February 19, 2012 1:29

Review: The Tarrasch Defence

Review: The Tarrasch Defence

The Tarrasch Defence is a great opening. It is both solid and active, it has a rich history and it has been endorsed by numerous world class players, including Garry Kasparov. Dr. Siegbert Tarrasch himself gave the good example when he wrote, more than a hundred years ago, of ‘his’ 3…c5: “All along I instinctively recognized this move as the right antidote to the Queen’s Gambit.” 

I’ve played the Tarrasch myself for over 10 years, and with tremendous results. It’s fair to say that if FIDE had only counted my games with this opening for their rating calculation, I would have had 200 points more than I have now. The only reason I recently stopped playing it is because I believe (stupidly, no doubt) that it’s necessary to vary one’s openings from time to time for fear of getting utterly bored with chess. 

Another interesting fact is that despite my excellent results, I’ve never felt the slightest need to read an opening book on the Tarrasch Defence – often Black’s moves are so natural that there really seems no point. This is a huge difference with other openings, such as the King’s Indian or the Benoni Defence, where a lack of theoretical knowledge is likely to result in an endless bunch of duck-eggs.

But a new book by Jacob Aagaard and Nikolaos Ntirlis, The Tarrasch Defence, published by Quality Chess, has made me change my mind. The two authors (though I got the impression Ntirlis did most of the original work) present so many fresh and fascinating ideas in this old opening that it’s impossible to put down. It’s also a very objective and sensible book, in which the old opening is both treated with respect and is challenged to defend itself against computer-age scrutiny and rigour. 

Although The Tarrasch Defence is extremely useful for White players (I’ll tell you why further down), it is mainly aimed at players who want to employ this opening with Black. Thus, the old main line with 9…cxd4 is only mentioned as part of the introduction, and in general the authors waste no time analyzing black alternatives that they regard as inferior to the ones they’re suggesting. 

Instead, Aagaard and Ntirlis spend most of the book on what they call the ‘Modern Treatment’: 

1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 exd5 5.Nf3 Nc6 6.g3 Nf6 7.Bg2 Be7 8.0-0 0-0 9.Bg5 c4! 10.Ne5 Be6 

PGN string

11.b3 h6! The “early h6”-theme recurs time and again in this book. Here, the authors confidently write:

The classical reply here is 11…Qa5, but extensive analysis has convinced us that the text is the best move, giving Black an active position with good expectations of equalizing.

In fact the only attention the move gets is in the introduction, where the authors mention German IM Blauert’s well-known shock novelty 12.Qd2! Rad8 13.bxc4 Nxd4!!N (Pedersen-Blauert, Gausdal 2004) and amazingly, Black seems fine in all lines. However, after the more prudent 13.Nxc6 bxc6 14.Rfd1 Bb4 15.Rfc1! (as mentioned by Schandorf in his 2009 book Playing the Queen’s Gambit) Black, according to Aagaard and Ntirlis,

has no realistic chance of achieving equality. 

12.Bxf6 Bxf6 13.Nxc6 bxc6 14.bxc4 dxc4 15.e3 Qa5

PGN string

A big tabiya in the Modern lines of the Tarrasch Defence. White has an impeccable pawn structure and a strong bishop on g2; Black has the two bishops, a passed pawn on c4 and the hope of breaking in the centre with …c5. The static features of the position suggest that White should be better, but Black has excellent dynamic possibilities at his disposal.

The authors don’t stop here but go on to analyze the subsequent positions in great detail. The same is true for other cases where 11…h6! proves to be surprisingly strong (such as after 11.f4, 11.e3 and 11.Rc1). Although this gives the book it’s a very ‘personal’ touch, I also found it somewhat disappointing that all these lines I came to love so much, like 11.e3 Nd7 (which is also not bad for Black) and 11.f4 Ng4!? (very exciting as well), are hardly looked at in this book. 

But then again: tough luck for me! If Black has more promising ways to play – and Aagaard and Ntirlis make a convincing case for it – then I should be prepared to kill my darlings and start looking at the new kids on the block. No boring old theory, but exciting new territory! 

And I soon discovered I might actually be better off after all. What to think of the following crazy and almost completely unknown line?

11.f4 h6! 12.Bxf6 Bxf6 13.f5 Bxe5! 14.dxe5 Qb6+ 15.Kh1 Qxb2! 

PGN string

Every true Tarrasch player should just love this stuff, which gets its own chapter. The only problem, of course, is that there’s so much new material here that improvements are probably just around the corner (although my Rybka and me couldn’t find them so far). That’s only natural, but it does require a little soberness from the reader from time to time. 

So unless you’re a pro, my advice is to just sort of browse through these pages and then simply go play the position – much more fun than trying to memorize all these unlikely moves that Tarrasch himself wouldn’t have found  behind the board either…

Despite such spectacular outbursts, I felt the book’s real asset is its excellent treatment of various endgames where Black can or cannot achieve perfect equality. Playing the Tarrasch has taught me to try and correctly evaluate certain highly typical endgames, for instance with an isolated d-pawn or with doubled c-pawns. These endgames are often not as bad as they look, and they are always very difficult to play for both sides. Despite this, the authors are very clear about what’s at stake: 

Let’s be honest about one thing. There are few openings in which it is easy to prove equality for Black. And the Tarrasch Defence is not known as one of them! Black has to work hard for equality (…).

They go on to discuss the endgame with doubled c-pawns. The following fragment is worth quoting in full:

11.b3 h6 12.Bxf6 Bxf6 13.Nxc6 bxc6 14.bxc4 dxc4 15.e3 Qa5 16.Rc1! Rac8 17.Qa4! Qxa4 18.Nxa4 Be7 19.Nb2! Ba3 20.Rc2 Bxb2 21.Rxb2 Rc7! 

PGN string

(…) Although the White pawn structure is more solid, it does not hold a lot of dynamic potential. The black pawns have both the weakness of a solitary existence and the freedom of no one relying on them. It is surely a blessing and a curse at the same time. 

The front c-pawn is passed, and though hard to support, you can easily imagine a situation where it moves one square forward and forces White to rapidly retrace his steps to control the first rank. Likewise with the pawn on c6; weak certainly, but also able to break up the white centre with …c5 at some point – but only if the d5-square is under control. White would love to answer d4-d5 and get his own passed pawn.

White does not have similar freedom. At the moment the pawns on f2 and h2 offer full protection for the entire family, but this is also a limitation. If White plays e3-e4, the most obvious active pawn move, the d4-pawn becomes an immediate target. This is very relevant, as Black will often try to play …Bd5 in order to provoke this advance. 

Black can challenge the soundness of this structure in the long term with …f5, …g5 and …f4 but the time this takes must be taken into consideration. This type of long-term plan requires a rather solid foundation in the centre and good coordination to achieve. 

A final idea for Black is to activate the rook from f8 via b8; here it may go to b6 or b4 (or maybe even b2) and then switch to the a-file. This is seen over and over again, and is in line with the endgame principle of activating your strongest pieces first (the rooks being stronger than the king, whose value in the ending is generally rated as 4 points). 

White does have one pawn advance that makes sense. If he manages to advance his a-pawn to the 6th rank, it can be used in a future invasion of the 7th. White’s advantage is of a static nature, long term and based on the quality of his pawn structure. He wants to neutralize the black activity based on the c-pawns, and advance his central pawns once all counterplay has been neutralized. (…)

I find such fragments irresistible, and in fact this is only the beginning of the discussion of this particular endgame, which goes on for no less than 6 more pages. 

I could go on and on about the many beautiful variations in this book, but the truth is that it is crammed with fantastic stuff - really too much to mention in one review. So let me just say that the authors treat the ever-important Timman Variation (9.dxc5 Bxc5 10.Bg5 d4 11.Bxf6 Qxf6 12.Nd5) with due adoration and skepticism (I’ve always felt the line to be both overestimated and underestimated at the same time!). Here, too, they improve existing theory as they go along in many crucial lines. 

Even more interesting is the authors’ “big discovery” that a ‘minor’ deviation in this opening turns out to be so dangerous it’s worth two separate chapters:

1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 exd5 5.Nf3 Nc6 6.dxc5!

PGN string

This is one of those simple-looking lines I myself have never dared to look at closely (and has fortunately never been played against me!). The problem for Black, which is of course well-known, is that after 6…d4 7.Na4 Bxc5 8.Nxc5 Qa5+ Black will regain the piece but White will have the two bishops. Yes, you’ll say, but isn’t Black terribly active now? 

Well, that’s what I always thought too, but the authors forced me to evaluate the endgame that Black gets after: 

9.Bd2 Qxc5 10.Rc1 Qb6 11.e3 Nf6 12.Bc4 dxe3 13.Bxe3 Qb4+ 14.Qd2 Qxd2+ 15.Kxd2 Be6 16.Bxe6 fxe6

PGN string

We think that Black looks okay, but things are never that easy of course. In our opinion Black has improved chances in the ending when a set of knights has not yet been exchanged, so that the bishop is less able to roam the board uncontested. (…) The endgame is ever so slightly better for White, but to call it a significant advantage is maybe a bit much. 

As so often in the Tarrasch, Black is comfortably within the drawing zone in the ending. People take on less attractive positions in the Petroff, Berlin, Queen’s Gambit Declined and other openings on a daily basis. If you are a technical player, this is possibly something you want to consider as part of your repertoire against weaker players, but probably not against your peers.

Jacob Aagaard and Nikolaos Ntirlis invite their readers to think about these positions for themselves, rather than to just memorize what they prescribe. Would you mind playing this position as Black? If you don’t, then you’ve got what it takes to become a real Tarrasch player – not scared of isolated pawns, bishops or your engine indicating +0.41. The Tarrasch Defence invites you to be scared of nothing.



Share |
Arne Moll's picture
Author: Arne Moll


Andre From Outkast's picture

A fantastic review of a stellar book. (:thumbsup:)

Rini Luyks's picture

The Tarrasch is very dynamic indeed, but this statement: "It’s fair to say that if FIDE had only counted my games with this opening for their rating calculation, I would have had 200 points more than I have now"...
Sorry, but here I want to be an "unbelieving Thomas": show me!

Arne Moll's picture

I've checked my personal database, RIni, and my score with Black in this opening is +17, =13, -3 against an average opposition of 2214. You get the point.

chessobserver's picture

That's a funny review for sure! The book is as good as it gets for a Tarrasch book, but the Tarrasch defence is a pathetic one and this book proves it. In all chapters black is desperately fighting for the draw in some pawn down or bad but "defensible if you are houdini" endgames. If you like computer chess then you will like the analysis, which stands until the next version of houdini...

Scruffian's picture

Haha spot on. Ridiculous amounts of impractical computer analysis to justify playing an at best inferior defence. The simple fact is that unless you're a GM a lot of the positions they give as OK for Black are far easier to play with White. A really hyped up dishonest and typically arrogant (just look at the issues around Excelling at Chess to see another example of how totally classless an author Aagaard can be) piece of trash. Only buy it if you need something to hit your computer with the next time Houdini suggests some inane line.

Arne Moll's picture

I think it's actually the other way around: GMs may find some of these positions unappealing for various reasons (that's why the opening isn't seen very often) but for amateurs they are absolutely fine because in practice White doesn't know how to handle them. That at least is my experience.

Thomas's picture

I think at GM, or rather world top level the opening ran out of fashion after Kasparov lost twice against Karpov in their first WCh match back in 1984, including this model game (47.Ng2!!):
These were the only games (out of 20 at chessgames.com) that Kasparov lost with the Tarrasch. Karpov may have simply been the wrong opponent ... Arne is perfectly right that the opening is playable at amateur level - like other openings that aren't fashionable among GMs and entail a certain strategic risk.

kevg's picture

**disclaimer - I have only looked at the sample of this book and several reviews**
For low-graded club-players like myself, with limited time for opening study, the Tarrach can be played against almost anything. I have enjoyed Jacob Aagaard's pevious work and as there aren't really any good books on the Tarrasch I really wanted an excuse to buy the Quality Chess offering. The problem is,in my games I rarely get to the point at which the book starts (frustratingly, most of my opponents refuse to take the d5 pawn). I suppose this is the reason this series is called 'Grandmaster Repertoire', but from my point of view there just aren't enough books dealing with early variations and general ideas. Anyway, reveiwers seem to like this book so I guess I can't really criticise it for being something it was not intended to be rather than what I would personally like. I just felt the need to air my feelings! Any suggestions for a lower-level book would be gratefully received (I have the Schiller Queen Pawn Opengin book).

AljechinsCat's picture

Despite I respect the Tarrasch very much the second ending looks a bit as an "Test your endgame strategy" exercise -- for the first player. Black might be okay but playing for a win for him will be difficult here (as far I can see he has nothing substantially to play for). I´m happy to face more Tarraschs from now on :).

DirkBredemeier's picture

I have played the Tarrasch for quiet a long time before switching to indian openings. I agree with Arne, that it is perfectly playable und a univeral weapon against 1.d4 und many 1.c4-Lines at amateur level.
The Problem with Argaards book seems to me the following: It goes too deep into details in the given Variations for players below - lets say - 2400. It does not make sense for an amateur - like me - to reach a 2600-Level in theory for the first 8 to 12 moves and then - after "end of book" - being struggling to keep a 2200-Level in the middlegame. I think kecg above describes the same problem.
In German there are already 2 or 3 Books about the Tarrasch, albeit 15 or 20 Years old. So use somethink like "Nunn´s Chess Openings -NCO" and Games with comments from a database.

Honest John's picture

I completely agree with Scuffian comments. When I saw that QC bought a computer with cables under water I became suspicious about their books and see this as a wrong path. I only have one so far and for me these books are not for GM´s because they simply turn on their computers and neither for amateurs because there are no explanation on ideas, themes, typical middlegames or endgames because that needs work and time and I guess Aagard only put his name on the cover. It´s totally impratical and a strong player can do better with a database and study of games. Today we see a lot of marketing and friedly critics but books made by computers are a piece of trash. Between 3...Be7, 3...Nf6 and 3...c5 any intermediate player and good student knows that the first two are far better moves. Kasparov, Kramnik and others came to these conclusions in the nineties about several other openings.

Andrew Greet's picture

First I should make it clear that I am an employee of Quality Chess, and played a role in editing the Tarrasch book. I would not normally stop to comment on a forum such as this, but I felt that the comments of the ironically named "Honest John" required a response.
Firstly, on the subject of the Tarrasch in general, I don't see much point in debating this as it depends on one's personal preferences and what you are looking for in a chess opening. If the Tarrasch is not for you then fine; to tell the truth it doesn't really suit my style either.
However, on the following factual points the above poster is just plain wrong:

* Jacob did much more than put his name on the cover. It is true that in the first instance Nikos came to us with a repertoire he had produced; however Jacob checked every single line in the book and made substantial improvements along the way.
* Just because we have a powerful computer in the office, it is a complete fallacy to assume that no human thought goes into the books. Plenty of recommendations in this book are not the computer's first choice.
* If the above poster had bothered to look at the book he would know that it contains a great deal of explanations about strategy, middlegame/endgame motifs and so on.

Honest John's picture

Ironic honesty? People can judge for themselves... but it was not me who made a book on the Queens Indian using a variation without putting the author name on it, but putting that minor fact aside and focusing on QC and speaking of honesty, truths and lies, I can say that the book "True Lies in Chess" from GM Fabregó (wich I don´t know) is just another example, where pairing some truths, we don´t know if the spanish GM is refering to some lies of the classics or some of his own about the classics!

Andrew Greet's picture

I don't know anything about the Queen's Indian book, and I don't understand what you are trying to say about the True Lies book. This is off-topic anyway. What I am concerned about here is setting the record straight about what is true and what is false - especially regarding the slanderous suggestion that Jacob merely slapped his name on the cover, which questions the ingetrity of both Jacob personally and Quality Chess as a publisher.
Anyway I've made my point so this will be my last comment on the subject. One thing I do agree on is that people should decide for themselves - preferably after having actually looked at the book.

Andrew Greet's picture

I made a type - should be *integrity* - perhaps the moderator could correct this and then remove this note?

FM Matthieu Freeke's picture

I must confess these harsh words on 'hard to remember' and 'endless streams of computer analysis' make me smile. At least 1.d4 players are still arrogant and not likely to invest time in this pretty defense which is good news! People will be people and say silly things! I guess some people are still looking for the =+ in their defenses!?

I want to compare this opening with the Petroff. Black seeks tactics early in the game and forces white to be either ambitious or settle for equality. Just as in the Petroff black gets early tactics but has to be prepared to defend some typical rook endgames. Theoretical tactics generally end up in endgames.
Just like some Marshall lines sometimes a pawn down for the bishop pair. Play is rather forced at times all the way to the endgame. The truth is these themes and endings are well explained. Studying just this book will improve your chess greatly. A fantastic effort.
I'd give it a 9/10

Mauro Casadei's picture

I find surprising (but maybe I shouldn't...) that some people are so angry at the authors/editors of this book.
A couple of points:

- the perfect book does not exist.
So it's a matter of tradeoffs. You get something you like and something you don't like.

- the question is "is what I like worth the money I spent?". I personally think it is. For 28 € (35$) the authors tell you the main lines that deserve to be studied more accurately,and many analyses are original. The ideas are described throughout the book.

- "a good player can do better with an engine and a database": maybe this is true, but how much time will it take ? 10 hours, 20 hours for selecting the right material and organizing it and sort out the main manoeuvres and strategica patterns ? One hour of work, depending on your profession may be worth 10-20-30-100$. A quick calculation tells us that such a book is good bargain for anyone.

- The book is honest. No doubt about that. There might be improvements (I'll post a wish at the bottom of this post), but its is quite aligned with the policy of the Quality Chess Seris "GM Repertoire". The new generation of opening books is much better that many database dumps produced in the 1990s.

- Finally, I do not understand the comment about the Tarrasch being a bad defence, . I personally do not agree, at least for <2600 players, but this is not the point. The point is "Why buy a repertoire book of an opening you find bad?"


Having said that, I would like first of all to thank the Staff of Quality Chess: finally we got some books that do not find the shortcut of sidelines to avoid the challenge of dealing in details with main lines; this task is tough but there must be someone that at least tries to do it.

One wish, though: I found A LOT CLEARER the format of the book "Beating the Open Games" by Marin, than any of the following GM repertoire books (Avrukh on d4, Marin on the English, and now Aagard on the Tarrasch).
The reason is that such a foirmat allows BOTH a summary table "Encyclopaedia-style", but also a good section on commented games with ideas.
This makes it easier to separate the parts that are useful for STUDY purpose from thos that will be used for CONSULTATION purpose.

Thanks a lot

Latest articles