It is not a particularly imaginative line. ... Although the Exchange Variation appeals to players who are trying to draw against stronger players, allowing equality on the third move as White may not be the way to go about that. Be aware that it's a strategy that has failed miserably throughout the years.
John Watson, Play the French, 3rd ed., 70

• The e-pawns cover important central squares (e5, e4, f4 and f5) and allows Black to concentrate on flank play.
• The open d- and f-files enhance Black's grip on e4 and f4.
• Black's king is relatively safe and useful in the centre.

Are chess books getting better? Take the first volume of The Complete Hedgehog by Sergey Shipov: although I’ve read many books on the Hedgehog system, I think this is the best one by far. It’s more accessible, better explained, better written, it’s much more interesting for readers who do not play the Hedgehog at all, and it’s funny on top of that. Now, did they make such chess books in the 50s and 60s or even the 70s and 80s? I, for one, haven’t seen them too often.  

If chess books in general are really improving over time, it would be another example of the incorrectness of the widely held opinion that everything becomes worse and worse over time. (Here’s a very interesting essay on the decline of violence over time.) One of the explanations for this in the world of chess books is, of course, that there are simply more chess authors around. Another reason is that chess education has improved a lot. Yet another is that it’s much more important to write in an appealing, accessible style because of the arrival of new media such as chess videos and online lectures. I guess all are valid, and I think we should be glad that we’re living in such interesting times. These days, it’s possible to be extremely enthousiastic about a chess book almost every month.

The Hedgehog is surely one of the most difficult and fascination opening systems around, confusing and often frustrating thousands of chess fans around the world, not only club players  but also strong masters. Still, grandmaster Sergey Shipov, editor-in-chief of the well-known Russian website www.crestbook.com shows that such ignorance can actually become a force and a source of joy once you’re willing to embrace the principles of the Hedgehog with full conviction.  The Complete Hedgehog, published by Mongoose Press, is one of those rare chess books that are both enthusiastic and completely honest.

So what is the Hedgehog? I’ve never seen a better explanation than the one Shipov provides in the introduction:

Opocensky-Saemisch
Bad Pistyan 1922

Observe: four of Black’s pawns have lined up along the sixth rank (sometimes they are joined by the g- and h-pawns) and with their short, strong spines (thus ‘hedgehog’, not ‘porcupine’!), they control the fifth rank in front of then. The hostile armies complete their reorganizations inside the space set aside for them. White has four ranks, Black three. The appearance of a pawn or a piece usually signals the start of sharp conflict, in which the winner will be the one who is better prepared. Besides the outward resemblance, these kinds of setups also resemble the woodland creature in the way they deal with an enemy who is superior to them in spatial measurement: Black spends a great deal of time in strictly defensive maneuvers [sic], under cover of his pawn-spines, in order to find the right moment to leap out suddenly and bite White. (…)

The possible permutations of the Hedgehog position are huge, many of them without any real theoretical significance. Thus, for example, I remember that in my school days, when facing weak opposition, I used to play the weirdest Hedgehog positions in blitz games. For instance, I might open a game as White with 1.a3, followed by 2.b3, 3.c3, 4.d3, 5.e3, 6.f3, 7.g3, 8.Bg2, 9.Ra2, etc., regardless of what Black played, reaching a very strange and non-viable version of the Hedgehog. (…) The Hedgehog is a garden of branching paths that suddenly can come back together as one. The problem of transposing or combining variations prevents us from describing these Hedgehog setups in encyclopedic fashion – that is, move after move. So the logical approach is to divide them up, not accordin to openings, but by the pawn structure that exists after the development of the pieces is completed.

This is a lengthy quote, but some very important things can be extracted from it. First of all, the reader will note Shipov’s style: often personal and anecdotal, and he’s not afraid to make lively analogies look more than just a play with words. In this respect, I was often reminded of Ilya Odessky’s book on 1.b3, which I also reviewed and liked a lot. Is it me or are Russian authors often funnier than Western European chess book writers? Well, perhaps not: a second thing that becomes clear from the above excerpt is that Shipov adopts a completely different style than Alexander Khalifman’s book series Opening Repertoire for White according to Kramnik, which also deals (Vol. 2) with the Hedgehog.

Khalifman’s series - not exactly ‘funny’ but very good nevertheless – does use a rather ‘encyclopedic’ approach to explain openings, and from Shipov’s explanation it becomes clear why, in my mind, Khalifman doesn’t always succeed here.  (By the way, Shipov’s book unfortunately does not have a bibliography, so I don’t know whether he was implicitly referring to Khalifman here.) At any rate, Khalifman in his book does not attempt to explain what the Hedgehog really ‘is’, anyway, primarily focused as he is on variations and moves. A book that does try to explain the system from a more conceptual, almost philosophical point of view is Mihai Suba’s classic The Hedgehog. The main difference between Shipov’s and Suba’s book is, in my view, that Suba still doesn’t go all the way in describing the Hedgehog as a holistic concept that can be applied to entirely different openings than just the English after 1.c4 or 1.Nf3.

To illustrate what I mean, here are two positions from Shipov’s chapter ’Getting to the Hedgehog Opening Structure’:

These are positions from the Paulsen Sicilian and the King’s Indian Defence - both resulting in Hedgehogs. However, this is not the end of it. The fact that the Hedgehog can result from many different openings doesn’t mean it should always be expected. In fact, even one of the players aims for a Hedgehog-type setup, this is not enough. As Shipov explains:

In order to reach the required structure, one only needs to exchange Black’s c-pawn for White’s d-pawn and allow White to occupy the center. (…) I should warn my young and impressionable readers that Hedgehog structures can occur only if both sides are willing; so there’s no point in studying the Hedgehog with the aim of making it your principal system for Black, because ‘wicked’ opponents might not allow you to set it up at the board. (…) And so, obtaining the Hedgehog depends first of all on White’s desire to attack Black’s apparently passive and vulnerable position.  

This is the kind of explanation that I missed in the book by Suba, who often seems merely overjoyed by the fact that the Hedgehog should appear at all in a game, and that it should always be the right strategy. Shipov himself dismisses such wet dreams best, when he reproaches his youthful self for trying to reach the Hedgehog at all times:

No, my friends – one should not make a fetish out of the Hedgehog, striving to set it up in every situation regardless of the consequences. (…) Chess is rich in possibilities, and can’t be restricted to a catechism of spiny little beasties. 

So what exactly are the characteristics of the infamous Hedgehog – in other words, what makes it such a feared, complex and respected system? Well, you should really read the entire chapter Shipov devotes to the ‘Hedgehog philosophy’, but here are a few of Shipov’s main points:

• “In the Hedgehog, Black operates in guerilla style: avoiding direct contact, he hides in the bushes, observes his foe, waits, and then attacks at the most unexpected moment.”
• Contratry to what common chess wisdom teaches about cramped positions, in the Hedgehog, “exchanges are bad for Black, because they decrease his fighting potential.”
• “Right away, and with no regrets, I will tell you that, in the larger sense, the Hedgehog is a risky opening.”
• In the Hedgehog, Black “sets up a solid wall of pawns, behind whose protection he can arrange a universal piece placement that’s guaranteed to be a good one.”
• Psychology plays an important role: “When [White] takes over the center without a struggle, he gets a feeling of superiority, regardless of his rating. (…) It’s a drive that frequently leads to an unprepared attack.”
• “The Hedgehog displays only an insignificant part of its possibilities. Its handlers must calculate many variations during the course of the game, and consider many nuances, the vast bulk of which never will turn into actual moves. (…) Literally at every move, the players must examine Black’s possible breaks with … b6-b5 and …d6-d5, as well as White’s active possibilites. (…) So time scrambles are an objective necessity for those who play the Hedgehog.”
• “Those who feel uncomfortable in close quarters – in elevators, for example – should not be playing the Hedgehog. (…) The blood of the Hedgehogger must run cold as ice – at least, until a certain moment arrives…”

At this point, perhaps you think I am giving away the contents of the book already. Well, not exactly: all my quotes are from the first 20 pages only, and the book has over 500. The rest of the book, of course, is more concrete and deals with variations and moves. The main focus of these lines is on the so-called ‘English Hedgehog’, arising after 1.c4 Nf6 2.Nc3 c5 3.Nf3 e6 4.g3 b6 5.Bg2 Bb7 6.0-0 Be7.

The book then divides into two main chapters: the classical continuation 7.d4 and the more modern system starting with 7.Re1. And Shipov deals wonderfully with the relevant games and variations and especially its ideas. Here’s an example of his clear way of explanations (and his talent for picking good examples):

DeFirmian-Zivanovic
Smederevska Palanka 1981

Let’s study this position carefully. The knight on c5 is pinned, the c6 square is weak: the white knight is heading there. Black’s king is far from the battlefield, so the invasion of White’s rook at d7 or d6 could be very dangerous. In addition, there’s a real weakness: the b6-pawn. White’s knight is very strong on e5; he also has a pawn majority on the queenside, with the possibility of creating a passed pawn there. All these nuances, taken together, define White’s advantage in this endgame as tangible and stable.

16…Bf8

The most natural reply. Black unpins the knight on c5 and draws the sting from the white knight’s leap to c6. In that case, Black would reply …Rd8-c8 and the rogue would be forced to retreat.

17.Bxc5! Yet another unpleasant surprise!

17…Bxc5 On 17…bxc5 White would also answer 18.Na4!, when the weak c5-pawn becomes a permanent weakness. (…)

18.Na4! A very unpleasant sortie from Black’s point of view. The ‘b6+Bc5′ construction is now under pressure.

 However, a warning seems appropriate. The book is mainly devoted to systems where after 7.Re1, Black does not play the critical moves  7…d5 or 7…Ne4 but instead strives for a ‘real’ Hedgehog with the black pawn on d6. All we read about these lines is this:

 The advance 7…d5 is the most logical move, from the standpoint of the principle of fighting for the center. After 8.cxd5, Black has two cardinally different paths. On 8…exd5 9.d4 0-0 10.Bf4 Na6, we have a standard Queen’s Indian type of structure. This is a great theme for a different thick book, and would probably also be best handled by a different author. In the variation 8….Nxd5 9.e4 Nb4 10.d4, a sharp clash of pieces begins in the center, which you may get a first-hand look at from the classic game B.Larsen-S.Gligoric, Bled 1979.

About 7…Ne4 we get just one variation and the assertion that “the continuation 7…Ne4 may be labeled perfectly safe; but it still doesn’t lead to a full-fledged Hedgehog.  The positions it produces are empty and boring – like a dinner without salt and pepper: tasteless!” To his credit, Shipov is the first to admit that this selection is biased and decided by taste rather than objectivity. Still, I can imagine readers who want to know all inside-out details of the Hedgehog will be disappointed by this omission. Shipov hasn’t written a compendium but a personal account, and readers who are more interested in objective variations only, should probably think twice before buying this book.

There are probably more things to this book that could be called a little odd: sometimes, the translation seems a bit forced (’the player of White‘ isn’t really a conventional way of indicating players). As said, there’s no bibliography and neither is there an index of variations (which is particularly impractical what with all the possible transpositions, although perhaps it’s done on purpose to avoid the ‘encyclopedia’ image). Finally, I have been unable to figure our what we are to expect from part 2. Shipov mysteriously (or vaguely, depending on your state of mind) ends the final chapter Looking into the Future, with the words “Time will tell! And everything will find its place…” and his Conclusion with “Play the Hedgehog! More to come…”. I honestly don’t know what to make of this.

But frankly, it doesn’t really matter. The Complete Hedgehog vol. 1 is a great book, probably the best ever on its subject. Shipov is a highly entertaining author, a true master in explaining ideas and the underlying stragies and psychology. And all this is written in an unmistakenly humouristic, erudite and personal style that distinguishes him from many of his predecessors; in short: Sergey Shipov is your ideal chess instructor.  Now go buy his book and enjoy your holidays.

Links

• Get yourself a copy of The Complete Hedgehog vol. 1
• Read more book reviews

• Ametanoitos doesn't trust our recommendation 1.d4 e6 2.Nf3 f5 3.d5 Bb4+ 4.c3 Bd6 because in his notebook he some years ago wrote 'Do not trust the ...Bd6 idea'. He doesn't remember his exact analysis but found that following some suggestions that were recommended as leading to equality (in some other books) didn't quite equalize against natural moves. So he instead decided to go for 3...d6.
  
  

• On page 168 we say that 1.d4 f5 2.Nf3 e6 3.Bg5 Be7 4.h4 'looks unsound and is likely to backfire after 4...Nf6.' Ametanoitos claims that 'this is not as bad as they say' and gives some examples demonstrating that the line can be quite dangerous but none of them with our recommendation 5.Nc3 Ne4.
  

• Firstly he is not satisfied with our 'recommendation' 7...Nbd7, offering the game Taimanov- Lisitsin, Leningrad 1949 which continued 8.cxd5 cxd5 9.Nc3 a6 10.Bf4 Bxf4 11.gxf4 0–0 12.Na4 with a quite clear advantage to White.
  

• Next Ametanoitos is unhappy that we after our recommendation 7...0-0 doesn't mention Cox' suggestion in 'Starting Out: 1.d4', 8.Ne5!?.

• How satisfied are you with the positions resulting from move X?


• How much better can move Y really be?


• How complicated are the variations following move Y?

• Being 'too consistent' - trying to transpose whenever possible - may cost you some advantageous options.


• Transpositions rarely are just transpositions; normally there is a trade of options. You stop some options and allow others.

1. Jeroen Bosch, "The SOS Files" (offering recent games with lines discussed in previous editions), pp. 8-21
2. Dorian Rogozenco, "The Blumenfeld Gambit" (1.d4 Nf6 2.c4 e6 3.Nf3 c5 4.d5 b5 5.Bg5 b4), pp. 22-28.
3. Jeroen Bosch, "The Lewis Gambit" (1.e4 e5 2.Bc4 Bc5 3.d4), pp. 29-33.
4. Dimitri Reinderman, "The Retreat Variation" (1.e4 Nf6 2.e5 Ng8), pp. 34-40.
5. Adrian Mikhalchishin, "Steinitz’s Anti-French" (1.e4 e6 2.e5), pp. 41-43.
6. Arthur Kogan, "Slav: the Bellon-Murey Variation" (1.d4 Nf6 2.Nf3 d5 3.c4 c6 4.Nc3 dxc4 5.a4 a5), pp. 44-53.
7. Jeroen Bosch, "Slav: a Marshall Gambit of Sorts" (1.d4 d5 2.c3 e6 3.Nc3 c6 4.e4 Bb4 5.Bd2), pp. 54-59
8. Alexander Finkel, "Queen’s Fianchetto in the Alekhine" (1.e4 Nf6 2.e5 Nd5 3.c4 Nb6 4.b3), pp. 60-65.
9. Igor Lysyj, "Grünfeld Indian: Kruppa Variation" (1.d4 d5 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Bg5), pp. 66-72.
10. Igor Khenkin, "SOS in a Flexible Caro-Kann" (1.e4 c6 2.Nc3 d5 3.Nf3 Nf6 4.e5 Ne4), pp. 73-78.
11. Sergey Tiviakov, "Queen’s Indian: Double Fianchetto" (1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.a3 Bb7 5.Nc3 g6), pp. 79-88.
12. Adrian Mikhalchishin, "Sicilian: Romanishin Variation" (1.e4 c5 2.Nf3 d5 3.Bb5+ Bd7 4.c4), pp. 89-92.
13. Jeroen Bosch, "Modern Provocation" (1.e4 g6 2.d4 Bg7 3.Nc3 Nc6), pp. 93-99.
14. David Navara, "Tricky Line vs the Slav" (1.d4 d5 2.c4 c6 3.Nc3 Nf6 4. Nf3 e6 5.Qd3), pp. 100-115.
15. Alexander Finkel, "Taking Chances in the Volga" (1.d4 Nf6 2.c4 c5 3.d5 b5 4.cxb4 e6), pp. 116-122.
16. Or Cohen, "Petroff for Beginners" (1.e4 e5 2.Nf3 Nf6 3.Nxe5 Nxe4), pp. 123-134.
17. Jeroen Bosch, "Winning Ugly in the Tarrasch" (1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 cxd5 5.Nf3 Nc6 6.Bg6 f6), pp. 135-143.

1961 Match (game 5)
Fischer, Robert

Reshevsky, Samuel
(After 10...Nd5-f6)
[FEN "r1bq1rk1/pp2bppp/2n1pn2/8/3P4/P1NB1N2/1P3PPP/R1BQ1RK1 w - - 0 11"]

When I researched this opening, I was surprised to find that most games reaching the diagrammed position started 1.e4 instead of 1.d4. A little more research showed why.

Of the ~150 games I found, about two-thirds started 1.e4, and two-thirds of those continued 1...c5. Now the most common move was 2.c3 (two-thirds again), followed by 2...d5 3.exd5 Qxd5 4.d4 e6 5.Nf3. Now there are many transpositions to arrive at the diagram, but the most common continuation was 5...Nf6 6.Bd3 Nc6 7.O-O cxd4 8.cxd4 Be7 9.Nc3 Qd8 10.a3 O-O. In this line, instead of using two tempi on ...Nxd5 & ...Nf6, Black uses them on ...Qxd5 & ...Qd8.

The players can vary at several points, but the roads eventually offer a return to the Fischer - Reshevsky game. For example, many games continued 2.Nf3, when White played c2-c3 later. Also possible is 2.d4 cxd4 3.c3 d5 4.exd5 Qxd5 5.cxd4.

Most of the other 1.e4 games that were not Sicilians, were Caro-Kanns. For example, 1...c6 2.d4 d5 3.exd5 cxd5 4.c4 Nf6 5.Nc3 e6 6.Nf3 Be7 7.cxd5 Nxd5 8.Bd3 Nc6 9.O-O O-O 10.a3 Nf6. Another path after 1...c6 is 2.c4 d5 3.exd5 (3.cxd5 cxd5 4.exd5 is the same) 3...cxd5 4.cxd5 Qxd5 5.Nc3 Qd8 6.d4, and it's easy to see that the game can follow the same direction as before.

Two other paths after 1.e4 are 1...d5 2.exd5 Nf6 3.c4 c6 4.d4 cxd5, and 1...e6 2.d4 c5 3.c3 d5 4.exd5 Qxd5. The diagrammed position can arise after different moves leading to closed games: 1.c4 (1...c6 2.e4), 1.d4 (as in our game), and 1.Nf3, but these are well known transpositions that can lead to other positions as well.

> I guess the only perk is I'll have to keep changing between white and black books depending on the color I'm playing (bit of a pain if you're playing multiple correspondence games simultaneously).
>

> How many moves do you think I should import up to? Default was 20. 
>

> Should I check/uncheck any of the other options during import?

John Watson is one of the best chess authors around. Currently, he’s writing a series of high-quality opening books, focussing (as always) on understanding and framing the opening in its proper context. I will take a look at the latest volume in his ‘Mastering the Chess Openings’ series, published by Gambit Books, which is about the English Opening. I was especially interested in this part, because I have almost no experience with this opening, yet have always found it extremely fascinating. So what’s in it for people like me?

Mastering the Chess Openings Volume 3 is the third volume in Watson’s series, which, according to the back cover, ‘helps chess players achieve a more holistic and insightful view of the openings’. But why not a review of the first two volumes? Well, first of all I thought it would be a nice challenge for me to delve into an opening I knew next to nothing about. Secondly, the English is one of the most difficult openings to write about, and I was curious how Watson, who is very good at explaining ideas, did it. Finally, it was a nice opportunity to compare it to some other books on the English Opening, for example by Alexander Khalifman and Mihai Suba.

Let me start with a little experiment. More than a decade ago, I played in a local tournament where the move 1.c4 was compulsory. For this tournament, I decided to play the off-beat 1.c4 e5 2.Nf3!? which I supposed my opponents hardly knew anything about. I, on the other hand, was able to prepare some interesting ‘novelties’ in this line that was relatively unexplored. I have not seen many high-level games with this obscure variations since, but Watson calls it ‘[both] an independent variation and a transpositional tool’. After reading Watson’s chapter about it (one of the first in the book), I was immediately inspired to look into this little side-line again, and this time with the help of some computer engines. Let’s take a quick look at some of the things I found - not to show off, but to get the reader inspired as well!  

1.c4 e5 2.Nf3 e4 3.Nd4 Nf6 4.Nc3 This is a fairly basic position in this line. Apart from 4…c6, Watson also analyses 4…Bb4!? 5.Qc2:

At first glance, this doesn’t look too bad for Black, but it turns out that defending his e-pawn is a bother. (…)  The gambit 5…0-0?! looks dubious after 6.Nxe4 Nxe4 7.Qxe4 Re8 8.Qf3.

It’s always funny to be on fresh ground after just five moves, so I decided to have a closer look. When I fed this position to Rybka, I was surprised to find the engine unimpressed by White’s material lead. It suggested, among others, the funny 8…Qh4!? after which White must already return the pawn because 9.e3?? fails to 9…Qxd4. Also, 9.Nb5 Qxc4 10.e3 Qc2! is not clear at all. Before I knew it, I was looking at all sorts of crazy lines, for instance 11.Bd3!? Qxd3 12.Nxc7 Re6! 13.Nxa8 Bd6 etc. 

This may look to you like a side-line from a side-line from a side-line, but the real point is that this is one of the main reasons why the English is such a difficult, yet so interesting opening: the fight starts not at move 20, but at move 2 or 3! And this not only gives opportunity for finding novelties as early as move 8, but also touches on more profound issues. As Watson writes in his introduction:

I should warn the reader from the outset that in the English Opening, transpositions and questions of move-order abound, so much so that they will sometimes prove annoying to all but the most sophisticated players. Working out and drawing attention to these myriad issues is also no fun for the author. But I would be remiss to ignore transpositions that have a significant practical effect. Understanding move-orders means no less than getting the position you want, instead of one ou don’t like or know nothing about.

Apart from move-order questions, which Watson explains in his usual clear and accessible-to-all prose, any opening book has to deal with with actuality - with incorporating new games and ideas. Even in an opening like the English, which is surely not as ‘hot’ as the Najdorf or the Moscow variation of the Slav, this can be tricky business. Take the famous Hedgehog variation. After you’ve explained its main ideas and possible move-orders, you have to choose which are going to be ‘main lines’ and which are to be the alternatives. My conclusion is that this simply cannot be done in the Hedgehog, which is why it’s such a devilishly difficult opening to learn. I will try to illustrate this by comparing Watson’s book to Alexander Khalifman’s analysis in his book Opening for White according to Kramnik part 2.

After 1.c4 c5 2.Nf3 Nf6 3.Nc3 e6 4.g3 b6 5.Bg2 Bb7 6.0-0 Be7 7.Re1 we’ve reached the stem position of what Watson calls the ’Modern Line’. (Khalifman, by the way, reaches it by 1.Nf3 Nf6 2.c4 b6 3.g3 Bb7 4.Bg2 c5 5.0-0 e6 6.Nc3 Be7, which already gives you an idea of the move-order problems to be dealt with!). 

First off, Watson strangely doesn’t mention why 7.Re1 is played in the first place. True, it’s rather obvious, but Khalifman at least informs us that White wants to play e2-e4. Black can easily prevent this both by playing 7…d5 and 7…Ne4, so to me, these would seem to be the main alternatives. However, Watson’s main line is 7…a6 whereas Khalifman’s is 7…d6. Here, it’s Khalifman who fails to explain why Black wouldn’t want to prevent White’s main plan. Watson says of 7…d5:

7…d5 was one of the main reasons why 7.Re1 didn’t catch on for so many years. Indeed, the rook move seems to have little role in this position. On the other hand, it can be argued that …Bb7 and …d5 often don’t go that well together throughout the Queen’s Indian/Bogo-Indian/Nimzo-Indian complex, so that White can hope for a pull by playing normal moves.

In my opinion, this explanation is absolutely essential for understanding what’s going on, but unfortunately, this is basically all we get on this important line - and it’s hardly enough to grasp the fine points. For one, the line has recently become quite popular with players like Carlsen and Aronian, playing it both with White and Black. Why is this?

Also, comparing Watson’s and Khalifman’s main lines after 7…d5 we find that after 8.cxd5 Nxd5 Watson recommends 9.d4 while Khalifman prefers 9.e4. Both authors, however, do not mention the pros and cons of these two methods. Khalifman says that ‘in case of 9.d4 Nxc3 10.bxc3 Black has 10…Be4!=’ and Watson promises White a plus after 10…Be4 11.Ne5 quoting a recent game Carlsen-Aronian, Elista 2007.

Of course, improvements in recent games are a problem for all opening books - always - but here the problem is especially clear. Watson, somewhat mysteriously, also ascertains us that ‘9.e4 can be answered by 9…Nxc3 10.bxc3 with a big center for White’. Yes, and? Is that bad or good? Here we see the problem of reducing the Hedgehog to mere variations: by making a selection, by choosing a particular ‘tree’, confusion can’t be avoided in such a positional line. It’s simply impossible to draw general conclusions about the moves 9.e4 and 9.d4 in just one page without looking at a lot of games in detail.

Let me be clear, though: while I think both Watson and Khalifman fail to categorize the various aspects of particular variation in a clear way, it’s impossible to hold this against them. I have never seen a book that systematically explained and elaborated on the subtle nuances of the Hedgehog. Even the acclaimed (also by Watson himself) The Hedgehog by Mihai Suba was mostly confusing to me, even though its enthusiam was highly contagious. Watson, to his credit, seems more aware of this aspect than Khalifman, who does try to categorize the line extensively.

In general, I liked Mastering the Chess Openings Volume 3 very much, and I’ve learned a lot from it - especially how much I don’t know about chess yet. One of the reasons is that Watson gives a lot of practical advice that tells a lot about grandmaster practice as well. For instance, why would you want to avoid the King’s Indian, for example by playing the Botvinnik set-up (1.c4 Nf6 2.Nc3 g6 3.g3 Bg7 4.Bg2 0-0 5.e4 d6 6.Nge2)? Well, before answering this question, which one would expect in your average ‘Play the X System and Win!’ opening book, Watson drily points out that

The majority of grandmasters who open then game with 1.c4 or 1.Nf3 are happy to transpose into the 1.d4 version of the King’s Indian Defence if their opponents play the moves …Nf6, …g6, …Bg7, …d6 and ..0-0. Why? 1) There is a wealth of interesting systems to play against the King’s Indian Defence. 2) There is no Réti Opening set-up (Nf3, c4 and double fianchetto) that promises a serious advantage versus the King’s Indian.

Indeed, this must also be the reason why Khalifman recommends entering the main line of the KID in his series, but he doesn’t explicitly say so. This is often the difference between Watson and other authors: Watson is not afraid to mention things that may be painfully obvious to insiders, but can be real eye openers to club players. For the same reason, I found Watson extensive explanation of the differences between ‘real’ Sicilians and ‘reversed’ Sicilians very instructive.

This book is not only essential for chess players who already play the English Opening, but can be a source of new inspiration for people who, like me, have always looked at it with a mixture of wonder, bewilderment and horror.

• More book reviews by ChessVibes
• Get your copy of Mastering the Chess Openings Volume 3

We have heard from the developers of both Rybka 3 (see last month’s column) and Rybka Aquarium, the new Rybka user interface (see the April column).

Now it’s time to get a sneak preview of what Aquarium is capable of. We’ll only examine one of it’s many interesting features, but it’s a feature that will be of great help for the serious chess player.

What is Interactive Deep Analysis?

Aquarium is in a class of its own when it comes to advanced analysis. It offers comprehensive game analysis and of course infinite analysis – both with features not found in other chess software – and then there is Interactive Deep Analysis (IDeA) which is the subject of this article. As Rybka Aquarium will not be released until later this summer and some of the implementation details may still change, I will concentrate only on some of the major features of IDeA instead of giving a detailed step by step description.

The purpose of IDeA is to dig deeply into a position and return as much information about it as possible. IDeA keeps its analysis in a tree structure which is unlimited in size and the user can browse at will, even while the analysis is in progress.

Besides having a live view of how the analysis is evolving the user can also direct the analysis into the most interesting positions by excluding or adding positions and variations to the analysis queue.

In short IDeA is is highly selective search controlled by Aquarium (and by the Aquarium user). Interesting lines are analyzed deeply but weak moves are only considered briefly or not at all.

There is no doubt that Interactive Deep Analysis is a very powerful tool in the hands of the serious chess player. Let’s start by seeing what IDeA looks like in action.

Caption: Interactive Deep Analysis in progress

When IDeA is running it displays four windows:

• The familiar board window, showing the current position.

• The tree window. No, this is not the opening book window! Instead, the tree window allows you to monitor the progress of the analysis and browse the variations.

• The status window shows basic information and statistics about the analysis, such as the name of the engine, how many positions have been analyzed and how long it has been running.

• The notation window allows you to examine the position and the analysis results while it is running.

The above description of the IDeA windows only tells half the story. It is pretty accurate if you run IDeA unattended, for instance in overnight analysis, and in fact IDeA will produce very valuable analysis when running on “auto-pilot.” I expect that this simple use of IDeA will be preferred by many users, but if you are a serious chess player you may have your own views about which moves and variations are important and should be emphasized in the analysis. If you are that type of a player then you’ll appreciate the interactive part of IDeA which is described later in this article.

Stopping and Resuming an Analysis Session

IDeA stores all its analysis on disk. When the analysis starts Aquarium checks if the position has been analyzed before. If it has, then the existing analysis is loaded into the IDeA tree and instead of starting from scratch, the analysis continues where the previous analysis left off. This means that you can stop the analysis at any time, exit Aquarium and continue at a later time. The same applies even if your computer crashes (as long as the hard disk is OK) or the power goes off. The reason is that the analysis is written to disk regularly throughout the analysis process.

Note that you don’t have to restart your analysis from the same position to take advantage of this feature. As long as the position you analyze exists in the analysis tree this feature kicks in. This is a great feature for those who work on analysis over a longer period of time and gradually want to expand and deepen their analysis.

Here is an example which shows how this works. The example I used was from the famous Nolot test suite which is a collection of 11 positions. At one time it was considered unsolvable by chess engines and it is still one of the toughest test suites available. One of the hardest problems in the Nolot test suite is position number 6, which comes from the game Melaniuk-Ivanchuk, USSR 1988. It is actually the position shown in the screenshot above. Here is what I did:

• I started IDeA from the Nolot 6 position and then exited Aquarium after about 20 minutes, while it still considered 1…O-O to be the best move but the correct move, 1…axb5, was at that time evaluated as worse.

• I started Aquarium again and resumed the analysis.

This screenshot shows how the old analysis has been loaded into the tree window after the analysis was restarted. About 9 minutes later 1…axb5 took the top spot but only after about 50 minutes it became the clear favorite with evaluation approaching equality. I should mention that for this experiment I only gave IDeA 15 seconds/position as you can see in the status window in the screenshot. This is quite a bit shorter than I would normally recommend for analysis.

Caption: Analysis can be stopped and restarted

Control the Focus of Analysis

The tree window doesn’t only show the the moves that have been evaluated in the root position. This is an interactive tree which you can browse while analysis is running to examine all variations and positions that have been considered. Additionally you can use the tree to control which positions are analyzed.

Caption: Analyzed moves in the tree with evaluations

This image shows the tree window while IDeA is running. The “Move” column shows the moves that have been evaluated. The number of positions following each move that have been analyzed is displayed in the “Positions” column. Finally the “DA” column shows the move evaluation. When you click on a move in the tree window it is highlighted like 1…axb5 in the image above. You can browse the tree using the arrow keys or the mouse.

If you run into a position where you want to concentrate the analysis on a particular move or moves, then you can mark them so that no other moves will be considered in that position. Here is an example. I think that 1…axb5 is the only interesting move in the current position and I don’t want IDeA to waste time analyzing other moves. To tell that to IDeA I color the move green as shown in the following image. Note that you mark the moves while IDeA is running.

Caption: Only 1…axb5 will be analyzed

In some cases I may not be sure what the best move is, but I know that a certain move or moves are not good and I want to eliminate them from further analysis to make better use of the analysis time. What I need to do is color such moves red in the tree. The image below shows an example where I have excluded 1…O-O from further analysis. All other moves displayed in the tree will continue to be analyzed and new moves may be added to the analysis in this position.

Caption: 1…O-O will not be analyzed further

I can color as many positions in the tree as I like and thereby focus the analysis on the positions that I think are most important.

Requesting Analysis of Additional Positions

The move coloring method only allows you to mark moves that are already in the tree. But what if you want to request analysis of positions that are not in the there? Can you add your own moves to the tree and get them analyzed? Yes, that’s also very easy. When you browse the tree in the tree window the moves are automatically added to the notation window. If you run into a position which you find interesting you can start experimenting with it by making moves on the board. Those moves are also added to the notation window. After a while it may look like you can see in the screenshot.

Caption: Requesting analysis of additional positions

Here I have added several variations of my own analysis. When I request analysis of a specific position it is marked with a light-blue background color. If you look at the notation window you can see that I have requested analysis of 4 different positions. Those analysis tasks will be handed over to the IDeA as high priority tasks.

The only remaining question is how do I request analysis of a position? There are several options. You can use them both to request analysis of new positions and deeper analysis of positions that are already in the tree. The next image shows the four buttons that are used to create analysis tasks and send them to IDeA.

Caption: Different types of analysis tasks

Current position. When you click the Current Position button, the position on the board is scheduled for analysis. It is analyzed in the same way as if IDeA had selected the position automatically.

Auto-play. The chess engine plays a number of moves starting from the current position and stores them in the tree with their evaluation. The user decides how many moves should be played.

Alternative. This option searches for a new alternative in the current position.

All positions. You are not limited to adding a single position to the analysis at a time. You can add as many moves and variations as you like to the notation window and then send them all at once to the analysis queue by clicking “All Positions.”

Adding the IDeA Results to the Notation

One advantage of IDeA is that you can leave it running as long as you wish. It just keeps expanding its analysis and adding new information to the tree until you make the decision to stop. If you leave it running overnight you can check the status in the morning and make a decision if you want to continue the analysis, perhaps after using the methods described above to make sure that the analysis will focus on the variations that are of interest to you. When you want to stop the analysis you click on the ‘Stop’ button.

When you stop IDeA all the analysis that was added to the tree will still be available, but you are additionally given the opportunity to add the lines that were analyzed to the game notation. You are given the four options shown in the “Analysis finished” dialog box.

Caption: Do you want to add the analysis to the notation?

You can select to keep all the analysis in the tree and not add any of it to the game notation (“Do not add lines”). Note that you can always add the lines to the notation at a later time. Simply restart IDeA from the same position and when you stop it again you can add the lines to the notation. If you select “Add best lines only” you will only see a few variations added to the notation and just the best moves that were found along with the evaluation of each variation. Transpositions are also noted. “Add all interesting lines” can add lots of variations to the notation as shown in the screenshot below. Finally, “Add current notation” copies all the moves that you entered in the notation window while IDeA was running.

Caption: All interesting lines were added to the notation

Interactive Deep Analysis is just one example of the fresh approach that was taken in the design of Rybka Aquarium. Other interesting features of Aquarium will be described in the coming months in ChessOK Cafe.

• The London as well as the Stonewall to a great degree are based on ideas rather than exact variations and involve a lot of possibilities for transpositions and move-order tricks.
  
• In both openings you normally attempt to fortify a central bastion rather than create immediate central activity.
• Both openings frequently allow you to switch to a kingside attack shouldn't your opponent play actively enough.
• Both openings lead to characteristic pawn structures that can frequently be recognized even in the endgame.
• There are certain London lines where White sets up a Stonewall formation (normally after the exchange of the light-squared bishops but not exclusively).
• Both openings are in my opinion a bit underestimated.