawn as long as it is safe. Develop quickly, avoid placing a bishop where ...g5 and ...f5 can trap it, and decide early whether you want queens on (7. dxe5) or an ending with an extra pawn (7. Qxe5). If you prefer to skip the theory, 3. Nc3 or 3. d4 leads to normal positions.
For Black
Against 3. Nxe5, take on e4 and answer 4. Qe2 with 4...Qe7, then 5...d6 so the pinned knight falls. After 6. d4 dxe5, aim every piece at the e5-pawn: ...Nc6, ...Bf5 or ...Bg4, and long castling so a rook arrives on the d-file. If a bishop lands on f4, push ...g5 and ...f5 to trap it. Against White's other third moves, develop, strike with ...d5 and trade off the attacking pieces.
Frank Marshall, Anatoly Karpov and Vladimir Kramnik all kept 2...Nf6 in their Black repertoires, and Fabiano Caruana relied on it against Magnus Carlsen in their 2018 title match. The name goes back further, to Alexander Petrov, a leading Russian player of the early 19th century who analysed the move together with Carl Jaenisch. Most databases file it as the Russian Game, and older English books spell it Petroff. Club games rarely look like those grandmaster ones. In our own users' games, Black answered 4. Qe2 with 4...Qe7 in only 46 of 242 tries, and closing that gap is what this course is built for.