Recommendations for Playing the French

Nikos and I have already discussed our lines after 1.e4 e6 2.d4 d5 3.Nd2 for Playing the French, but as he has analysed everything with usual tenacity, it was quite interesting for us to see what the public thought. Especially as 3…Be7 is the current trend in repertoire books (and Emanuel is working on 3…Nf6) and we were heading more for 3…c5 in general.

We are still analysing, so we take everything on board, although we have some ideas. I am uncertain about 3…Be7, especially as when we looked at the recommendation in THE MODERN FRENCH it was straight to +- for White when we compared our recommendation for John’s 1.e4 books (which are a team effort as always, when John’s name is on the cover – but remember that John is the brutal dictator of Quality Chess, who has chosen to do the King’s Gambit alone!). All you had to do was one logical move (top 3 in all 1.e4-players’ candidates for sure) and tap the spacebar every 30 seconds. TMF seems a bit shaky at places, but is still a very interesting book that deservedly has received rage review.

82 thoughts on “Recommendations for Playing the French”

  1. Gilchrist is a Legend

    I voted for 3…c5 4. exd5 exd5. When my coach taught me the French when I was around 2000 I have played this for each time I played against the Tarrasch Variation. I have never lost four out of four times from opposition ranging from 2250-2400, but no one else whom I know plays this line. I like 3…Be7, but then if not, I suppose 3…c5 4. exd5 Qxd5 is what I would play.

    You said that lines have been discussed but that Nikos has analysed–so does that mean that the lines are almost finished or that the framework for the book is almost finished, or does the paragraph mean that there is consideration about other lines (3. Nd2 and/or others)? I am slightly confused with regards to that.

  2. While I wouldn’t advocate for the Rubinstein as the response to the Tarrasch, it certainly got a lot of support in the poll. I wonder if, given the didactic goals of the book to which Nikos has earlier alluded, you might consider an addendum on playing the Rubinstein. To date it has been very poorly served in the literature, as far as I can see.

  3. About the poll on what to play after 1 e4 e6 2 d4 d5 3 Nc3 Nf6 4 Bg5, I really don’t get the popularity of 4…Bb4. It seems to me that it leads to positions resembling the Winawer but if I wanted positions like that I would not play 3…Nf6 in the first place!

    So I strongly prefer the ‘classical’ 4…Be7 5 e5 Nfd7 where positions are similar to the Steinitz variation.

  4. Every recent book on French has covered MacCutcheon (Vitiugov, The Modern French and Watson, although he also covers the classical) but when was the last time you saw a comprehensive coverage of the classical French? I remember Gufeld’s book from 1999 (!) but nothing beside that …

  5. Kostas Oreopoulos

    Coverage of Nd2 in French is really hard

    a) c5 exd5 , is a chess school for every french player and every one should know and play it. You improve as a player playing both sides

    b) c5 Qd5, not my taste. For me French is a closed position with wing breaks

    c) Be7, h6… I would’nt base my whole rep on those moves.

    d) Nc6. This is really underestimated. Tzermiadianos covers it well (not fully from black’s point of view). I like Nc6

    e) Nf6. This is huge. A book by itself (and a big one)

    These days i am analysing Nc3 Nc6. I don’t know what you think, but its fully playable (at least as far as i have analyzed), with unique and interesting positions, holding nicely even in corr games, and basically very few know theory for it.

  6. @Hesam
    There is a big majority for 4…Bb4, I suspect. Not to cover it would be weird. But we need to do it better than the opposition, of course.

    But I am deeply tempted to cover 4…Be7. I am unsure if it equalises; probably not, but it is really interesting and non-standard to me.

  7. @Kostas Oreopoulos
    Emanuel will probably spend 2/3 of a book on 3…Nf6. It is natural for us to spend a lot of energy on 3…c5, but maybe look at something else as well.

    I am always happy as White after 3.Nd2 Nf6. I did not get a double pawn and I have time to get organised!

  8. Gilchrist is a Legend

    @Jacob Aagaard

    When I learnt the French I learnt 3. Nc3 Nf6 4. Bg5 Be7 as a solid choice where one can play positionally to outplay the opponent. It seems as if one slight mistake will not lose such as in Najdorf 6. Bg5 positions.

    Is it possible to include two options against 3. Nd2? 3…c5 4. exd5 Qxd5 if that is what the main plan was, but then I think 4. exd5 exd5 is interesting as well. However no one else whom I know wants to study this, nor play it, which I think is slightly odd. When I played 1. e4, 3…c5 4. exd5 exd5 was the line against which I hoped to never play, but never did since none of my opponents played it against me, instead opting for 3…Nf6 or 3…Be7.

    Also I am not sure if it is possible, but I play Winawer only with 7…0-0 and never with any of the lines with sacrificing the kingside pawns. I had wished at some point a book would be published based on the French as a repertoire book, with Winawer and 7…0-0 in 2012, but this has not materialised.

  9. Gilchrist is a Legend

    @Jacob Aagaard

    I found on ICC one of my games against a 2200 that was for some reason broadcast live against whom I played 3…c5 4. exd5 exd5 quite a few years ago. Here are some of the moves from the game:

    1. e4 e6 2. d4 d5 3. Nd2 c5 4. exd5 exd5 5. Ngf3 Nf6 6. Bb5+ Bd7 7. Bxd7+
    Nbxd7 8. O-O Be7 9. dxc5 Nxc5 10. Nb3 Nce4 11. c3 O-O 12. Be3 Re8 13. Qd3 a6
    14. Nfd4 Qd7 15. f3 Nd6 16. Bg5 Nc4 17. Rab1 g6 18. Rfe1 Nh5 19. Bh6 Bd6 20.
    Rxe8+ Rxe8 21. Nd2 Bf4

    I think during that game I was trying to play positionally, and I won in around 40 moves, but I think this line deserves consideration.

  10. Jacob Aagaard :
    There is a big majority for 4…Bb4, I suspect. Not to cover it would be weird. But we need to do it better than the opposition, of course.
    But I am deeply tempted to cover 4…Be7. I am unsure if it equalises; probably not, but it is really interesting and non-standard to me.

    Why are you unsure about 4…Be7, which line do you think poses more difficulties 6 h4 or 6 Bxe7? My take is that the consensus about these lines were formed long before engines became mainstream and due to its bad reputation people have avoided them. So I am quite confident that a thorough Houdini/Critter assisted analysis will reveal many lines as playable (or even better) for Black.

  11. Jacob Aagaard :
    @Gilchrist is a Legend
    No, I think this will not happen.
    But we will give two options against 3.Nd2.

    I am late to all this but personally I would prefer 3…c5 with IQP. However given that you have written a book on QGD Tarrasch, if you choose 3…c5 here as well, you can count on Herr Siegbert’s ghost helping with any analytical problems that might arise when you are working on the book.

  12. @Kostas Oreopoulos:

    I think that 3.Nd2 Nc6 is better for White than 3.Nc3 Nc6, when White still can get some edge. Have you seen the recent coverage of 3.Nc3 Nc6 in the latest CBM? (I think it is an article by GM Kritz).

  13. I thought you guys had already decided on c5 with exd5 (in case of the exchange). I applaud and think this a wonderful idea, and sorely lacking in the literature. I don’t understand the appeal to people’s opinions on this……Be7 sucks and Nf6 has too much theory and is over-rated, and the other moves are just plain bad or boring. C5 strikes me as the professional choice but also the most correct one. Nf6 would be alright but it’s been done to death, and as you mention, it’s being done already. So that leaves c5. The only choice , really, should be what piece to capture with on d5 in the event of e takes.

  14. I played the frenche with black for a while. I wanted to play the sharp and dynamic Winawer variations, but in practice nobody played this sharp mainlines and most of the time the I faced the Tarrasch. Against the Tarrasch I played 3…c5 and 4…Qd5 and altough it was ok for black, I found the positions a little bit dull and I always felt that white remains too much in his comfort zone.
    I hope you will find something challenging against the Tarrasch, if so I will consider to play the french again :-).

  15. Gilchrist is a Legend

    I played 3. Nd2 for the almost 10 years I played 1. e4 before I switched to 1. d4. In ranking of playing certain lines from most comfortable to least comfortable:

    3…Nf6 with 11…0-0 (most comfortable against which to play)
    3…Nf6 with 11…Qc7
    3…c5 4. exd5 Qxd5
    3…c5 4. exd5 exd5 (least comfortable against which to play)

    Given that I usually have played 3…c5 4. exd5 exd5 and hoped to never play against it when playing 3. Nd2 as White, my favourability for this line is obvious. 🙂

  16. Gilchrist is a Legend

    I am not sure why, but when I played 3. Nd2, 3…Nf6 arose positions in which I was very comfortable, specifically the main line with 11…0-0. I cannot remember losing against 3…Nf6 in at least 10 games, because for some reason many of my opponents played 3…Nf6. I played a game that was drawn after more than 130 moves in 3. Nd2 Nf6 against some IM because I stupidly declined a draw offer on move 20 and probably overestimated my position because I thought my position looked excellent. But still as a former 3. Nd2 player, 3…Nf6 is most comfortable against which to play, in my opinion and practise.

  17. Kostas Oreopoulos

    @Nikos Ntirlis

    No, i am waiting for it, CBM 49 i mean (along with some other books)

    I dont remember details well on Nd2 Nc6, but there was nothing very special for white.

    For me, getting a nice position is not enough. It has to be easy to play and maintain the edge.

    This is why i extensively use 2-D evaluations (from IDEA). You get a complete picture of the position (and branching factor , to understand how much one has to memorize )

  18. Hesam :

    Jacob Aagaard :
    There is a big majority for 4…Bb4, I suspect. Not to cover it would be weird. But we need to do it better than the opposition, of course.
    But I am deeply tempted to cover 4…Be7. I am unsure if it equalises; probably not, but it is really interesting and non-standard to me.

    So I am quite confident that a thorough Houdini/Critter assisted analysis will reveal many lines as playable (or even better) for Black.

    No, I don’t think. I guess the French is lightly better for White. This doesn’t mean White must absolutely win or White has a large advantage but objectively White must have a little plus, a little something, with the good moves. I don’t think Black has a real absolute equality.

  19. @Kostas Oreopoulos:

    I agree for the 2-D evaluations. It is a very usefull tool. Wait for the John’s 1.e4 books to see a nice way for White to play in the 3.Nd2 Nc6 variation!


    I showed Jacob at Istanbul a line with 3.Nd2 c5 4.exd5 exd5!? and he agreed with me that it is playable for Black. So, i think that we’ll have this line in the book. I think also that still 4…Qxd5 is best theoretically and that you’ll find our analysis there quite good.

  20. @garryk:

    You can refer also to the Yusupov series! (i cannot remember the book where he recommends the Advance, which by the way is a good line for White)

  21. Having been a French player for 10 years before quitting playing it outside of Transposing to the MacCutcheon thru the Veresov (i.e. 1.d4 d5 2.Nc3 Nf6 3.Bg5 e6, and if 4.e4, then 4…Bb4), I can tell you that there are certain lines that deserve no respect at all, and others that are severely underestimated:

    First off, after 3.Nd2 Nf6, one line Gilchrist never accounted for was 3…Nf6 with 7…Qb6 (or 8…Qb6 or 11…Qb6, all transposing to one-another, the first two allowing White to play 10.Nc3 instead of 10.exf6, but I like that line for Black). I think this is stronger than 11…O-O or 11…Qc7, the latter often leading to passive play for Black, and positional dominance by White via ownership of e5.

    However, I have absolutely no respect for the Tarrasch at all, and would never recommend it to anybody as White because of 3…c5. If 4.Ngf3, Black can steer back to the Closed Variations with 4…Nf6 as White has now committed to the Korchnoi Gambit instead of Ne2 and the other Knight to f3. Black gets a good game in the Korchnoi Gambit if he plays it right. If 4.exd5 instead, Black takes back with the pawn, and answers the check with the Bishop Interpose while answering 5.Ngf3 with 5…Nf6! instead of the old 5…Nc6?!. White has problems here!

    That leaves 3.Nc3 and 3.e5, both of which to me give White an advantage, and are the main reasons I’ve moved away from the French. Maybe a new book with updated analysis might get me to play it again some day, but I will never fear the insipid 3.Nd2. In my neck of the woods, everyone around here seems to like to play the Winawer against 3.Nc3, and more specifically the Poisoned Pawn variation, which I’ve typically destroyed, with those Kingside passers playing a major role in the attack, but more often than not, it’s parallyzing everything else, and only then either rolling the h-pawn or having a decisive g5 or g6 break.

    The Classical I respect more than the Winawer, but it lacks the winning chances that the Winawer has. If White knows what he’s doing, he can make Black’s game miserable, where he’s playing for 2 results, a draw with correct play, or a loss without it. I give the Classical the same respect that I give the Castling Long lines in the Classical Caro-Kann. Black’s playing to draw and that’s it. The MacCutcheon is riskier than the Classical, and you’ll probably have more losses with it than the classical, but it also gives Black better winning chances than the Classical. It’s like comparing the Najdorf to the Petroff.

  22. I agree with Patrick although if you really want a plus then you should probably choose 3. Nc3 rather than 3. e5

    By the way, I had read a Sveshnikov book on the advance French where he said himself against 3. e5 probably Black must be able to equalize but just with a very precise play and if Black avoids many pitfalls. He said probably for a real plus 3. Nc3 is the way to go…

  23. Gilchrist is a Legend

    @Nikos Ntirlis

    I am pleased that 3…c5 4. exd5 exd5 is being seriously considered for the second line in addition to 3…c5 4. exd5 Qxd5. I think the former is a good line, and the only line in the French that I was told to play against 3. Nd2.

    I think maybe someone had said it before on the blog, but since you wrote a Tarrasch (Defence, not 3. Nd2) book, IQP positions you know quite well. I think the line 3…c5 4. exd5 exd5 is another good IQP position for Black. In 3…c5 4. exd5 exd5 5. Ngf3 Nf6 6. Bb5+ Bd7 7. Bxd7 Nbxd7 8. 0-0 Be7 9. dxc5 Nxc5, which if I remember correctly, is one of the main lines, Black manages to exchange the light squared bishop with free piece play. I have two wins against a 2200 and a 2300 as Black in this line, and I think this is a comfortable position to play.

    The other line I remember from years ago, is the other temporary pawn sacrifice main line with 5. Bb5+ Bd7 6. Qe2+ Be7 7. dxc5, then White castles 0-0-0 and some piece exchanges leave Black with the IQP but Black also exchanges the light squared bishop and I think also one of the bishops to eliminate the advantage of the bishop pair. I never had any problems in this main line either.

  24. Nikos Ntirlis :
    I showed Jacob at Istanbul a line with 3.Nd2 c5 4.exd5 exd5!? and he agreed with me that it is playable for Black. So, i think that we’ll have this line in the book. I think also that still 4…Qxd5 is best theoretically and that you’ll find our analysis there quite good.


  25. @Gilchrist is a Legend
    As far as i know White must play 4 Ngf3 in order to tp present “problems” for Black.

    4. exd5 exd5 and then later Bb5+ Bd7 gives Black the chance to develop actively with Nf6 as you pointed out.
    After 4 Ngf3 Black can only go for Botvinniks setup with Nc6, Bd6 and Nge7, which is solid, but less active.

  26. Gilchrist is a Legend

    @Jacob Aagaard

    Perhaps, but in both main lines 5. Bb5+ and 5. Ngf3, the light squared bishops eventually are exchanged with IQP positions. I would much rather play these positions as Black.

    Do you mean 5. Ngf3, after 4. exd5 exd5? Because 4. Ngf3 then I think Black has a few options already. 5. Ngf3 then Nf6 as mentioned by some above instead of 5…Nc6. 5. Ngf3 Nf6 6. Bb5+ Bd7 7. Qe2+ Be7 8. dxc5 0-0 9. Bxd7 Nbxd7 10. 0-0 Nxc5 is a line instead of the setup with 5…Nc6 6. Bb5 Bd6 7. dxc5 Bxc5 8. 0-0 Nge7 9. Nb3.

  27. Gilchrist is a Legend

    4…exd5 and 4…Qxd5 will not be as difficult as two completely different variations, since they both begin 3…c5, and then I think most critical is obviously 4. exd5, but 4. Ngf3, 4. c3, etc. can be met simultaneously for those who both play 4…exd5 or 4…Qxd5.

    Also with regards to 4. Bg5 Be7, it would be interesting to cover this as well as 4…Bb4, but I am not sure if this is in plans or not.

  28. I was just looking at Avrukh’s Slav ideas; the question of a Slav book I know has been asked. I just can’t remember the answer.

  29. Gilchrist is a Legend

    Interesting I think is the line 3…Be7 4. e5 c5 5. Qg4 Kf8, where I have had problems playing against this as White, but if I had to choose between this and 3…c5 4. exd5 exd5, I rather would play the latter.

  30. Has anything yet been decided or revealed regarding which line the book will recommend versus the Steinitz, please?
    (After 7.Be3, 7…a6/7…cxd4/7…Be7) ?

  31. Gilchrist is a Legend

    @Nikos Ntirlis

    Do you find 4…Bb4 a better move than 4…Be7 after 3. Nc3 Nf6 4. Bg5? To me I think both are fine but each has its own character, with 4…Bb4 sharper and 4…Be7 quieter and more positional generally, except for some lines with White playing 0-0-0 and Black 0-0 and there are mutual kingside attacks. 4…dxe4 and 4…Be7 seem to be positional but sound to me.

  32. Hey Jacob, I was just wondering when the playing 1.e4 books are coming out and if you can give us a teaser as to what some of the recommendations are (besides the Tarrasch French, which made me salivate). Also is a KID book possibly in the cards? The one advertised on this blog earlier by Kotronias? Keep up the good work!

  33. I second the suggestion made here on 3. Nc3 Nf6 4. Bg5 Be7.

    Ziegler’s DVD is the only recent source. There are lots of new lines to discover here with engines, e.g. I was surprised to see in latest NIC yearbook that Alekhine-Chartard 6. h4 h6!? is playable and even some GMs have tried 6. h4 Nc6. After 6. Bxe7 I think Black’s chances for equalising is no worse than rubinstein variation (i.e. White’s advantage can be contained to a normal “+=”) but Black has more interesting typical french play than in the rubinstein.

    There are simply too many books on the MacCutcheon at the moment!

  34. What about the “French Volga”? After 3…Nf6 4.e5 Nfd7 5.Bd3 c5 6.c3 b5!? even engines believe that Black gets decent compensation!

  35. Gilchrist is a Legend

    Interesting to hear about 3. Nc3 NF6 4. Bg5 Be7. I think it would be excellent as well to include both 3. Nc3 Nf6 4. Bg5 Bb4 and 4…Be7 as well as 3. Nd2 c5 4. exd5 Qxd5 and 4…exd5. I think in the 4. e5 line, there are many seventh move choices that are playable for Black, so maybe one choice, where you said 7…a6 I think it was, will suffice, unless of course you plan 7…a6 as the main line and another move as alternative.

    @Nikos Ntirlis

  36. Hi Jacob,
    I asked myself where to put this question and found no better answer than this post; I’m wondering when the GM Rep on the Open Lopez will be published and if it only covers the Open Lopez after 5.0-0 or if it also gives recommendations against Whites other tries in move 5,namely the Worrall(5.Qe2) and the 5.d3-System..
    Thankys in advance 🙂

    P.S.: Keep up the good work! I already own..well,actually I lost count of the QC-books in my bookshelf. I’m especially fond of the Schandorff books on 1.d4 and the Yusupov books. Muuuuch better layout than the “Tigersprung” versions here in Germany ;D

  37. Kostas Oreopoulos


    In short, 2-D evaluation , tries to answer not only who is better, but how easy it is to play the position.

    It all starts with the comparison of the best N moves returned by the computer, and how frequently a player plays the 1st – 2nd -3rd….Nth choice

    One can say a lot about how a strong player picks a move, and when the move is “critical’ (meaning all other moves are clearly worse) how often does he pick the best move… etc
    You will be “surprised” to know that very good (ultra top gm’s) are not “THAT” much better in playing better moves . They usually play the 1st-2nd and 3rd choice at the same % as a much worse player. What makes a difference is that the play bad moves much more rarely than players with less rating.

    Now with the 2D evaluation you can see (if you limit for example to the first 5 choices of the computer, which means if you play BY YOUR SELF, without knowing everything) what is the percent that the position will end up beeing +=, what = , what =+ and so on.

    So 2 positions might be = with best play, but over the board, the one might have 5% of = and 95 of +/- and the other 100% of =

    What that means is, that move (a) leads to equality, but over the board it would be very hard to find the way to equality. On the other hand , move (b) is dead equal, whatever you play.

    So depending on the situation you enter this or that variation.

    if you want to read more, you can read this

  38. @Maxi
    We are still some months away from finishing this book. Victor is working steadily towards the completion and things are in general going well.

    The book starts with 4.Ba4.

  39. I have a question for the QC Team. The Playing 1. d4 and Playing 1. e4 series are great works but what about for Black? I mean, why not the same work for Black? E.g. 1st vol. Playing against 1. e4 and 1. d4 then 2nd vol. Playing againt 1. d4 deviations (all systems with 1. d4 without c4), 1. c4, the case of 1. Nf3 and irregular openings (1. f4, 1. Nc3, 1. g4 …)

    Maybe someone like Schandorff could make this sort of work… With e.g. the Caro against 1. e4, the Nimzo and Queen’s Indian against 1. d4 etc

    Thank you by advance for your answer.

  40. It is now October, should we look forward to 2 new books this month (Avrukh and the second series of your GM hardbacks?)

  41. Gilchrist is a Legend

    Does that mean Positional Play and GM11 and to be released concurrently at the end of October or similar?

  42. From all your comments it looks like the classical slav, modern benoni and open spanish books will not be out until at March/April at the earliest next year, kings indian might not be until 2014 and that the French book by Nikos might be out first quarter next year. Am I right?

  43. Gilchrist is a Legend

    I think the focus should be for October–I am contemplating pre-ordering Avrukh’s GM11 quite soon, probably this week. Any new updates on it?

  44. If I wanted to order a third book to get the free postage, is the next book you think today (obviously realise things can change) you will publish after these two available on your “coming soon” section so that I can order it?

  45. Gilchrist is a Legend

    @Jacob Aagaard

    So I suppose that means the pattern is GM11 and Positional Play websales scheduled to be shipped from Glasgow on Monday 30/10. I suppose I will pre-order tomorrow or Thursday, but how do you know how many copies are posted to the QC office in Glasgow to be shipped? For example, if there are 300 copies to be posted to the office, what happens if 400 people pre-order with the last 50 pre-ordering couple days before posting?

    Regardless of postage time, it should not take that long hopefully to get myself a hardback copy of GM11 down here in Manchester, maybe on Hallowe’en 😀

  46. Gilchrist is a Legend

    I know Avrukh wrote the 616(?) page book on GM2, what is the page count for this GM11 tome now? 😀

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top