Category : Openings

Openings for the Practical Player

Kasparov’s blitz comeback prompted me to think about how chess openings have changed since he retired 11 years ago. I grew up with Kasparov as my idol in the 1980s and 1990s and I adopted his general game strategy – with white obtain an advantage thanks to superior opening preparation in the main lines and then proceed to win (actually the same strategy was employed by Fischer, my other idol); with black go for sharp openings and try to take over the initiative thanks to superior opening preparation in the main lines (of Najdorf and King’s Indian mainly). Obviously the “proceed to win” was the more problematic issue for me, as generally I managed to fullfil the first part of the plan.
This general opening strategy worked until Kasparov lost to Kramnik in 2000. The reason was the emergence of the Berlin endgame – all the hard work into the (semi-)forced main lines was made irrelevant in the mud of the Berlin endgame. The beginning of the new millenium also saw the incredible rise of the chess engines – they became so strong that soon enough it was useless to try and play against them (the last human-computer match on equal terms was Kramnik’s 4-2 loss to Fritz in 2006). This levelled the field, now everyone could have deep and profound analysis in the opening and Kasparov’s opening advantage over his opponents was gone. The new generation was coming, Kasparov wasn’t getting any younger, the game required even more energy and there were other interests in his life – all this culminated in 2005 when he officially said his goodbye.
The powerful hardware and software changed opening theory. If before some lines could be evaluated as “unclear” on move 20, now they were analysed to a definite assessment until move 40 or beyond. The Najdorf is now filled with immesurable number of drawing lines and endless computer games that end in forced (!) draws on move 30, 40 or 50. The once popular Sveshnikov Sicilian, offering fantastic counter-attacking chances to black, is nowadays rarely played because most main lines lead to a draw and white discovered that playing 11 c4 (after 1 e4 c5 2 Nf3 Nc6 3 d4 cd 4 Nd4 Nf6 5 Nc3 e5 6 Ndb5 d6 7 Bg5 a6 8 Na3 b5 9 Nd5 Be7 10 Bf6 Bf6) he can obtain a solid and risk-free position with a tiny edge – definitely not what the Sveshnikov players want. And these are only two examples from the many drastic changes in various lines. This set a new problems for the chess professionals – how to memorise all that theory? But they are chess professionals, they have the best hardware and software available, they can hire coaches and seconds to work for them, they can have sparring partners to practice and try to remember all those lines, they have all the time in the world to devote to this. And what about the practical player, the one who, apart from the software, which is widely available, doesn’t have all those resources?
I happen to be that kind of practical player with such practical problems. I play less nowadays, mostly because of the ongoing trend of worsening conditions in the open tournaments, but this gives me more time to think about these issues, the main one being the memorisation of my analysis. I have always been very well prepared and whomever I played (including guys rated 2650+) I rarely had opening problems. I played main lines, Najdorf and all, with ideas and analysis of my own and I fared well. With time my files were expanding and lack of practice makes it more difficult to remember all that’s inside (when you play often during your preparations you revise all those lines, which makes it easier to remember, repetitio est mater studiorum!).
One night as I was lying in bed I started to think what to do about this problem. I started thinking for black. I realised that the main problem is the choice we give to our opponent: if he has a choice of many main lines, then that’s not so practical (if the choice is of secondary lines, then it’s not a problem as they can be dealt with more easily). So, for example the Najdorf Sicilian is not very practical: white can play 3 Bb5 (made popular by Carlsen and nowadays no less a main line than the traditional ones), he can take 4 Qd4, and then he has more or less all the 6th moves at his disposal as they have all become main lines! The other Sicilians are not much better, there is the Rossolimo again on the way to the main lines, plus the main lines. Perhaps there is some merit in 1 e4 c5 2 Nf3 Nc6 3 d4 (3 Bb5 needs to be studied as well) cd 4 Nd4 Qb6 as it cuts down white’s options. Against 1 e4 the French doesn’t seem practical, all three main lines, 3 e5, 3 Nd2 and 3 Nc3 are critical. The Caro-Kann, 3 e5 is huge today, while also 3 Nc3 (or 3 Nd2) is doing fine, while 2 Nc3 d5 3 Nf3 has also become popular. The Spanish is better from a practical perspective, especially if 3…Nf6 is played – there are many side-lines leading to it (Vienna, King’s Gambit…), plus the Italian Giuoco Piano (pretty popular lately) and the Scotch (Kasparov’s choice in the blitz), but these can be dealt with. Going for the main lines with 3…a6 gives white too much choice of opening plans, hence it’s not very practical. Playing 3…Nf6 you scare white with the notorious Berlin, so white has only 2 main options – to play 4 d3 and avoid it, or play 4 0-0 and enter it. That’s already an achievement, white’s options have been heavily restricted. The Petroff can also be an interesting choice (mind you, I speak here of practicality, not of “playing for a win” or other issues) as white is lured immediately on black’s territory, with the main line nowadays being the not forcing 3 Ne5 d6 4 Nf3 Ne4 5 Nc3. The Pirc/Modern is not really an option, white has way too many plans at his disposal, while the Alekhine is somewhat dubious in the main lines with 4 Nf3. The conclusion is that it’s important to bring white to our territory as early as possible.
Against 1 d4 there are some openings, like the Grunfeld and King’s Indian, that aren’t very practical – they allow white way too much choice, even though the Grunfeld is at the moment perhaps the best opening for black against 1 d4. But the lines there are so complex and forcing, it’s a maze! The Nimzo-Indian also allows too much choice, but lines are less forcing there than in the Grunfeld, so that’s a plus. The various Slavs used to be good, but lately white added sacrificial lines with g3 in their arsenal, making it more difficult to remember. The Queen’s Gambit seems a good choice, especially the Accepted. White has 3 main options after it, 3 e4, 3 Nf3 and 3 e3, but against all of them black is the one that chooses the system he prefers to play. Another interesting idea is Mamedyarov’s 1 d4 d5 2 c4 c5, which looks dubious, but is probably impossible to refute.
Thinking about white the problems are somewhat different. I have always strived to put pressure with white, to try to achieve an advantage, but that is elusive nowadays, especially against a well-prepared opponent. I believe that the main lines should still be played, the chance to put pressure are higher there. But then there’s the memorisation problem… Luckily, new main lines emerged, especially in the Sicilian. Here we have the Rossolimo after 1 e4 c5 2 Nf3 Nc6 3 Bb5, the check on b5 if black wants to play the Najdorf or Dragon, or the quirky 3 g3 (or 3 b3, as Kasparov played in the blitz against So) after 1 e4 c5 2 Nf3 e6. All these were made popular mainly by Carlsen, the man who epitomizes the practical approach to the openings today. The other openings are easier to cover as it is mostly white who gets to choose which line to play: any main line in the French, Caro-Kann, Pirc/Modern and for example lines with d3 in the Spanish and there you go – your repertoire is set. No forcing main lines yet still main lines that with sufficient work can pose problems in the opening. I mainly discussed 1 e4 for white here, but it is possible to do the same after 1 d4 – white gets to choose which lines to play in the Grunfeld, King’s Indian, Nimzo or Queen’s Indian. The third option, 1 c4 or 1 Nf3 has become popular today (a lot of 1 c4 in the Candidates) but there it is black who chooses the system, so more work is required. And of course white can play Nf3, g3, Bg2, 0-0, d3 and then depending on circumstances, but this doesn’t fit in with my understanding of how the opening should be played. However, that doesn’t mean it’s not a valid opening strategy!
It actually took me several nights to figure all these things out, but once I did it I was content. It is still possible to play the opening with an advantage in mind, you only need to adjust the means a bit. If in Kasparov’s era we tried to bust black in the opening, in Carlsen’s era putting even a small pressure is considered an achivement. How chess evolved!

Refutation of the Budapest Gambit

Some openings are easier to refute than others. I never had a high opinion of the Budapest Gambit – black sacrifices a pawn and then spends time to take it back, all in order to achieve a worse pawn structure. The compensation he is theoretically promised, smooth development and some chances of an attack may have been valid in the past, but today there is no attack and the pawn structure turned out to be more important than the development.

It is important to define here what I mean by “refutation.” It is not a forced win, that is for sure! If white obtains a stable and safe advantage out of the opening in all the lines and at the same time black’s position lacks perspective and has no way to evolve then I consider an opening, or a variation, to have been refuted. I hope to show in the analysis below that this is valid for the Budapest Gambit.

The following analysis is based on a lot of my own work and games and two great books – Squeezing the Gambits by Kiril Georgiev (2010) and Grandmaster Repertoire 2 – 1 d4 Volume Two by Boris Avrukh (2010).

White’s key move in several lines is Nd2, followed by Nde4. This has a poweful effect – it prevents exchange of one pair of knights, which would help black if allowed (he has less space and it also opens the 6th rank for the rook maneuver Ra6-g6 or Ra6-h6), it attacks the Bc5, another important black piece, and it liberates the way for the f-pawn. Hence I have taken as a main line the move 7…Nge5, the only way black can prevent the Nd2-e4 maneuver. But life is not easy for black there either.
Here is the full analysis:

Here is also a .cbv version for download and viewing in Chessbase.


The Charm of the King’s Indian Defence

I have been looking at some KID lines recently and I remembered how exciting this opening is. I played the KID with black until around 2000 (with pretty good results) and then from 2008 I switched sides and started playing against it with white.

The following game is from the Berane tournament I won back in 1998. It was a category 8 round-robin tournament with 14 participants and lasted for two weeks (there was even a free day in the middle of the tournament). The blasphemy of two games a day didn’t even occur to people back then. My score of 9.5/13 also brought me my first GM norm. I remember that my daily routine was get up around 11am, take a walk, eat lunch, prepare for the game, play the game, eat dinner and read Shakespeare’s Complete Works before sleep. In those two weeks during the tournament I read all the plays and almost all of the sonnets!

This game is also one of the most spectacular games I’ve ever played. The following diagram shows the star move of the game:


My opponent was Julian Radulski, who would later on become a good friend, a GM, and achieve rating of over 2600. Julian tragically died in 2013 at the age of 41, but he is fondly remembered by anyone who knew him.

The opening phase of the game is not of much interest today, but the character of play that followed is very much in the best spirit of the KID. I start my comments from the moment the real chaos begins. Here’s the game and you can enjoy it together with the lines that remained unplayed:


The Fianchetto Grunfeld and Must-Win Situations

Here’s the second article I wrote for Informator 120

The Fianchetto Grunfeld and Must-Win Situations by GM Aleksandar Colovic

The last round of the Shamkir super-tournament saw a very exciting situation – Caruana had to beat Carlsen with black to win the tournament. Even though they were equal on points he needed a win because he had an inferior tie-break. So the first question was how he would approach the opening in this delicate situation.

Caruana showed his aggressive intention as early as move 5 when he offered a pawn for central domination. He could have taken on d4 instead and transposed to the well-known exchange variation of the Fianchetto Grunfeld, known for its solidity and drawing tendencies – it served Kasparov well in his matches with Karpov as he never lost a game in it. But certainly this isn’t the way to play when you need to win. Or is it?

Let’s go back in history a bit and see what happened in another elite game in a similar situation. Round 12 of the Palma Interzonal in 1970 saw the clash of the leaders – Geller was sole first with 8/11 ahead of Fischer with 7.5/11. He was white and a draw would have kept him in the lead, so he started with 1 Nf3, 2 c4 and 3 g3, similar to what Carlsen did against Caruana. Admittedly, the situation in Palma wasn’t as critical as in Shamkir, as a round 12 game in a 23-round tournament shouldn’t be that important, but here it was a principled fight – Geller had been Fischer’s bete noire, beating him in their last three encounters, so even though the tournament victory didn’t depend on this one game, we do know that for Fischer every game was a must-win situation. So how did he react to Geller’s obvious intention to sit and make a draw? He did not lunge forward like Caruana and calmly went into the exchange variation of the Grunfeld. Geller must have misinterpreted this as he offered a draw as early as move 7, the moment he took on d5. A big psychological mistake, but he was probably thinking that he was putting Fischer under pressure with the offer, as if telling him “if you don’t want a draw, try to beat me in this symmetrical and most solid position.” Fischer laughed at the offer and simply continued as if nothing had happened. This seemed to get Geller out of his comfort zone and soon he lost a pawn, but he defended well and should still have drawn, if not for his blunder on move 71. Eventually, Fischer’s decision proved to be right.

In the 44 years since the Palma Interzonal theory has advanced immeasurably, so I am pretty convinced that if white really wants to make a draw in the Fianchetto Grunfeld, he can do that rather comfortably. So Caruana was probably right not to go there. But where did he go?
After Carlsen took on c5 and both sides castled we were actually in yet another Fianchetto Grunfeld variation, but with colours reversed (and hence a tempo up for white) – now it was Carlsen playing the Grunfeld! This line was used (rather unsuccessfully, as he drew one and lost one game with it) by Romanishin in his match against Anand in 1994. White (or in Carlsen’s case black) sacrifices a pawn in order to establish a powerful centre and have chances for an attack. But if Anand was able to difuse the line with black, certainly Carlsen was in much better situation being a tempo up? He used that tempo to land a knight on d6 to obtain an advantage and win a good game.

Was Caruana’s choice on move 5 right? I’d say yes and now. Yes, because he avoided a probable draw in case of taking on d4 and gave himself a fighting chance to try and outplay Carlsen; no, because the position objectively was better for white, a whole tempo up compared to a line which is considered good for black when a tempo down. And giving Carlsen a pawn and a tempo is rarely, if ever, a good idea.
1 4 5 6