One of the most exciting tournaments to watch, the World Cup in Khanty Mansysk, is well under way.
With every passing round the tension rises and as the usually say for knock-outs, the players with best nerves win. As I write this Round 3 finished and we already have players like Anish Giri, Sergey Karjakin and Hikaru Nakamura out of the tournament.
(I wouldn’t worry about Karjakin though – with the Candidates Tournament already announced to take place in Yekaterinburg, he can easily get the wild card spot.)
In this post I’d like to share some of the more original opening ideas the players have come up with. It is worth noting that especially in the shorter time controls these experiments paid out, at least out of the openings. Here they are in chronological order.
I always enjoy it when players come up with original ideas in the opening. Let’s hope they continue to do so!
By Alex Colovic
The non-stop playing of the top players is taking its toll. There is no time to rest, there is no time to think things through, there is no time to invent new ideas.
The players play and it is their seconds and helpers who do the hard work of constantly coming up with new moves and lines to surprise the opponents. Some are obviously better than others, but even the best run out of ideas at some point.
It is not surprising therefore when there are so many games played with the same openings. Just sticking to what is safe ensures against big surprises, but that also leads to less excitement.
In the penultimate round of the Zagreb Grand Prix Mamedyarov employed the Scandinavian (also known as the Centre Counter) against Caruana. It was yet another tournament that wasn’t going well for Mamedyarov and it appeared as if he didn’t care. Hence, 1 e4 d5.
Black got the usual slightly inferior but extremely solid position from the opening and eventually drew the game. Players note very carefully what everybody else is playing and they probably concluded that the Scandinavian isn’t good enough for classical games, but the opening exploded in the Paris Grand Prix in both the rapid and the blitz.
Mamedyarov played 2 more games with it – he beat Anand and lost to Dubov. He stuck to his treatment of 3 Nc3 Qa5, the same line he played in Zagreb against Caruana.
Other players tried their luck with the Scandinavian as well and they all went their own ways. Nakamura also played 2 games in it, choosing 3 Nc3 Qd6 in his loss to Caruana and draw with Giri.
It was Duda who employed it the most. Perhaps tired a bit of his trusty Petroff, he kept choosing the peculiar development after 3 Nc3 Qd8 that involed the double fianchetto. His results were great – he lost 1, won 3 and drew 1. The good thing about his opening was that he always played the same, which is very practical as it requires very little preparation.
While things worked out better than expected for Duda, I doubt the Scandinavian will catch up on the highest level. It will now be subjected to serious analysis and they will find ways how to deal with it. Surprises only last so long on that level. That is why they need a constant supply of fresh ideas.
The public had its quick rush with the Scandinavian and now we’re waiting for the next one. I wonder what it will be.
By Alex Colovic
This one is from my newsletter. Enjoy it!
Nowadays chess theory develops by the minute. New ideas appear constantly and it is difficult to follow every single one of them. But sometimes it is not necessary to do so.
The openings in chess have their cycles. A characteristic cycle of an opening is that it becomes popular as a lot of new ideas and plans appear in it; people start to play it a lot and it is analysed in great depth; in these stages it is vitally important to follow all the latest developments and novelties; after a period of this process, which depends on the complexity of the opening, the main directions and variations are crystallised and it becomes easy to determine which are the most critical lines; here already the impact of the novelties is not as strong; then the main directions are explored and if the ideas are getting exhausted the opening is slowly abandoned or other, new directions appear, usually at an earlier stage.
To be aware where a certain opening is in this cycle is very important from a practical aspect. It is not very practical to start playing an opening when it is in its first stages, when it’s not clear what the main lines are and which plans are the most promising – though it is in exactly this phase that the elite players are thriving. A typical example for this is the Giuoco Piano – there are many plans and move-orders for both players: White has plans based on Bg5, Be3, the queenside expansion, the knight-transfer to f1 and then g3 or e3, the central d4-push, while Black can play with …a6 or …a5 (favoured by Ding Liren), go for …d5 or not (depending on White’s move-orders), go for …Ne7-g6 or/and …Be6. While this leaves scope for surprises and new developments for the elite, it is very confusing for the rest.
A different example is the Najdorf. This ever-green opening can never be abandoned, and is in the stage of new ideas appearing, as early as move 6, with White players trying out moves like 6 h4, 6 a3 or 6 Nb3. The reason for this is that in the traditional main lines the variations have crystallised and Black usually has a choice of a couple of reliable and straight-forward lines to choose from. For example, in the 6 Bg5 line he can either go for 6…Nbd7 or the delayed Poisoned Pawn 6…e6 7 f4 h6 8 Bh4 Qb6, championed by Vachier-Lagrave. Or in the English Attack, he can choose the lines with …h5, which are all the rage now, the sharp variations with 10…a5 (after 1 e4 c5 2 Nf3 d6 3 d4 cd 4 Nd4 Nf6 5 Nc3 a6 6 Be3 e5 7 Nb3 Be7 8 f3 Be6 9 Qd2 0-0 10 0-0-0 a5) or even the complex lines after 10…Nbd7 11 g4 b5 12 g5 Nh5.
Therefore it is practical to take up an opening in these later stages, when one only needs to study the current theory, which has narrowed down thanks to the decades of “distillation.” In these stages the effect of novelties is diminished because normally they cannot drastically impact the variations, so they are mostly of a “local” character.
A similar case is the Sveshnikov Sicilian. As the last World Championship match implied, and Carlsen’s later games showed, Black is perfectly fine in the main lines after 7 Bg5, so White players are looking for alternative ideas, Caruana’s attempt of 7 Nd5 being such a case.
I understand that in order to be able to recognise these cycles a certain opening erudition is required. This however shouldn’t discourage the opening student. Speaking from personal experience, if you dedicate your efforts to a study of an opening, meaning you go back in time to the opening’s beginnings and then analyse its development over the years, you will start to recognise these changes that all openings experience.
As an additional advantage, having studied an opening in such a way will often enable you to be ahead of the curve and even foresee the changes that still haven’t occurred! That is one of the rare superpowers in chess!
By Alex Colovic
As with all things in life, the answer is “it depends.”
It depends on many things – how you play it, against whom you play it, which lines you play. And also there is the distinction of “simple” and “simplified.”
I made the transition from 1 e4 to 1 d4 back in 2008 when I realised that I won’t be becoming a GM soon. My rationale was since I wasn’t getting the title then I may as well study and learn something new. Creating a repertoire is never a simple task, but I knew what I wanted. I wanted safety so I immediately started with “everything g3.” With the benefit of hindsight, not a bad approach. Later on I expanded on it, but the love for the fianchetto has remained ever since.
Chessable’s author IM Christof Sielecki followed up on his Keep It Simple: 1 e4 with Keep It Simple: 1 d4. Not surprisingly, he also advocates the fianchetto against pretty much everything.
After watching the introductory video I got a general idea of all the lines he is suggesting. I very much liked his sincerity where often he would say that Black is fine in certain lines, admitting that there is no advantage in the variations he’s suggesting. And this is true, as in modern theory it is impossible to claim an advantage in any sound opening.
Christof’s “simplicity” is shown in the suggested move-order for the repertoire: 1 d4, 2 Nf3, 3 g3, 4 Bg2 and 5 0-0 against pretty much everything. But then things get complicated.
The repertoire does not shy away from entering mainstream theory and mainstream theory is anything but simple. The author recognises the danger of suggesting off-beat lines just in order to avoid theory – this is a dead-end street in the long-term – so he enters the main lines when that is the correct way to go. Good examples for this are the transpositions to the Catalan, the Fianchetto Grunfeld, the Tarrasch Defence and the Queen’s Indian.
I checked several lines that were interesting to me. I was curious to see how the repertoire compares to my notes.
In the Tarrasch Defence the proposed line is to play the main line with g3 without Nc3. I discovered that Sielecki improves on the proposed line by Aagaard and I couldn’t find anything wrong with it.
In the Reversed Grunfeld lines (1 d4 Nf6 2 Nf3 d5 3 g3 c5) he digs deep, following a forced line from many games and then considers 4 candidate moves for Black on move 14, with White having a pull in all of them.
In the Catalan you cannot avoid theory and the repertoire doesn’t intend to. The good aspect of the proposed move-order is that it avoids all the Catalan lines with early 4…dc so Black’s choices are limited to the Closed System without the subtlety of …Bb4-e7 or the Open System with 6…dc.
In the Open System the very recently played game Grischuk-Nakamura from the FIDE Grand Prix posed serious problems to Black in the line 7 Qc2 b5. Sielecki couldn’t have known of this game when he worked on the repertoire and his analysis follows an earlier game, Gelfand-Ponomariov from 2011, but stops short after 16 Nc4, limiting himself with a verbal explanation when theory goes on and Black is probably fine later on. As with all theory, it never stops evolving, and the improvement by Grischuk (16 Nd4!), who in turn was following an analysis by Avrukh, will serve well the diligent student.
In the line 7 Qc2 a6 the suggested line is 8 a4. This leads to microscopic (if any) advantages for White, but it is extremely solid so it fits well with the whole idea of the repertoire.
I was somewhat surprised that against the Grunfeld with …c6 the recommended line is 7 b3, allowing the sharp equalising try 7…dc 8 bc c5 and after 9 Bb2 Qb6 10 Qc1 (instead of the more popular 10 Qb3). This is a very concrete line requiring good memorisation, somewhat at odds with the generally slower pace of many other lines.
Against the KID the repertoire doesn’t enter the main lines but suggests 6 b3 instead. Apart from the move-order issues (as Black can play …c6 instead of …d6, toying with an idea of a transposition to a Grunfeld) this is a very practical approach – not many KID players like the sight of 6 b3 and the lines are not bad at all for White either!
The only lines where White doesn’t fianchetto his bishop are the ones where Black plays an early …c5, where the suggested plans are based on playing d5 and follow mainstream theory which guarantees White an advantage.
As usual, Christof is very thorough and I couldn’t really find a weak spot in the repertoire. Curiously enough, he often says that he used LeelaZero to suggest interesting ideas.
After going over the repertoire I cannot say that it is “simple” at all. It is simplified to a certain extent, but it requires a lot of study in order to absorb the whole material. If this doesn’t bother you (and it shouldn’t!) you have a very solid and high-quality material to rely on that will most likely serve you for many years to come.
By Alex Colovic
Have you ever wondered how an elite chess player’s preparation looks like? If you have, now you have a chance to find out.
Chessable managed to get Pentala Harikrishna to create a repertoire for White against the French Defence. He is by far the strongest player to have a repertoire on the platform and I was curious to see what material he prepared.
As an 1 e4 player who has always played 3 Nc3 against the French in the past I was excited to find out that Harikrishna also suggests the same choice for the repertoire. My reason for more or less abandoning 3 Nc3 was that all 3 main Black defences: 3…de, 3…Bb4 and 3…Nf6 became so popular that every single one of them required a ton of work to keep updated with many viable options appearing for Black.
So what did Harikrishna come up with?
As usual the Chessable repertoire comes with a full video material, so instead of going over the variations (which I did later) I decided to watch the video. While watching I was looking at my own notes and compared the lines.
Harikrishna covers a lot of 3rd moves for Black, but these are inferior so I concentrated on the 3 main ones.
The biggest surprise for me was that against 3…Bb4 he suggested the exchange variation 4 ed. At first I was disappointed, as I couldn’t believe White can pose problems there. Still, the facts that an elite player suggests that line and that the move has already been played by other strong players were enough to convince me to give it a closer look.
Harikrishna’s concept is curious. He is trying to put pressure in a symmetrical position a-la many lines in the Petroff Defence. He does in fact revolutionise the line by not playing in the old-fashioned way with a3, forcing Bxc3 when White takes bxc3, but rather goes for the more positionally solid approach with the maneuver Nc3-e2. A useful rule of thumb he gives is that White should always put his king’s knight on an asymmetrical square from Black’s: if Black plays …Ne7 then White goes Nf3 and if Black goes …Nf6 then White plays Ne2.
After the initial scepticism I found the lines quite compelling. The main reason was the easiness to play them as White’s play is simple, straight-forward and very safe. Quite a different picture than the lines after 4 e5 c5 5 a3 Bc3 6 bc and now either 6…Qa5 or 6…Ne7 when White can easily be put to the sword of the counter-attack if he messes up.
Against 3…Nf6 I saw that Harikrishna’s suggested line is 4 e5 Nfd7 5 Nce2. I felt inner satisfaction when I saw this because I studied and prepared this line back in 2000 (!) as a back-up for my main choice of 4 Bg5. Since my analysis is rather dated, I was glad to find out many important improvements Harikrishna has discovered, perhaps the most shocking one being 12 Kd2! in one of the lines.
Against the Rubinstein 3…de he suggests the latest trend of 7 Ne5 (after 4 Ne4 Nd7 5 Nf3 Ngf6 6 Nf6 Nf6). I have never studied this line in depth so I took time to check it. One of the main characteristics of the course comes to the fore here – Harikrishna doesn’t shy away from showing his own preparation and novelties. For example, after 7 Ne5 Nd7 he analyses the move 8 Be2, a novelty on move 8! Both Karjakin and Dominguez have played 8 Bf4 against Meier, but Harikrishna goes for the endgame and gives deep analysis of his new idea. He could have saved this idea for an important game, but instead he chose to share it with the Chessable students. I found this very sincere.
As with any elite player, the analysis is deep and thorough, all carefully checked with an engine. Another example is his own admission that he forgot his preparation in the game against Rapport from March this year and then he proceeds to show an improvement and an important novelty on move 10 in one of the lines.
I found the video pleasant to watch as Harikrishna’s voice is calm and his exposition is measured. I invested several hours in watching him explain his lines against the French and I feel I came out of it equipped with new lines to try against this opening. For me, that is a lot.
Normally when it is sweet we want it to last and this particular one lasts for 50 minutes. Does it qualify as short then?
You probably realise by now that I’m talking about Chessable’s Short & Sweet courses that are free and provide a short overview of an opening.
In the course of my recording for the Najdorf Sicilian I took time to record and create a Short & Sweet repertoire. I aimed to provide a sustainable overview of the opening, covering as many different lines as possible.
With the Short & Sweet Najdorf I wanted to show that by studying a relatively small number of lines you can gain confidence to try the Najdorf, without fear of not knowing theory or being busted because of it. The main difficulty lay in the selection of the most characteristic lines for each option by White. Therefore I took care to select lines that show the most typical development of the play in a nutshell.
In the video I expanded on the presented variations as much as possible. I wanted to explain with words the main ideas and concepts so that even with a concise course as the Short & Sweet you could feel confident about the Najdorf.
Since we’re talking about short and sweet, I’ll keep this short. The sweet is at the end of this text following the link.
After the release of my Najdorf repertoire for Chessable I received a lot of positive feedback – people were happy with what I managed: to condense the Najdorf to a manageable size.
You can read the details in my previous post where I outlined my principles and selected lines. They appear to have stricken a chord with my students – with emphasis on understanding rather than memorisation it seems that I have made the Najdorf much more approachable.
With Chessable’s video sync feature in full operation the next step for the repertoire was to produce video explanations for the material. So at the end of April I sat down and recorded more than 11 hours of video material.
In the video course I took special care to take a look at the problem moves my students have encountered. By carefully checking the situations where the students have made the most mistakes I paid special attention to these positions and tried to explain them as thoroughly as possible. Often I drew parallels between similar lines or where ideas from one line could be implemented in another (like the …Qb8 idea in the 6 Be2 and 6 g3 lines), thus trying to clarify any confusion and to aid the learning process.
Another bonus for the video material is that I took the time to analyse all the model games I provided for the repertoire. The purpose of these games was to explore typical Najdorf positions, themes and maneuvers. Even though the games not always correspond to the actual repertoire lines, their intent is to give you a feeling for the Najdorf and I tried to convey this in my video analysis.
I am very satisfied with the work I did and I believe this video update fittingly complements the written commentary. However, none of this would have been possible without the professional help of David, Chessable’s CEO and the studio in his office in Swindon, the lovely English town where he resides. Thank you, David!
Or perhaps I should have named this post “How to Make a Draw Among Friends?”
The ongoing tournament in Ivory Coast, part of the Grand Chess Tour, saw the game Karjakin-Nepomniachtchi in Round 1. It featured the Gothenburg Variation of the Najdorf.
A line with rich history and alas, no future. Introduced in Round 14 of the Gothenburg Interzonal in 1955 by the Argentinian trio of Najdorf, Pilnik and Panno in the games against the Soviet players Keres, Spassky and Geller, respectively, it was handsomely met by the spectacular move 13 Bb5! (first played by Geller, as the story goes) and it resulted in the Argentinian fiasco with three beautiful wins for the Soviets.
It was considered that the line was refuted by this move, but three years later, citing sources from a Soviet magazine, 15-year old Bobby Fischer improved with the incredible 13…Rh7! in the crucial game against Gligoric in the last round of yet another Interzonal (in Portoroz) to secure qualification for the Candidates Tournament in 1959. (Another story goes that Fischer actually asked Gligoric about this line at some point earlier in the tournament – while swimming – and Gligoric said he didn’t know much about it.)
Fischer’s move is considered best even nowadays and the analyses have shown that this line ends in a draw. This has been known for a very long time.
However, another thing has also been known for a very long time. And that is the fact that the move 11 Qh5! (instead of the flashy sacrifice 11 Ne6) leads to an advantage for White, thus basically refuting the Gothenburg Variation.
Here’re the details (note that I’m using lichess for this one as chess.com has been having some issues with the game viewer):
Please bear in mind that by “refutation” I don’t necessarily mean a lost position for Black, but rather a prospectless position at the end of the line, making the whole variation unappealing to play. Similarly, you can take a look at another refutation here.
It is quite apparent that both Karjakin and Nepomniachtchi played this “game” in order to make a spectacular draw, as I am pretty sure that both knew the best way how to play against the Gothenburg Variation.
Some time ago Carlsen brought up the subject of these kind of “games” and it gave rise to some controversy with the accused Karjakin and Mamedyarov denying vehemently. But if you have been in the business for long enough, you learn to detect these things and understand what is happening under the surface.
Circumstances aside, bringing up the forgotten page from chess history, the Gothenburg Variation, is something I appreciate, so thanks to both players for that. After all, they could have played the Exchange Slav instead…
Bobby Fischer said that a turning point in his career came when he realised that he can play for a win with Black too, even against the strongest players.
I assume that this realisation must have happened during the period 1960-1963, because at that time he was still playing 1…e5. Fischer had relatively short draws in the Cordell Variation of the Ruy Lopez where he didn’t get chances at all in the Buenos Aires tournament and the Leipzig Olympiad in 1960. Even though his last two 1 e4 e5 games are from 1963, more or less from that period onwards he switched completely to the Najdorf. From the mid-60 he also incorporated the Alekhine Defence, but he never returned to 1…e5.
Nowadays there is no elite player whose repertoire does not include the 1…e5 move (barring the honourable Frenchman with 2 surnames and Nepo). After 1…e5 all of them play the Berlin Defence.
The World Champion is not an exception. He built his solidity around the Berlin and it has served him well for a very long time.
Then something happened during the preparation for the match with Caruana. Whether he thought that he should play different type of positions against this particular Challenger, or he wanted to change his approach with Black generally, or perhaps something else, Carlsen decided to adopt the Sveshnikov Sicilian as his main defence against 1 e4.
An interesting decision, the Sveshnikov. Kasparov used it several times as an alternative to his Najdorf in the early 00s and Gelfand made it his main opening in his World Championship match against Anand in 2012. Both did rather well. Still, generally speaking, it was considered that the opening lost a great deal of its former glory thanks to the line 11 c4:
This line stabilises the centre and largely kills Black’s dynamic counterplay. This change of character wasn’t to the liking of many Sveshnikov players and slowly the opening gave way to the eternally dynamic Najdorf.
So when Carlsen adopted the Sveshnikov, the first question I asked myself was, what does he have in mind against 11 c4? We are still waiting for the answer, as the only time so far this was played against him was in the blitz game with Nepomniachtchi, where the Russian was late for the game (!) so Carlsen accepted a draw after 11…b4 12 Nc2 a5 13 g3.
Theory aside (for this I invite you to join my newsletter using the yellow box on the right, since soon enough I will send a theoretical overview of Carlsen’s treatment of the Sveshnikov), I am very curious about the psychological implications of the use of this line of the Sicilian.
Let’s start with what Carlsen has said himself. He was quite open saying that he liked playing the position where he felt Black had easier play. This concerns primarily the 7 Nd5 line, played very often against him since the match with Caruana. In his own words, the engine gives an advantage to White, but when preparation ends then it is easier for Black to play who often has excellent compensation for the often-lost h5-pawn in view of an attack. His careful study of these positions has helped him understand the dynamics much better than his opponents – after the match he’s beaten van Foreest, Navara and Karjakin with only Caruana drawing.
Dynamism is present in the whole Sveshnikov, but at various degrees. If the above-mentioned line with 11 c4 lowers it to the minimum, the lines where White takes 9 Bxf6 gxf6 are one of the most dynamic ones in the whole theory of chess openings, while the positions after 9 Nd5 Be7 10 Bxf6 Bxf6 are somewhere in the middle. The level of dynamism is determined by White’s choice of lines, meaning that Carlsen is prepared for a different type of battle in each game he plays the Sveshnikov.
Does Carlsen’s newly-found liking for the dynamics mark a new phase in the development of the World Champion? Carlsen’s results in 2019 show that it appears that he has found his reliable mainstay opening against 1 e4 that, and this is crucial, allows him to play for a win with Black against any opponent.
Just like Fischer and Kasparov before him with the Najdorf, Carlsen may be equally successful with the Sveshnikov. It won’t be much about the actual theory and preparation, but more about the understanding of the inherent dynamics of these positions that will determine the outcome of his future games. So far, Carlsen seems to be ahead of the others in this aspect.
Still, nowadays the moat is always narrow as people learn to catch up very quickly. They will learn to understand the dynamics and Carlsen’s advantage may disappear. To quickly adapt and change is a vital advantage in today’s game. But that time still hasn’t arrived for Carlsen and, besides, he can adapt too.
While I don’t think this played a major role in his choosing of the opening, compared to the Najdorf, the Sveshnikov has one important advantage – it is more practical. If on White’s move 2 the possibilities are the same, after 2 Nf3 d6 White has quite a lot of sensible ways to avoid the Najdorf: 3 Bb5+, 3 Bc4, 3 c3 (not to mention the other, less critical ones) and even after 3 d4 cd there is 4 Qd4 and 4 Nd4 Nf6 5 f3. White also has the move-order 2 Nc3 with various ideas based on g3 and Nge2 while keeping the option to play d4, in addition to the Grand Prix Attack.
In the Sveshnikov these options are very limited. There is 2 Nc3 Nc6 3 Bb5 (as the Grand Prix is considered harmless after 2… Nc6) and after 2 Nf3 Nc6 only two options remain – the Rossolimo 3 Bb5 and 3 Nc3. This fact that White doesn’t have too many ways of avoiding the Sveshnikov may be very important in practice.
Carlsen’s choice of the Sveshnikov is definitely good news for the chess public. We already saw some and will definitely see more exciting games with the World Champion being aggressive with Black. The only danger lies in the possibility that people stop playing 1 e4 against him, but then again, as they say, you can run, but you cannot hide…
The Najdorf. The ultimate desire for many, a common-place occurrence for others. I have a feeling that when it comes to the Najdorf there are two possible feelings about it: you either love it or you’re petrified of it.
I have always fallen into the first category. The reason is that I’ve played it since I was a kid and the main reason for the fear of the second category, the mind-boggling amount of theory, didn’t bother me because I learned it as I was growing up and that knowledge built itself inside me.
I have been approached way too many times with lamentations of “I would like to play the Najdorf, but there’s so much theory, I cannot study all that.” And I understood them, starting to learn the Najdorf from scratch is not a task for the faint-hearted.
However. When I stopped to think a little about it, I realised that a lot in the Najdorf is based on good understanding. And that understanding mostly revolves around the d5-square and what should be done about it. When I pushed the concept a bit further I realised that with the help of modern theory there is a chance for not-too-theoretical repertoire to be created.
My main goal with this repertoire was to provide the student with a feeling for the Najdorf positions. Once you have that feeling then even a surprise in the opening will not be enough to disturb you or prevent you from finding a good antidote. I provided ample textual explanations of the critical positions and the chapters with model games and typical strategic and tactical motifs were aimed exactly at this.
The second goal was of course the theoretical knowledge. The Najdorf is theoretical, there is no going around it. I tried to present the basic theoretical knowledge necessary for the student to be able to play the Najdorf. I concentrated on the modern lines, the ones that are most popular today, with enough information on the traditional variations that are less trendy nowadays.
The repertoire consists of the following chapters: Introduction, The Positional 6 Be2, The Sozin 6 Bc4, The English Attack 6 Be3, The Aggressive 6 Bg5, The 6 f4 Line, The Fianchetto 6 g3, The Modern 6 h3, Odds and Ends and also the afore-mentioned chapters Model Games and Strategy and Tactics. The theoretically heaviest ones (meaning with the most forcing lines requiring memorisation) are The Sozin and the Aggressive 6 Bg5.
Wherever possible (or practical) I proposed the typical Najdorf move 6…e5 as a reply to White’s 6th move attempts. Theoretically speaking Black is in fantastic shape in these lines so there was no reason not to take advantage of it!
I think that the main advantage of this repertoire, and this aspect took a lot of hard work, was that I succeeded to narrow down the theory to a manageable level. A lot of secondary variations were explained in textual terms rather than lines – from my experience of working with students I discovered that they remember better when things are described with words rather than with moves.
The books I used to learn the Najdorf from, The Sicilian Defence by Lepeshkin (in Russian), Najdorf for the Tournament Player, The Complete Najdorf: Modern Lines and The Complete Najdorf: 6 Bg5, all by John Nunn, were hefty tomes. I never found them difficult and always enjoyed working with them, but to be honest, had I had the resources available now, I would have definitely chosen the much more efficient ways provided by modern technology.
My effort to provide a concise yet profound Najdorf repertoire and to give a chance to everybody to try this wonderful opening is now before you to judge. I can only hope I did a good job.