Travail Pursuit #61: Into the deep

As the sun sets on our adventure into the French Defence I invite you to take one final plunge with me into the murkiest depths of opening theory. This image is used under Creative Commons terms and sourced from the NOAA Flickr photostream

As the sun sets on our adventure into the French Defence I invite you to take one final plunge with me into the murkiest depths of opening theory. This image is used under Creative Commons terms and sourced from the NOAA Flickr photostream

Here it is then, the final part of my mini-series on this very particular line of the French Defence, Tarrasch Variation.Time to ditch your snorkel and flippers, pop the hatch on your submersible, climb inside and join me as we dive down to explore chess theory’s equivalent of the Marianas Trench. Today we’ll turn on the sub’s spotlights and take a look at some of the strange and fascinating creatures that can be found down here as we examine variations occurring after Black’s 18th move! Just remember to decompress before you return to the surface.

Last time out we looked at some perfectly reasonable Black alternatives to 18…Qa5+. This time we’ll examine what can happen after this ‘mainline’ (can you call it a mainline when it’s the 18th move!!) is played.

The key point behind the check of course is that Black displaces the White king and makes it just a little bit more time consuming to get him to safety. Unfortunately I must start out on this post by recollecting a couple of chastening experiences:

Black has played 18...Qa5+

Black has played 18…Qa5+

White responds to 18…Qa5+ with 19.Bd2?
Shapland vs. Firth, Calderdale League 1, 2008 (see game 1 in the viewer at the end of this post).

This was the very first occasion upon which I ventured to play 9.Nf4!? having looked at the theory and prepared for this opponent. Obviously I didn’t prepare very well because when we quickly reached move 19, I couldn’t remember the best way to play, got very concerned about moving my king and decided instead to play 19.Bd2.

The consequences should have been disastrous. My opponent responded with the natural-looking 19…Qb5 (preventing king’s-side castling) whereupon I compounded my error by castling long into the teeth of a ferocious assault. Black won a tempo with 20…Rc8+ and after 21.Kb1 Ba3+ Black is already completely winning. I have no idea how my opponent managed to fluff the attack to the extent where he felt he should take a perpetual check. I breathed a massive sigh of relief and went home to learn the opening line all over again!

Having almost learned in the hardest way possible that 19.Kf1 is forced I went on to have some success with this variation although I’ve inevitably lost some games too.

White has played 20.Rc1?

White has played 20.Rc1?

White doesn’t exchange on f6 when the pin is broken
Inter_Mezzo vs. VampireChicken,, 2014 (game 2 in the viewer)

Just to demonstrate that I’m still getting it badly wrong from time to time, here is a more recent game which amply demonstrates the serious dangers that lurk in wait if the Black knight on f6 is allowed to activate itself. I’ve mentioned in previous posts that, once the knight has been unpinned, either by Be7 or the Black queen moving White is well advised to play Bxf6.

This game saw 18…Qa5+ 19.Kf1 g6. The point of this last move is to allow the knight to come to f4 via h5. Now it really is compulsory to play 20.Bxf6 but instead I decided to try 20.Rc1. You can see what happened in the game, but suffice it to say, I did not escape with a draw on this occasion!

Black has played 19...Qb5+

Black has played 19…Qb5+

Black plays 19…Qb5+
Shapland vs Patrick, Calderdale League 1, 2011 (game 3 in the viewer)

Let’s get all the painful stuff out of the way first. Here’s another harrowing loss for me. This time it occurred over-the-board. In this game I also didn’t exchange on f6 and once again the knight played a crucial role in my downfall. On this occasion though, the origin of my sins was not the failure to exchange on f6 because I’d already blundered before that became critical.

In this game, my opponent decided not to continue with 19…g6 but instead played 19…Qb5+. This move of course helps White to get his king safe but the Black queen initiates a threat on the White b-pawn in exchange. It’s an interesting idea. My opponent followed up with a more familiar manoeuvre 20…Bc5. We know what this is all about. Black blocks the c-file. I responded in principled fashion playing 21.Rac1 to tie the queen to the defence of the bishop but after 21…b6 I lost my head and played 22.a3?? This move simply allows Black to clean up on the queens-side and my opponent never let me back into the game again after that.

Instead of my blunder the correct way to continue would have been 22.Rc2 possibly followed by 23.Rd1 or Rhc1 and then Rd2. Either way, White has to defend the b-pawn here as he doesn’t get any compensation chasing the queen in this position. Conversely, remember all the way back to part one of this series where we saw that grabbing the ‘poisoned’ pawn can be dangerous for Black in the mainline of the closed system.

Black has played 19...Be7

Black has played 19…Be7

Black plays 19…Be7
Bellerophontis vs nosykes,, 2015 (game 4 in the viewer)

Here’s another ‘readers game’. My thanks to Nick Sykes for sending it on to me. This time Black tries another typical bishop move which also prepares the g6 advance. Again in this game we see Black playing the d4 advance fairly swiftly. This time he’s vacating the d5 square for his knight and not his queen.

On this occasion White responds by occupying the c and g files with his rooks. He’s able to tuck his king away on h1 but again, Black is able to land his knight on f4 and this piece once again plays a decisive role.

We also see another thematic idea in this game. With the bishops swapped off on g5 and Black’s rook still at home on a8 White advances his b-pawn to try and deflect the Black queen from c7. He then overlooks the most accurate continuation which was to put his rook on c5 and double on the c-file but he still gets some good play on the king’s side with his rook on the g-file and his queen on the h-file.

Finally, White manages to infiltrate his queen to the back rank but Black still has threats of his own and the game sharpens significantly. This well-played and very interesting game should probably have ended in a draw by perpetual check but White plays for a win and blunders to defeat.

Black has played 19...Rc8

Black has played 19…Rc8

Black plays 19…Rc8
Inter_Mezzo vs calexico, Red Hot Pawn, 2009 (game 5 in the viewer)

19…Rc8 looks sensible. Black takes his chance to occupy the c-file before White has had the chance to connect his rooks. However, there is a drawback to this plan. Because Black no longer has a pair of rooks, White can (and does in this game) simply put his king out of harm’s way on g2 and then contest the c-file. If Black exchanges rooks then White will hold the file and should be able to penetrate decisively.

In this game Black tried the interesting 21…Rc6!? with the idea of playing either 22…Qc7 or if White exchanges with 22.Rxc6 bxc6 23.Rc1 then c5 or Qb6 simply shuts down the file. That said, 22.Rxc6 was still White’s best option. Instead, I chose to employ another typical idea in the position which we’ve seen in a previous post, namely Qh3 and f4 with a view to re-deploying the queen to the queen’s side.

I thought I’d found a neat idea to land my rook on c7. It shouldn’t have been decisive with precise play but Black went astray by blocking the check along the seventh rank with his bishop instead of moving his king. This created a weakness back on the king’s side which I was able to exploit with 27.Qh3.

After some further adventures eventually I was able to get a decisive attack on the g and h files by advancing my h-pawn and giving back the exchange to open lines.

Black has played 19...g6

Black has played 19…g6

Black plays 19…g6
Inter_Mezzo vs. CaptainTod,, 2014 and
Inter_Mezzo vs. sramesh73,, 2013 (games 6 and 7 in the viewer)

Playing g6 directly on move 19 resolves the tension on the h4-d8 diagonal immediately. As we’ve already seen, White is obliged to exchange on f6 when Black uses a little intermezzo 20…Qa6+ to recapture on f6 with his queen. Of course, White should avoid the exchange of queens and switch over to the queen’s side with 21.Qa4.

In game 6 my opponent chose 23…Bc5 placing his bishop on a familiar square. We’ve seen several games where this move is played at various stages. I won’t cover the ideas again but, suffice to say, with Black’s queen already deployed on c6, White doesn’t need to be offered a second chance to play 24.Rac1 and pin the bishop. Subsequently, having managed to lure the Black king to f6, I was able to exploit this pin on the c-file to transfer my queen back to the h-file via e3. This gave me my first toe hold in the Black camp. Once my queen had reached h6 I pushed my h-pawn to put pressure on the Black king’s position and my opponent made an error.

Although I didn’t follow up consistently (I should have carried on pushing the h-pawn), and subsequently squandered my advantage, I did ultimately win the game when my opponent made a second blunder that lost on the spot.

In game 7 we really do journey into the deepest depths of the theory on this line and I’ve added into the annotation examples from some high level games to demonstrate that a number of others have passed this way before me. Black deviated from game 6 when he chose to play 23…Bh6 which I think in this instance is an improvement on the game above as it exploits the White queen’s absence from the h-file and prevents White from putting a rook on c1.

Despite this, White finds a way to get a rook on the c-file via 24.Rhd1, 25.Rd3 and 26.Rc3. I then selected to continue manoeuvring this busy rook with 27.Rc5 although I also think that 27.Qc2 is a good option and it’s been played before as you’ll see in the annotations to this game.

Finally, on his 29th move (!!) Black plays a move that appears to be new – 29…d3!? Of course this makes a lot of sense. We’ve seen on plenty of occasions already that the advance of this passed pawn is one of Black’s key ideas in this line. I tried to exploit this move directly by playing 30.Qc3, attacking the Black e-pawn and after 30…Bf4 31.Rb4!? Black makes the mistake of advancing his king to g5. The game ended in short and quite unusual tactical sequence where it seems Black could have survived by voluntarily stepping his king back into a pin on f6. Once he’d missed that brief window of opportunity it was curtains.

You can down load all the games in this database by clicking on the ‘Chessmicrobase’ logo on the top  right of the game viewer.

Time to have a quick refresh of some of the main themes that I think have emerged over the course of this series.

Lighting the fuse: in part 1 I looked at all the possible deviations play can take before the game reaches the position where White can play 9.Nf4!?

  • 1.) From these games it was clear that Black really does have to play in the hypermodern style in this line. With 3…Nf6 he gives White   the centre and he must then play actively to attack it otherwise he’ll just get strangled. Typically this will involve playing f6 and later e5 as Black aims to create a passed d-pawn.
  • 2.) If Black doesn’t attack the White centre then the first player can often take the initiative on the dark squares and build his play up slowly, keeping Black’s counter play under lock and key.
  • 3.) When Black plays lines with Qb6 in order to avoid 9.Nf4, then White can consider offering the b-pawn as a gambit in exchange for activity and a strong initiative. The half open b and c files can be more important than the extra pawn in such positions.

Disturbing the peace: in part 2 we examined some interesting deviations that can occur between 9.Nf4 and 14.Qh4 or 14.0-0. The theory books tend to suggest that the sequence following 9.Nf4 is forced, but I don’t think that is the case.

  • 1.) Both 9…fxe5 and 11…gxf6 are interesting options. They may not offer Black the best chances against a really well prepared opponent but they have good surprise value. Properly prepared players of the White pieces need to have an idea of what to do against these options
  • 2.) 14.0-0 looks like a safer option for White if he wants to avoid the most sharp and double-edged lines. There are still possibilities to put Black under pressure especially if White opens the a-file by exchanging knights on b3. The Black defensive manoeuvre Bc5-b6 looks like the sternest test of White’s efforts to graft out a win.
  • 3.) When White plays 14.Qh4 he invites some pretty wild complications that can’t guarantee a theoretical advantage with best play but do offer a wealth of practical chances. Here 16…Bb4+ looks like a very good alternative to the deepest and most well analysed lines. Should Black want to deviate from these paths then this is really his last best chance.

On the precipice: in part 3 we jumped into the detail of the lines that follow the sequence 14.Qh4 e5 15.Nf3 Nxf3 16.gxf3 Bf5 17.Bxf5 gxf5 18.Bg5. In particular we looked at the variety of approaches Black can take here if he doesn’t want to choose the main move 18…Qa5+.

  • 1.) Advancing the d-pawn immediately with 18…d4 is principled and initiates some dangerous ideas but can Black afford to neglect his development for another move? This might be a good line for Black if, having arrived at this point, got vertigo, and wants to bail out to a draw
  • 2.) In certain circumstance (for example after 18…Bc5) White can castle long, but in most cases he should go short. The best squares for his rooks looks like c1 and d1 when if Black does advance his d-pawn White can blockade on d3 with his rook.
  • 3.) Two ideas of Black’s that White must be very wary of are allowing the remaining Black rook to get to g8 or h8 from where it can help launch a counter attack on White’s king. Allowing the knight to reach f4 via e5 or h5 is not recommended. As soon as the knight is unpinned, exchange it.

Into the deep: in this final post we’ve looked at games where Black has played 18…Qa5+. Whilst there are some unique ways in which play can develop here, some of the ideas we’ve seen in previous posts are still valid.

  • 1.) Primarily, we saw again that the Black knight will need to be eliminated as mentioned above. Black should again aim to keep White at bay and advance his passed d-pawn whenever he can.
  • 2.) White should be aiming to get his heavy pieces into the Black position. Most commonly this can be done via the c-file or h-file but the g and b files can also be used.
  • 3.) Very often White uses the fourth rank to manoeuvre his queen from one side of the board to the other, stretching and probing the black defences. Concessions can be extracted on one flank before moving back to the other.

So there we have it folks. Over this series of four posts I’ve given you no fewer than 29 annotated games in the closed variation of the French Tarrasch. I do hope that, if you’ve survived the hardcore theoretical jousting, you’ve at least enjoyed some of the games that have emerged for me and others in this really interesting and complicated variation. I think I’ve been able to demonstrate that, like all other openings, knowing the themes and ideas is much more important that knowing the moves off by heart. Yes, there are some forcing lines here, but there are also some surprising sidelines that players of both colours can and do consider.

Travail Pursuit will return…


11 Responses to “Travail Pursuit #61: Into the deep”

  1. Matthew Webb

    Nov 19. 2015

    Wahooo! I’ve been looking forward to this article. I may have to clear up my schedule for this evening :-)

    **A more suitable comment coming soon!**

    Reply to this comment
  2. Eric Gardiner

    Nov 19. 2015

    I can’t resist being the first pedant to point out that it’s the Marianas Trench!

    I too will give the article the respect it deserves and come up with a less flippant comment in due course :)

    Reply to this comment
    • Dave

      Nov 21. 2015

      Thanks for correction Eric. I’m glad one of my readers has good enough trivia knowledge to put me right. I should have checked the spelling before I posted. I’ve corrected the article. Cheers

      Reply to this comment
      • Eric Gardiner

        Nov 21. 2015

        No problem, I found it quite funny :D I appreciate the time you put into writing these articles. I have Chessbase notes I’ve wrtten on various openings but they aren’t as entertaining to read!

        Reply to this comment
  3. Nick Sykes

    Nov 19. 2015

    In my game after 22…Bxe7 Nf4+ was my intention, but of course this leads to a perpetual check after 23.Kg3 Ne2+

    Of course I knew after 23.Kg1 Rc8! was good for Black after 24.Qxf4 exf4 25.Rxc8 Kxe7 26.Kg2 g5 when Black is clearly has the better chances!

    Yeah after moves 30can easily get a perpetual check. Not ideal against a guy who was quite a bit lower rated then myself.

    Reply to this comment
  4. Nick Sykes

    Nov 19. 2015

    Pretty good game to say the French is my 3rd choice defence!

    Reply to this comment
  5. Eric Gardiner

    Nov 20. 2015

    In the position after 21…Nd5 in the bellerophontis vs nosykes game, White tried 22.b4 !? in Arkell, K – McDonald N(1986), presumably with the points
    a.22…Qxb4, 23.Rb1 and
    b.22…Bxb4, 23.Qh7.

    It’s not a line of the French I’ve played from the Black side. My instinct is that I’d be happy to take Black in the stem position after 18…Qa5+. However a Chessbase online search gives a 62.5% score for White (from 180 games) which sounds quite good for White. Therefore I’ve probably missed some of White’s resources and I’ll continue looking through the other games you’ve analysed for us.

    With regard to mainlines on move 18, I think you’ll find that the Semi-Slav ‘trench’ is deeper (& maybe the Najdort & Dragon ?) – but this is also not an opening I’ve played from either side.

    Reply to this comment
    • Martin Carpenter

      Nov 20. 2015

      Move 18 mainlines are rather common nowadays. Even if there isn’t one already, it only takes a few games/their analysis to generate them…..

      I guess one rather distinctive thing here is how very forcing things are here after 9 Nf4 – so you will tend to get it reasonably often.

      That’s very much more Najdorf like – and *very* unFrench like of course! There’s plenty of lines in the 6 Bg5 Najdorf like this.

      Still a bit amazed he’s managed it in practice quite this often :)

      Reply to this comment
      • Dave

        Nov 21. 2015

        Eric and Martin, it wasn’t my intention to suggest that this line was the most deeply analysed in all of chess. You’ve got to allow me a bit of room for hyperbole and colourful metaphor! :)

        I agree that the Semi-Slav, Najdorf and Dragon run deeper in places and I’d suggest that the Grunfeld, Kings Indian and possibly even the dreaded Berlin variation of the Spanish are similarly deeply analysed.

        Reply to this comment
    • Nick Sykes

      Nov 21. 2015

      After 22.b4?! Black should play 22… Qd8 23.Bxe7 Qxe7 24.Qxe7 Kxe7 Black is better

      Reply to this comment
  6. Eric Gardiner

    Nov 20. 2015

    Yes, agreed about the forcing nature. A database search shows that 9.Nf4 is the second most commonly played ninth move so Black players reaching the position after move 8 shouldn’t be surprised by it.

    Reply to this comment

Leave a Reply