1. e4 e6
The French Defense is a chess opening for Black that answers 1. e4 with 1... e6, preparing 2... d5 to challenge White’s center immediately. After the main sequence 1. e4 e6 2. d4 d5, Black has a solid pawn chain and a clear target: White’s center. Compared to its structural sibling the Caro-Kann (1... c6 and 2... d5), the French keeps the light-squared bishop at home behind the e6-pawn but gains faster, sharper counterplay with ... c5 against White’s pawn on d4.
The defining position arises after 1. e4 e6 2. d4 d5. Black’s first move looks modest, but it sets up an immediate strike: the d5-pawn attacks e4 on move two, and White must make a real decision right away. White’s four serious options define the whole opening: 3. e5 (the Advance) grabs space, 3. exd5 (the Exchange) releases the tension, 3. Nd2 (the Tarrasch) defends e4 while avoiding the pin ... Bb4, and 3. Nc3 (the main line) defends e4 with the most active piece. Against every one of them, Black’s counterplay follows the same recipe: hit the d4-pawn with ... c5, develop pieces toward White’s center, and let White’s space advantage become a liability.
1. e4 e6 2. d4 d5 3. e5 c5 4. c3 Qb6 5. Nf3 Nc6 6. Be2 cxd4 7. cxd4 Nge7
White grabs space and fixes the pawn chain; Black attacks its base on d4 with everything available. The queen on b6 hits both d4 and the b2-pawn White can struggle to keep defended, and the knight reroutes via e7 to f5 for a third attacker on d4. If White ever runs out of defenders, the center simply falls, and the b2-pawn is sometimes free to take along the way.
1. e4 e6 2. d4 d5 3. exd5 exd5 4. Nf3 Nf6 5. Bd3 Bd6 6. O-O O-O 7. Bg5 Bg4
White releases the central tension and frees Black’s problem bishop in one move: the c8-bishop now develops easily to g4. The structure is symmetrical, so the fight is about piece activity: Black mirrors White’s most active setup, contests the e-file, and plays for ideas like ... Nc6-b4 to trade off White’s good bishop on d3. Equality is comfortable; winning chances come from outplaying White piece by piece.
1. e4 e6 2. d4 d5 3. Nd2 c5 4. exd5 Qxd5 5. Ngf3 cxd4 6. Bc4 Qd6 7. O-O Nf6
White’s 3. Nd2 defends e4 while dodging the pin ... Bb4, at the cost of blocking the c1-bishop and loosening the grip on d4. Black strikes immediately with 3... c5, and because no knight sits on c3, the queen can recapture on d5 without being harassed. Black gets free development and open lines for every piece; White gets activity against the slightly exposed queen. It is one of the most direct, principled ways to meet the Tarrasch.
1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. Bg5 Bb4 5. e5 h6 6. Bd2 Bxc3 7. bxc3 Ne4
Against White’s most natural setup, Black answers the pin with a counter-pin: 4... Bb4 ignores the threat to f6 and creates one against c3. After the forcing sequence Black trades on c3, plants a knight on e4, and damages White’s queenside pawns. The middlegame is a genuine fight: White attacks on the kingside, often with Qg4, while Black targets the wrecked c-pawns and the dark squares White’s bishop trade left behind.
1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Nfd7 5. f4 c5 6. Nf3 Nc6 7. Be3 cxd4 8. Nxd4
White pushes 4. e5 to gain space and backs the chain up with f4. Black retreats the knight to d7 by design: from there it supports the standard breaks ... c5 against d4 and, later, ... f6 against e5. Piece pressure on d4 with ... Nc6, ... cxd4 and a quick ... Bc5 or ... Qb6 forces White to defend accurately; the big white center is a target as much as an asset.
If you play 1. e4 and keep running into the French, pick one of White’s four third moves and learn its actual plan; random development hands Black exactly the game the opening wants. 3. Nc3 is the most ambitious: it defends e4 with a piece, keeps every option open, and leads to the richest fights, but you must be ready for both 3... Nf6 and the sharp pinning lines. 3. Nd2, the Tarrasch, trades some ambition for comfort by sidestepping ... Bb4; meet 3... c5 with accurate development and play against Black’s early queen sortie. The Advance with 3. e5 is the most thematic try: you get real kingside space and attacking chances, but the entire line is a referendum on the d4-pawn: count the attackers and defenders on d4 every single move, and never lose b2 for nothing. The Exchange (3. exd5) is solid and easy to learn but gives Black free development and a symmetrical game, so choose it for safety, not for an edge. Whichever line you pick, two evergreen ideas pay off: exploit Black’s sleepy c8-bishop by keeping the position closed on the light squares, and use your space advantage to attack the kingside before Black’s ... c5 and ... f6 breaks dissolve your center.
Use the extra space. In the Advance and Steinitz structures, defend d4 reliably (c3, Nf3, sometimes Be3), then attack on the kingside where the chain points; f4-f5 against e6 is the classic lever. Keep the position closed so Black’s c8-bishop stays bad, and meet ... c5 and ... f6 breaks with precise calculation rather than reflex captures.
Attack the center immediately: ... c5 against d4 is the soul of the French, often paired with ... Qb6 and ... Nc6 to overload the defenders. In blocked positions, prepare ... f6 to chip at e5. Solve the c8-bishop with trades or the ... b6 and ... Ba6 route, and remember the structure forgives almost everything except passivity.
The French Defense takes its name from an 1834 correspondence match between the cities of London and Paris, in which the Paris side adopted 1... e6, and the label stuck. It has been a fixture of serious chess ever since: one of the oldest and most respected answers to 1. e4, with a reputation for soundness and fighting spirit rather than fashion. Generations of strong players have trusted it as a lifelong main weapon, including world-class grandmasters such as Viktor Korchnoi and Wolfgang Uhlmann, who defended the Black side of these positions for decades at the highest level. Today it remains one of the most popular replies to 1. e4 at every rating, prized by players who want a solid structure with a built-in counterattacking plan.
NM Heisman provides a firm foundation for those who are interested in learning how to answer 1.e4 with the French Defense 1…e6. Defenses to all of the popular lines for White are addressed.
We are going to play the French Defense against pawn to e4, so let's play pawn to e6. The idea of the French Defense is to support pawn to d5, hitting White's e-pawn and challenging the central light squares. The main drawback of the French is that our light-squared bishop is immediately blocked on the h3-c8 diagonal by the pawn on e6, so that bishop becomes the opening's "problem child".
We not only play pawn to d5 against White's pawn to d4, but we can also play pawn to d5 against pretty much every other White second move. That's a good tip to remember! So even if White plays, say, knight to f3, play pawn to d5.
White's third move, pawn to e5, is called the "Advance Variation" because he advances the e-pawn to close the center and gain space. This move is very common at the amateur level, so it's good to know how to play against it. Now we should answer with the thematic pawn break move (also called just a break move), pawn to c5. Without that move, our pieces would lack space; for example, if we instead blocked the break move with knight to c6(?), it would very likely result in a cramped position.
We now want to pressure the fixed White pawn on d4. There's no hurry to capture it, as after we take on d4 and White takes back with the pawn, the great square c3 would open up for the White knight on b1. Better is to develop a piece and pressure d4 with knight to c6.
Now queen to b6 continues to add pressure to d4. We could also continue with bishop to d7, but queen to b6 is thematic.
We have several moves here, but knight to h6 is good, or even 6...cxd4 7.cxc4 and then similarly 7...Nh6. The plan is to later go to f5 to continue to harass the d4 pawn. Don't be afraid to let White double your pawns with bishop takes on h6 - we get compensation with the "advantage of the bishop pair" as well as the semi-open g-file.
Now 7...Qxb2? is a trap as after 8.Be3! Qxa1 9.Qc2 our queen is trapped in the corner. So the simple recapture with the pawn on h6 is best.
Get ready to castle with bishop to g7. Yes, castling kingside is not as dangerous as it looks, since the dark-squared bishop guards the dark squares and we'll later be able to break with the pawn to f6 after we castle to further activate the dark-squared bishop and the rook.
White is better, but an unbalanced game is in store, so we have practical chances.
Now is the right time to capture the d-pawn, now that the other knight can go to f5. Knight to f5 is hasty, when White can play pawn to g4! and we have to retreat, as knight takes on d4 is not safe.
White is in trouble! After the natural knight to f5, there's no good way to guard the d-pawn, as bishop to e3 allows our queen to take on b2.
Castling was a mistake for White, who now must lose his critical d-pawn, as the only additional defense of d4, bishop to e3, allows our queen to take on b2.
We can try to hold our pawn with pawn takes on c3, as White cannot save his bishop due to the threat of pawn takes on b2.
Now we must recapture. White has a normal, slight advantage.
White has compensation for the pawn. The engine rates White slightly better after knight to a4.
Now we cannot win the d-pawn with 6...cxd4 7.cxd4 Nxd4?? allowing the famous trap 8.Nxd4 Qxd4? 9.Bb5+ Bd7 10.Bxd7+ (better than 10.Qxd4) 10...Kxd7 11.Qxd4 winning our queen. So just stop all the checks first with the solid bishop to d7, when we would be threatening to win the pawn on the next move.
This allows us to win the d-pawn starting with pawn takes on d4. This is a known gambit called the Milner-Barry Gambit, but we're fine winning the pawn.
Now after knight takes on d4, we have a line where White has sacrificed the d-pawn: the Milner-Barry Gambit.
We must recapture. Importantly, White does not have bishop to b5 check winning the queen, which is the reason we played our bishop to d7.
Now we could take the e-pawn with queen takes on e5, but this is very complicated. Instead the best move, engine-approved, is pawn to a6! This stops later tricks with their knight jumping from c3 to b5.
This is the main line of the Milner-Barry Gambit and the engine likes us here, so this move order has made the gambit fairly rare.
Saving the pawn this way is not good for White, but many amateurs try it because they don't know the Milner-Barry Gambit and don't want to lose the d-pawn. We can take advantage in the following way, by starting with pawn takes on d4.
Attack the bishop with knight to b4
If White tries to save the bishop this way, we can dominate the d3 square with queen to a6!
Our control of the light squares, specifically d3, gives us a big advantage. This line shows why White is pretty much committed to playing the Milner-Barry Gambit and giving up the d-pawn after playing bishop to d3 on move 6. I often see this line continue with the bishop going to c2 on move 7 in amateur games, but if we play precisely, we get the advantage.
Thanks for the pawn: knight takes on e5.
Knight takes on d4 is just bad. I can see from the database that even amateurs avoid that bad move.
Normally we don't want to take the pressure off the center by pushing to c4, which is anti-positional in most French lines. But here the weakening of the b3 square makes it playable, so we push to c4, leading to a closed game and lots of maneuvering to come.
This is one of the main lines of the Advance French. The very locked position is unusual, and we can often find shelter for our king on the queenside in this line.
Now get out of the pin with bishop to d7.
This mistake transposes to the same "rubber band" trick seen in another line. Take the guarded e-pawn with 6...Nxe5!
Thanks for the pawn.
We win a pawn. My student, radio personality Howard Stern, called this his favorite trap! :)
Win the pawn by capturing the bishop: bishop takes on b5.
Again, we've won the critical central pawn and have a large advantage.
If the center is locked with White's pawns on e5 and d4, and we have pawns on e6, d5, and now c5, and White does not keep his chain intact with pawn to c3 (here playing knight to f3), it is usually correct to trade our c-pawn on the wing for White's central d-pawn by taking on d4, since White cannot replace his d-pawn with another pawn.
We've got some advantage by trading a side pawn for a central pawn.
With only 1 White pawn in the center, we can now develop normally starting with the natural knight to c6.
Guard the knight while developing the bishop and breaking the pin.
Both knight to c6 and bishop to d7 are good.
Now develop the bishop to d7, get out of the pin, and as a bonus, set a trap that amateurs commonly fall into!
Allowing a well-known "rubber band" trap, but amateurs commonly play this! We can now win a pawn by taking a guarded pawn with our knight: knight takes on e5!
Of course, we must recapture with the hanging knight. Recapturing with the queen would lose a piece, as their knight takes on e5. So of course the right move is knight takes on d7, and we've won a pawn.
We're just up an important central pawn. The engine says with perfect play we're already theoretically winning.
Most lower-rated players would automatically recapture with the bishop here to "avoid doubled pawns and not block the d7 bishop". But notice the pawns can be undoubled by force as the c5 pawn is attacking the d4 pawn, and then later we can make a second break: we take on d4, White takes back, and we play pawn to c5. So capturing with the pawn here is actually slightly better according to the engine evaluation.
We have more central pawns and a good position.
We can play bishop to d7, hoping White makes the mistake of pawn to c3, when we take on e5 with the knight: the rubber band trick seen in other lines. But it's simpler to just trade a side pawn for a central pawn by taking on d4.
Guard the knight while developing the queen's bishop.
We're already better with more central pawns and the threat of knight takes on e5.
White is playing the Exchange Variation. Of course we shouldn't recapture with the queen as White can eventually start attacking the queen with tempo. Clearly better is taking on d5 with the pawn.
There are plenty of ways both White and Black can develop their pieces in this line. I am going to recommend a line where we can get some excitement by castling queenside. But we can also just bring out our kingside pieces and quietly castle kingside. In any case, don't leave your king too long on the open e-file!
We need to develop our king's knight. Normally we play knight to f6 here, but in the recommended line we are going to try to castle queenside, so we need to not only block any potential checks on the e-file while we develop the queenside pieces, but also deal with potential pins from their bishop on g5 by blocking with pawn to f6.
Knights before bishops. Move the queen's knight to its most active square, knight to c6.
Next to develop is the queen's bishop. With the White bishop on d3 and White castled, the pin with bishop to g4 gains in strength.
We're getting ready to play queen to d7 and castle queenside, with a potentially exciting game in store.
Attacking the White d-pawn, now that it is no longer guarded, is a natural developing move: knight to c6.
We have plenty of reasonable moves like knight to f6 and bishop to d6. But in this line we are suggesting getting ready to castle queenside (and threatening White's d-pawn) with bishop to g4.
Now taking on d4 with our knight is not as accurate, because of queen to e1, check: after knight to e6 and their knight to e5, White has play for the pawn. But after the suggested bishop takes on f3, removing the guard, we can safely take the pawn next move.
Having removed the guard by taking on f3 with our bishop, win the pawn by taking on d4 with the knight. Don't be afraid of queen to e3 check, because that double attack can be safely met with knight to e6, and we remain ahead a solid central pawn.
The engine says White has some compensation for the pawn after 8.Re1+ Be7 9.Qe3 Ne6, but we're solid and doing OK.
We'd like to develop the knight to e7, but that would block the bishop, so first we play bishop to d6.
Now block the e-file with knight to e7, preparing queen to d7 and castling long.
We now need to get ready to castle queenside. Note that with the knight on e7, castling kingside is a disaster after 8...O-O 9.Bxh7+! Kxh7 10.Ng5+ followed by 11.Qxg4, a famous tactical pattern in which White is winning. So we play queen to d7.
This setup was all about castling queenside, so we castle long. It's also not too late to castle short, as the trick 10. Bxh7+ Kxh7 11. Ng5+ Kg8 doesn't work here because the bishop on g4 is guarded by the queen on d7.
Normally the Exchange Variation is boringly equal when both sides castle kingside, but this line can be exciting!
This is a common amateur move. We can just pin the knight and get ready to castle kingside with the bishop to b4.
This is a solid line where we'll quickly castle kingside and get "normal" Exchange Variation play.
We can now develop with a check to speed up castling.
Now the knight can go to f6, but we recommend knight to e7 to prevent pins with a bishop on g5 and stop any immediate checks.
"Castle early and often" :)
We're fully equal.
We now have three main lines: bishop to b4 (the Winawer), pawn takes on e4 (the Rubinstein), and, the one we recommend, knight to f6 (the Classical). By attacking the e4 pawn again, we force White to save it with bishop to g5 or pawn to e5.
Pawn to e5 is often called the Steinitz Variation, after the first official World Champion. In positions like this, if we play knight to e4, White can take it with their knight, and after we recapture with the pawn, our pawn on e4 would be very vulnerable. So it's much better to solidly retreat the knight to d7.
OK, time for the — you guessed it — thematic break move, pawn to c5, which is almost always played when White closes the center by pushing the e-pawn to e5.
As usual, once our pawn has gone to c5 to pressure d4, we can logically follow up with the knight to c6 to develop while adding more pressure.
Simply play bishop to e7.
Castling kingside in the French when White has a solid pawn on e5 can sometimes be dangerous, but here it is OK.
An exciting game is in store as White may or may not castle queenside.
Time to castle and await further indications as to White's plan.
Time for the second break move, pawn to f6.
We're considered to be fully equal, if not slightly better. White has a little more space, but we have a "good French" type position with our usual problem child, the bad light-squared bishop.
White's move, knight to f3, is common at the amateur level but not very accurate, as after we break with pawn to c5, we will first undermine the d4 pawn and consequently the e5 pawn as well.
White could not support the center with pawn to c3, so it must hold onto the e5 and d4 squares. Now take the pawn before White can repair with pawn to c3: pawn takes on d4.
Develop and attack e5 with knight to c6, and we have a good game.
Our pressure on the White e-pawn and extra center pawn gives us full equality.
The main move for us here is bishop to e7, but I am recommending a fun line, the McCutcheon, with bishop to b4.
Now the only way to save the knight from the pin is the counterattack pawn to h6.
We cannot now play 6...Ne4 since 7.Nxe4 wins a piece ("rubber band trick"), so it's best to first capture on c3 to clear the e4 square for the knight.
Now e4 has been cleared for the knight.
White is now threatening to take on e4 with the bishop and, after our pawn takes back, to play queen to g4, double attacking e4 and g7, so we need to capture on d2 first, which also gets rid of White's advantage of the bishop pair.
As always, when the center has White pawns on e5 and d4 and we have pawns on e6 and d5, we should almost always play the break move, pawn to c5, as soon as possible, like now!
The engine thinks the chances are about even after a move like knight to f3 or pawn to h4.
We have to save the g-pawn. It has long been known that king to f8 is a reasonable move here, but engines have confirmed that theoretically pawn to g6 is a little better.
White has some advantage, but there is unbalanced play with chances for both sides.
Now move the knight to e4.
White has his normal slight opening advantage if he plays bishop to d2.
This bad move by White loses a pawn (a "Counting" tactic!). Now save your bishop by capturing the d-pawn — this is even better than bishop takes on d2, check, queen takes back, and our queen takes on f6, which is also good for us.
Capture the pawn. We're ahead a pawn with control of the center. Big advantage to us.
If White takes on f6 with the pawn, we reach a position that is theoretically already winning after our bishop takes on d4 and then on g7.
Gotta take back a piece with pawn takes on g5. No reason to play bishop takes on c3 with check first, as the knight is pinned.
Gotta save the rook and prevent promotion with rook to g8.
We can regain our pawn with queen to f6.
We'll win the g7 pawn, and the engine evaluates this as a nice advantage for us.
In this line the only way to save the piece is to break the pin with pawn to g5.
Gotta take back a piece!
We're doing well after targeting the f6 pawn with our knight to d7.
The f6 pawn can be won, and the engine evaluates this as very good for us.
Now just save the knight by moving it to e4, with good play.
After knight to e2, the position is about even. We'll soon get in our thematic break with pawn to c5.
This is the Janowski Variation, which has become more popular in recent years. We must save the knight, and here knight to e4 is safe and best.
In the MacCutcheon, White often plays this move and we have to decide whether to save the pawn with the king to f8 or with the pawn to g6. Both have their advantages and disadvantages. We recommend the pawn move to g6 in this variation, where White may have a small advantage.
The MacCutcheon usually results in a fun, unbalanced middlegame.
We now have bishop takes on c3 with check, or the recommended queen takes on d5, which often transpose.
Take the knight first.
Now regain the piece.
The engine says we should be fully equal. Both sides have doubled pawns and our centralized queen is safe.
White is playing the Tarrasch Variation 4. Nd2, popular at the master level but less so among amateurs. Now there are two main lines, knight to f6 and pawn to c5, and several other acceptable third moves for us. I recommend knight to f6, attacking the White e-pawn again.
Now 4...Ne4 is not good because 5.Nxe4 dxe4 and the Black doubled e-pawn is weak on e4. So solidly retreat the knight with 4...Nfd7.
With the center closed (White's pawn has moved to e5), we have the standard pawn break, pawn to c5. Without that move, we would get a very cramped game!
Keep adding pressure to d4 with knight to c6, a common theme in many of these lines for us when White closes the center with pawn to e5.
In this line, take the pawn on d4 to set up the next break with the pawn to f6.
We now have a second break, pawn to f6.
Capturing with the knight is best.
Develop the bishop to the best square, d6.
Castle to bring the king to safety and get the rook onto the semi-open f-file.
White has a normal opening advantage, but the imbalances in the pawn structure give chances for both sides! Course author NM Dan Heisman has played both sides of this position!
Play queen to b6 to keep pressuring d4 — a recurring theme in lines where White pushes to e5 against the French.
Play the thematic break move in these closed centers: pawn to c5.
Keep the pressure on d4 with knight to c6.
The engine thinks now is a good time to play the quiet bishop to e7, where it thinks White has a normal opening advantage... with a fun game to play.
White has a normal opening advantage. His best move is bishop to e2.
Knight to f3 was not White's most accurate move. You guessed it — play the "French break" with pawn to c5. Easy, right?
Keep pressuring d4 with knight to c6.
Play queen to b6 to keep pressuring d4 — a recurring theme in lines where White pushes to e5 against the French.
It looks like we can just win the d4 pawn since it is attacked more than it is guarded, but always be careful when doing that... In any case, start by taking on d4 with the pawn.
Here we can just play pawn to a5, the engine's top move. But most players can't resist winning the pawn, which isn't bad, so let's see: knight takes on d4.
Gotta recapture!
Save the queen. This is similar to the Milner-Barry Gambit, but White has better play here and more than full compensation for the pawn. Still, a pawn is a pawn...
White has full compensation - and maybe a little more - for his pawn. We'll castle kingside.
Believe it or not, the engine's top move here is pawn to e5, but let's play more simply with knight takes on e4, which another engine likes better! :)
Gotta recapture!
Develop the knight to d7, where it can eventually go to f6 to threaten the bishop. It also supports the important c5 break, which is also playable right away.
White has a normal opening advantage, meaning of course that we're "OK" :)
Lure the bishop to where you can attack it with a knight: pawn takes on e4.
Attack the bishop with the knight.
Now break up White's center with pawn to c5
We have full equality.
Bishop to e7 here is good, but slightly better is immediately attacking the center with pawn to c5, and we're OK!
In this line the engine already thinks we're slightly better.
Pawn to f3 is another very rare move that beginners or gambiteers occasionally play to guard the e-pawn. Take the e-pawn to leave the recapturing pawn unguarded.
Now we can accept the gambit.
Develop the knight and transpose into the Euwe Defense to the Blackmar-Diemer Gambit. The engine thinks White does not have enough for his gambitted pawn.
White is ahead in development and has more open lines. We have the extra pawn and no big weaknesses. The engine says that White does not have enough compensation for the pawn, but an exciting game could be in store.
Taking back on e4 with the pawn was a blunder. Double attack with the queen check!
Win the pawn and we're clearly way ahead.
We're up a pawn and the White king is in danger. Theoretically we're already winning, assuming best play.
Pawn to g3 is a further mistake, loosening the rook on a1. Double attack again by taking on e4 with the queen, giving check.
Take the rook...
Now play the pawn to b6 so the queen's bishop can help the queen escape from h1. With some care, we're winning once we extract our queen from the corner.
We should be able to extricate the queen, after which the extra rook gives us a winning position.
This is the rare Alapin Gambit. When in doubt, take that pawn! Pawn takes on e4.
Save the pawn with the natural knight to f6.
Now the engine prefers the counterattack with knight to d5. Now there aren't too many ways for White to save the hanging bishop on e3.
We can trade the pawn with 6...exf3 and be a pawn up, but White would be ahead in development after 7.Ngxf3. Slightly more accurate is to counterattack the d-pawn with knight to c6.
Now save the e-pawn by capturing on f3.
Develop the bishop.
White does not have enough for the sacrificed pawn.
Clear the d-file by capturing on e3 with the knight.
Much better to win the pawn with the undeveloped queen than to make the mistake of taking with the knight, when White can pin the knight by castling long or with rook to d1. We're ahead, so offer the trade of queens by taking on d4 with the queen.
We're just winning with the extra pawn and good play.
Same as in the line above. Take the bishop and then the d-pawn, winning material.
Taking the d-pawn with the queen is far better than 8...Nxd4.
We're ahead a pawn and have the better position.
Oops! Easy pawn fork wins: pawn to e3. Don't want to miss those easy ones.
We win a piece and the game!
If White doesn't play the main line with pawn to d4, just play pawn to d5 anyway! The attack on the pawn on e4 usually forces White to take on d5, and after we take back with the pawn and White plays pawn to d4, we transpose into the 2. d4 d5 3. exd5 exd5 lines already considered.
Transpose into the Exchange Variation with pawn takes on d5.
In this Nf3 line in the Exchange Variation, we can post our pieces aggressively. Normally it's knights before bishops, but here the bishop is best placed on d6, and we wait to see what White does before deciding whether to put our king's knight on f6 or e7. Other moves like pawn to c6 or knight to c6 are also good. It's hard to really misplay this move.
Now knight to e7 would likely transpose into the lines we considered under 2.d4 d5 3.exd5 exd5. Of course other moves like knight to f6 or pawn to c6 are acceptable.
Now we can castle kingside, but if so, we would probably have wanted our king's knight on f6 instead of e7. Now we will develop our queenside pieces with the aim of castling queenside and getting an exciting "castle on opposite sides" game.
Keep the pin and don't give up the bishop pair by taking on f3.
Time to get out the other knight. This follows "Move every piece once before you move any piece twice unless there is a safety issue (tactic)."
Prepare to castle queenside with queen to d7.
This exciting line suggested by IM John Watson in his book "Play the French" is somewhat good for White - his attacking chances on the queenside are a little faster than ours on the kingside. But it's very early and it should be an exciting game where either side can win. Play aggressively!
We can save the d-pawn with pawn to c6, knight to e7, or knight to f6. I suggest the last of these, which is also the engine's top choice.
Get the king off the open e-file!
Pin the knight, with a slight advantage for us.
Another solid Exchange Variation line.
Guard the pawn with the developing knight to f6.
Although the d-pawn is attacked twice and guarded once, we can ignore it for now and castle.
We will now play knight to d7 with the possible idea of a later knight to b6, and White's doubled, isolated d-pawns will be vulnerable.
The engine rates the position as almost dead even if White finds the best move, bishop to g5.
Save the bishop with bishop to e7. White's pawn advance to c5 was reasonable but not dangerous, as it takes the pressure off our center.
White has a very tiny advantage. We can soon break up White's pawns with pawn to b6 and have play.
Gain space and get ready to capture on d4 with pawn to c5. We're already doing well.
Capture the pawn before White can reinforce with pawn to c3. In general we don't want White to answer our capture on d4 with a recapture by the c-pawn. This transposes to the line 1.e4 e6 2.d4 d5 3.e5 c5 4.Nf3.
We've got some advantage by trading a side pawn for a central pawn.
With only 1 White pawn in the center, we can now develop normally starting with the natural knight to c6.
Obviously we have to recapture, but it is worth noting that "More pawns in the center is good!" - especially in the opening and middlegame. By capturing toward the center we have converted our b-pawn into a more powerful c-pawn that can later advance to c5 and control d4.
Engine evaluation is full equality for us. I actually find it easier to play Black in this type of pawn structure.
White wants to transpose into the Advance Variation if we play knight to c6 and they follow with pawn to d4, which is obviously OK for us. If we wish to thwart this, we can now play pawn to d4.
Recapture
Block and guard the d-pawn.
Now bishop to d7 manages to hold the d-pawn.
This attempt to remove the guard doesn't work.
Now 9...Qxd4 10.Nxd4 Bxg2 wins back the pawwn but the engine greatly prefers developing the knight.
Despite White's extra pawn, we have the bishop pair and all the play. The engine says we're much better.
Now knight to f6 is perfectly playable, but we can play differently to take advantage of White's omission of d4 by gaining space with pawn to d4 ourselves.
After almost any move White plays, we should play pawn to d5.
We should now transpose to the 3.Nc3 Nf6 line with knight to f6.
This transposes into the main line 1. e4 e6 2. d4 d5 3. Nc3 Nf6.
Now knight to f6 is perfectly playable, but we can play differently to take advantage of White's omission of d4 by gaining space with pawn to d4 ourselves.
Our d-pawn is attacked twice, so it needs to be saved. It's much better to do it with the c-pawn than to block that pawn with knight to c6. So I recommend pawn to c5.
We have a pleasant game with approximate equality.
We must recapture to regain material equality.
White has transposed into another Exchange Variation, which could be reached by 1.e4 e6 2.d4 d5 3.exd5 exd5 4.Nc3. We should just play normally with a knight to f6 and go for quick kingside castling. Because White has blocked his c-pawn with the knight, the engine already thinks we're slightly better.
A solid Exchange Variation line for us where we should aim to castle kingside quickly.
We can take advantage of this strange move some amateurs play by striking at the center with a pawn to d6. Of course the normal pawn to d5 is also good, but that allows White to transpose into the Advance Variation with pawn to d4.
Now we're ahead in development with the bishop capture.
We're ahead in development and already slightly better.
We can now force an advantageous trade of queens to stop White from castling.
Trading queens is good here!
Our light squared bishop is blocked by the e6 pawn, so we prepare to put it on the clear long diagonal by fianchettoing after b6, and we stand well.
White can't castle and we have no exploitable weaknesses. We're already better.
White has just put his bishop on a square where our natural pawn to d5 is not only safe, but attacks the bishop. I call attacking a piece with one of lesser value "AWL" (attack with lesser), and in the opening, when that move also develops and forces the opponent to waste time, we call that "winning a tempo". So here we of course win a tempo with pawn to d5.
We have to recapture, but we want that bishop to move again!
We should now develop naturally and get ready to castle kingside as fast as we can. Usually knights before bishops, so we play our knight to f6.
White's loss of time moving the bishop from c4 to b3 is why the engine now thinks we're a little better.
White's second move, pawn to d3, is an indication he wants to play the King's Indian Attack against the French. While much rarer than pawn to d4, it is not unknown; even Bobby Fischer was known to play this. We of course should answer with our normal pawn to d5.
We should expand in the center and control d4 with pawn to c5.
Either knight to f6 or knight to c6 is good here. I recommend the latter.
We should now get ready to castle kingside.
We should now avoid moving the bishop to d6, as this sets up possible later forks for White with a pawn push to e5. So playing the bishop to e7 is solid.
Castle early and often
We've set up what GM Soltis calls "The Standard Defense". The engine rates the chances as even.
White's natural move is a big mistake. It is too soon to advance, and the e-pawn will be in danger after we play knight to d7.
How can we safely get a third attacker on e5? Play the queen to c7.
White will lose his e-pawn and his center will collapse. We're already pretty much winning, assuming best play from now on.
We should recapture with the pawn. Not 3...Qxd5 allowing the AWL 4.Nc3.
Get out the knight and get ready to castle kingside as quickly as possible. White has wasted time moving the d-pawn to d3 and then to d4, so we have transposed into the French Exchange Variation except that now we have the extra tempo and are playing what White would play (reversed)! So we're already slightly better.
White has wasted time and we're now playing the normal "White" side of an Exchange French. Tiny advantage to us.
Queen to e2 is one of the few White second moves against the French where we should not play pawn to d5, because after White takes on d5 we recapture with the queen (taking with the pawn is now illegal because our e-pawn is pinned to the king), then knight to c3 follows, and this Attack With something Less (AWL) is awkward for us. So it's better to transpose to a good position in the Sicilian(!) with a pawn to c5.
Fight for the d4 square with knight to c6. This is a very rare line. We already have a slight edge due to the awkward position of the White queen blocking White's light-squared bishop.
This has transposed to a Sicilian-looking position where White's queen is awkwardly placed. The engine already rates us as slightly better.
Another rare second move for White. As almost always, we should play the "automatic" pawn to d5.
We want to be able to capture a pawn moving to d4 with a pawn, so now the thematic pawn to c5. This is a common idea in many rare French lines.
Keep fighting for d4 with knight to c6.
The engine already thinks we're a little better.
Some players play pawn to c4 to discourage us from our favorite pawn to d5, and indeed other moves are possible, but best is pawn to d5!
Gotta recapture!
White wants to lure our queen to d5 so he can "AWL" it with Nc3 and win a tempo. But we can delay the capture with ...Nf6.
If White had not checked, we could play knight takes on d5 with no problems, for example, after knight to c3, we take on d5. Now we delay winning back our pawn with knight to d7.
Put the question to the bishop to get out of the pin.
Break the pin with pawn to b5.
Hit the bishop with the knight to also discover the queen's attack on the d5 pawn.
The engine gives us a nice edge. White cannot hold the extra d5 pawn and we have active play.
Yes. The plans repeat across variations: build the e6/d5 chain, strike at d4 with ... c5, develop against White’s center. You learn ideas instead of memorizing long forced lines. Its solidity also means one inaccuracy rarely loses the game on the spot.
The defining moves are 1. e4 e6 2. d4 d5. From there White chooses between 3. Nc3 (the main line), 3. Nd2 (the Tarrasch), 3. e5 (the Advance), and 3. exd5 (the Exchange), and Black’s core plan of attacking d4 with ... c5 works against all of them.
The famous drawback is the light-squared bishop on c8, which starts the game blocked by its own e6-pawn. Good French players plan around it: trading it off, rerouting it via d7, or freeing it with pawn breaks. In return Black gets one of the most solid structures in chess.
They are siblings, not rivals: both meet 1. e4 with ... d5 behind a supporting pawn. The Caro-Kann (1... c6) frees the light-squared bishop earlier; the French (1... e6) gets the thematic ... c5 break in faster and tends to produce sharper counterplay. Better is a matter of taste.
You do not refute 3. e5, but you do pressure it: after 3... c5 4. c3 Qb6 5. Nf3 Nc6, Black piles attackers onto the d4-pawn at the base of White’s chain, often adding a knight via e7 to f5. White must defend d4 precisely every move or the center (and sometimes the b2-pawn) falls.
Reading about an opening is step one. The trainer at the top of this page drills all 64 lines against the moves real opponents play. The first lines are free.
Train the French Defense now