You've played six World Championship matches till date. Which one of those do you think resembles the Magnus Carlsen-Fabiano Caruana World Championship match the most?
I don't think anything resembles this. Maybe my match with Boris Gelfand comes the closest. But we managed one pair of decisive results at least.
The first game of the match began with a Rossolimo; Caruana played 3.Bb5. What are your thoughts on his opening choice?
Well, Magnus has been dabbling in 2.Nc6 this year. So, you know, it was in the air. And about the Rossolimo, what can I say about the third move of an opening " it's there and if you have studied it, you can play it.
Indian chess legend Viswanathan Anand takes a boat trip during the resting day of Norway Chess 2018 tournament in Stavanger. Reuters
Also, in the first game, Carlsen was winning after he took the pawn with 31.Qxh5. The position looked like something Carlsen could easily convert. Why do you think he messed up?
It is very strange. This should have been routine for Magnus. Magnus' own hero from five years ago would have won this position quite easily! It was one of the big misses of the match. What was striking is that Magnus' play until this point had been exemplary. He played at a very high level; everyone praised him. Before cashing in, he could have slowly improved his position. I would only put one caveat in this situation " it is very easy to sit at home and say this is how you should have done things. God knows, I have spoilt many winning positions as well. It's not my aim to make light of this. But it's a missed opportunity. There is no way of going around this.
After the first game, did you feel that this World Championship would be a very exciting one given that the first game itself was so full of fireworks?
Yes. Especially, if he (Magnus) had won this game, the impression of someone simply crushing his opponent would be hard to get past because Magnus, with black, outplayed Fabiano so easily. And then, had he won the point, Fabiano would have had to chase him in every game. So that opens up a lot more space¦ it's a miss.
In the second game, Fabiano had played Rd8 on his 10th move. What was your impression of this? Have you looked at it or prepared it? Or do you think it is a one-off idea that Caruana pulled off?
I had looked at it and I had some notes but, evidently, Fabiano went much further. I know Hikaru (Nakamura) looked at it as well because we had discussed it briefly. It was good opening work. This is the sort of thing that should happen in a World Championship match. You have so much time, you can go really deep and clean some stuff and play moves that might not be the most fashionable. I don't think it was one-game stuff. I think Fabiano intended to repeat it, had Magnus gone for 10.Rd1 again.
The third game was rather dull but in the fourth game, we saw this 6¦Bc5 idea in the English Opening. This is becoming a lot more popular than the usual lines in the English.
I guess the idea of 6¦Nb6 followed by 7¦Be7 is still leading if you look at the total number of games. I wouldn't say that 6¦Bc5 is completely taking over. But if you see, last year, I played this line against Fabiano at Saint Louis, and then there was the famous Dubov-Karjakin discussion. And so it flared up and then dropped out of the radar for a while. But after this match I think this is just going to be one of the main lines.
And what would you say about the pawn sacrifice from game five " the move 6.b4?
It's well-known; and the theme is well-known. It was also seen in the 1992 Fischer-Spassky rematch where 4.Bxc6 bxc6 was met with 7.b4. And even what Caruana played in this game has been played before. I would expect both sides to be well prepared. Obviously, Fabiano because he played it; but also Magnus seemed to know it well.
And then this move, 4.Nd3 was played in the Petroff Defence by Caruana in the sixth game. I mean, these moves like 4.Nc4 or Nd3 in the Petroff have never really been taken seriously but they are now seen being played in a World Championship. What's your take on this?
You try to play these unusual moves to get a game. But even these unusual moves can be analysed with a computer. You might have heard this famous conversation between Grischuk and Kovalev that Grischuk had looked at the sequence that happened in their game with Nf4-Nd5 and Black going Nc6-Nd4 (laughs). It's quite funny that they had gone that far. Grischuk knew it was dead and it was not a big deal.
So it's like you can prepare anything and play it at least for one game?
Yes! And when it works, it works. There are cases when your plays like that are harmless. But if your opponent doesn't know it on that day, you get something. I guess, Fabiano's preparation has been thorough but we don't really know because Magnus has probed very little. I think he has just probed the surface.
Conversely, on the other side, because Fabiano stuck to only one thing and kept on chipping away, we now know that Magnus' Rossolimo preparation was very good and his Sveshnikov preparation in this 7.Nd5 line gives the impression of being slightly more improvised.
Yes, usually, World Championships are meant to probe this way. But Magnus didn't seem to be doing that.
Funnily enough, I had predicted for this match, and I had hoped for this match, that there would be very little probing. I thought both of them would take the fight to each other with both colours. I expected more wandering around because both of them are capable of multiple openings. But in fact, Fabiano stuck to 1.e4 and Magnus stuck to the Sveshnikov for all of the six games (with black). I expected that Magnus might even dabble into three (openings) but this just goes to show that I was completely wrong. But it has made the Sveshnikov much more interesting. Suddenly we have discovered a new position which is much richer than we had thought. Let's see what tournament practice tells us.
On the other side, I think Fabiano chose the most solid openings. He decided on the Petroff and the Queen's Gambit Declined¦ I think his idea was to give very little space to Magnus. This is a strategy I had tried as well, and even Sergey (Karjakin). It's a reasonable strategy because, when you play a World Championship, your opponent's preparation is going to be quite good. So, you need lines that would withstand a lot of scrutiny. That's the concept. But then, Magnus, at several points just seemed to not even want to try with white.
In fact, the funny thing is that, I think Magnus had been looking for a fight much more with black while Fabiano looked for the same fight with white. So, this meant six games were quite interesting whereas the other six, frankly, hadn't been anything to watch. I didn't even bother looking at them much afterwards. I just checked the openings for the exact details and just moved on.
Let's talk about Caruana's "missed win" in game six where the challenger was held to a draw despite having an extra knight on material count. The computers suggested that Caruana could have won had he moved his bishop to the h4 square on move 68.
It was obviously not easy. I haven't delved into it entirely but it seems that this is not a forced win to begin with, which means that white could have used his bishop to corral the knight in some other way. I saw (Peter) Svidler's recap of this ending and still I had forgotten all the details. And the thing is when Svidler did the recap of this ending, he kept on checking the notes and the corresponding squares. It's a very difficult endgame. But I think I understand the position after 70.Ng1 pretty well now. The essential idea is to lose a move.
The computer shows Black wins with 68..Bh4 here. But had Caruana played the incredible 69.Bd5 Ne2 70.Bf3 Ng1!! they would request metal detectors immediately! No human can willingly trap his own knight like that. pic.twitter.com/ypBjXv3QCA
- Garry Kasparov (@Kasparov63) November 16, 2018
However, I have not yet grasped what exactly white should have done a move ago or why the bishop on c4 is lost and something else¦ I need to go over this bit slowly and try to understand it. But it's fascinating because suddenly you discover something new about an endgame.
I sort of concur with everyone else when they say that they might have solved this if they were given this as a puzzle by the method of elimination but there is no other way. There is no way to know that this position (on move 68) is significant and the position from two moves ago is not. Not a single human being knew that until the computer pointed it out.
So would you spend time studying this endgame? Does it benefit you as a chess player?
Yes, some positions are just interesting enough that you must study them. It is conceivable that someday I might have a similar position on the board and this knowledge might be useful in some vague way. But I am not necessarily studying it to score points.
If classical chess wisdom were to be applied, white should have had a clear advantage after Caruana's 10¦Qd8 in the seventh game. What's your thought on it?
Why should white have a clear advantage? Even white has moved his knight back to d2. It's interesting to note that these Nd2-b3 plans are the most effective when the black queen is on a5. But you just go back to d8 and ask: 'what is your knight doing on d2?'
I have stopped trying to understand chess by the logic you mentioned because the computer keeps pointing out exceptions. I remember having conversations with my second where he would tell me about a move and I would say this has been played in 100 games before with an extra tempo. How can this move make sense? But I realised that chess doesn't seem to lend itself to that sort of logic " if you can make use of the extra tempo, good for you. But more often than not you just study the new position on its own merits. If it works, it works! That was always the joke in our team. If someone said something illogical, and we couldn't refute it, we had to agree it works!
And do you think the 7¦Nd5 line in the Sveshnikov has the capacity to come at par with the main lines of the opening?
I think, now, it definitely does. Already the fact that Kramnik played this in the Olympiad shows that the line is worth revisiting. I have never really played someone who tried it, so I haven't really gone into this so much.
Given that Magnus was going to play the Sveshnikov the whole match there was an argument for spending more time here, but you know this logic can take you anywhere. You can always say you should have spent more time here and if they play something else the criticism moves on.
In the eighth game, Caruana was playing quite aggressively and seemed to be on the brink of a victory when he put the brakes on with 24.h3. Do you think he should have continued his attack in the spirit of the position?
You know, neither 21.c5 or 23.Rd1 particularly strikes me as being difficult. They are fairly obvious moves. And about Rad1 and Rfe1, I would say a beginner can play it without understanding why and a computer can play it by seeing the mate.
What was really amazing was these lines where the computers flung white's pawn up to h6 to kick the bishop to h8 and then brought the rook in with Re1-e6. Here, black is helpless despite the fact that he has a protected passed pawn. If you see all that then it's remarkable. But, frankly, for players of Caruana and Carlsen's level, to make these moves with the hand is trivial.
And the devil is always in the detail. Caruana might have seen Qh5 and Rfe1 but there might have been some little move that bothered him and he couldn't work it out. I mean, after you have seen the computer line, everything looks logical. But's it's not logical before. So, it's a miss but not a trivial one. And the computer's insight is fascinating only if you see the whole line.
And, I don't know, with a lot of these moves, the problem is that once even if you know the evaluation " if someone were to tell me the position is +1.9 " I would start checking the moves I have rejected before. I think, at the board, it's much harder. That's what it goes on to show.
And do you think Caruana's 17.Bxf3 was some sort of a World Championship mistake? I mean, one wouldn't play a move like this in a normal game.
Indian Grandmaster Viswanathan Anand plays against Bulgarian Grandmaster Veselin Topalov (not pictured) during the first round of the Candidates Tournament in 2016. AFP
Not at all; I actually agree with Fabiano's logic. I think computer's evaluation is much too high. What this is for white is a lasting advantage, which is unpleasant in nature. But it is nowhere near winning. First of all, if black swaps all the rooks, I think white's advantage is very small because black has many fortresses.
Having said that, white can play on endlessly, especially with one pair of rooks. It is not that the advantage is big, but it can never be neutralised. So if Carlsen would have kept one pair of rooks and played h5 at the right moment, it could be torture.
The tenth game was the tensest game of the match. You were commentating on it live and you felt that black had some chances of attack there. What's your opinion now?
I mean, when you have pawns on e4 and f4, you definitely have chances of attack. The question is whether they are any good. There was this moment when Caruana could have simply played Bxb5. I was on the St Louis show and Maurice asked me, "Bxb5. The computer is screaming an advantage for White. Do you agree?" After that I couldn't answer his question in an intelligent way, because of course I agree. Once the computer says it's good, I understand why it's good. But if the computer would have said it's bad, I would have spent a few minutes trying to understand why it is bad! And this is always the problem when you try to understand the game with that crucial nugget of information. It's all you need to know " whether it is good or bad. And I had mentioned my game with Aronian (Tata Steel Chess 2013). All I knew I was better here. I couldn't remember a thing otherwise. But that was enough for me to find everything. Because in every position you have any doubts, you can tell yourself this is supposed to be better so a move must exist. And this is very different from going to a position saying I don't know how this is going to turn out.
After Bxb5, the plan with ...Rf6 to g6 or h6 turned out to be trickier than it looked. So there are some problems for white to solve, but he does solve them. It was an amazing game. I don't think either side was lost at any moment, but it felt dangerous, practically dangerous. And this is the interesting point about this match. At many points we have landed in positions where I would say to myself, I could lose this with white and I could lose this with black in a practical game. And still we got six draws! These positions have been so rich, but neither side fell apart. So we have to compliment them for that rather than just speak about missed opportunities.
Game eleven was sort of dead but the final game of the match witnessed this sudden draw offer by Carlsen. Were you surprised by Carlsen's decision?
I was shocked. It was strange for me that Magnus didn't want to make even three more moves. He could have played Bd7, or even Ra6, Rb8 and b5. Make three more moves and you can still offer a draw. Why offer it here?
I tried to think of a similar example. Sometimes when I am lost out of the opening and then I am defending the game for 20-25 moves, really suffering. And then my opponent makes a move that not only loses him the advantage, but puts him in a slightly worse position. Essentially you should wipe your head and start thinking I have an advantage and start playing for a win. But sometimes it is very difficult to make this switch. You have been thinking for a draw for so long that it is very difficult to make this switch.
But Carlsen was never lost in this game. He didn't have to switch. I think he was fighting a battle in his own head. He wanted a draw badly in this game, because was worried that he might lose it and he was not able to switch that thought. After making all the allowances, and understanding the circumstances, I am still surprised that Carlsen offered a draw here.
Another big surprise was Caruana's downfall in the tiebreaks. Having matched every wit of the world champion in the classical leg, he simply collapsed in the rapids.
I think the match started to go wrong for Fabiano very fast. In the first game, I think, there was a crucial moment where Magnus played 4.e4, right in the opening. And, I think, Fabiano had come to the board with the intention of not falling behind on the clock but, unfortunately, this was one of the situations where it was worth spending a little time, just to understand what you should do. After Magnus had castled, I think he had exactly what he wanted to get.
It was actually unpleasant for Fabiano, though he played reasonably well. There was one moment where he could have gone 19...Nb7 instead of 1¦Nb5 but this was well within the margin of error of these things. I think, he had seen this resource of taking his knight to d4 and felt that he was holding. It was very hard to calculate all that.
This was one of those positions where if are thinking clearly and if you hold, it gives you some confidence. And I think if Fabiano had held this draw " and he was very close " the match could have gone differently. But it's typical also that the person who is trying very hard wins after a couple of mistakes. You never lose these positions with one mistake, you lose them with two or three. So, Fabiano, I think, let it slowly slip away. But full credit to Carlsen for seeing under the circumstances that 37.Re7 was crucial.
And after that, I think, it's basically hopeless. Magnus got back his confidence. I know how wonderful it feels to get a point's lead after not being able to break through (for 12 games).
In the second game again, Carlsen played really well. On top of that, I think, Fabiano underestimated the counterplay that black had after 21.c5 and 22.c6. You need more time to see that it looks deceptively good for white.
Game three was basically Fabiano trying to hold on in a hopeless match situation. He got some very small chances at the end. It showed that even Magnus' nerves can affect him. After 26.e5 and the simplifications, suddenly 31¦Qb2 happened. But he had the presence of mind to play 32.Qd6 and stabilize the position. And then one or two further mistakes and it was over. But you can't really fault someone in the situation. When you are two points down (in the match), it's pretty hopeless.
In the end, I think, Magnus showed his strength in rapid chess. I think his ability to play almost any kind of position and play it reasonably well is a huge advantage for him in these situations. As for Fabiano, he just didn't get into the match in the rapid stage.
Sagar Shah is an International Master and CEO of ChessBase India.