In Part 1 we covered the incredible story of how Sergey Karjakin managed to come back from 2:0 down and a lost position to level the scores after the classical games of the 2015 World Cup final in Baku. That was only half the story, though, as on the very next day a no less incredible tiebreak followed. Peter Svidler and Sergey Karjakin provide the inside story of how Karjakin eventually emerged victorious after six more drama-packed games.
In the 2015 World Cup final drawing wasn’t an option:
In the first part of this article we covered the four classical games. Now it was time for the tiebreaks, which started with 25-minute rapid games. Once again, Vladimir Barsky interviewed both players for the chess magazine 64, with the interviews republished on the Russian Chess Federation website (Svidler | Karjakin). We've translated and combined the narratives of the two players.
The first game initially went Svidler's way, before he ended up losing a third game in a row:
Svidler: Yes, there were a lot of flaws in technique and some things didn’t go right, but in terms of content the first game was also good, up to a point. Then I took the decision with two minutes remaining to continue to play quite sharply for a win. Of course I wasn’t sacrificing a lot of material but, nevertheless, the decision was taken to sacrifice a pawn which wouldn’t be won back immediately and, perhaps, wouldn’t be won back at all, and to continue to play with two minutes against eight. Ultimately that cost me the game, because at some point I failed to keep a lid on things. The opposite-coloured bishop ending a pawn down was also holdable, though all things considered it shouldn’t have arisen at all. When queens were exchanged there was very good compensation for the pawn; the fact that I had to go for an opposite-coloured bishop ending was already quite alarming.
Karjakin: In the first game I perhaps ever so slightly caught him out in the opening… Although no, it’s wrong to say that, since in the end I got nothing, but at least he spent a lot of time. In principle, independent play started quite early for both of us. At some point he seized the initiative, but I managed to exploit the fact that he thought for a long time in the opening. Objectively his position was very good, but he was down a pawn and I maintained the tension. I perfectly understood that if I could at some point transfer my knight to g4 then he might potentially be in danger. I haven’t yet looked at the game on a computer, but it seems to me that purely in practical terms I didn’t play badly. In the end we got an opposite-coloured bishop ending where White had an extra pawn. At first I wasn’t sure it was drawn, but then I realised the position was holding without any particular problems. Nevertheless, I managed to find an idea which paid off in the game.
Now was the last moment for the move 78…f4! and, however I capture, he’ll have a pretty simple fortress. I read on the internet that it seems later on Black still had a way to survive, but it’s absolutely crazy – the king has to race to the queenside…
78...Bd7? 79.Bf4 Be6 80.d5+! Bxd5 81.Kc8 Bb3 82.Kd8 Bc4 83.Ke7 Bb3 84.e6 Bc4 85.Kf6 Bb3 86.Bc1 Bc4 87.Ba3 Bb3 88.e7 Kd7 89.Kxg6 Black resigned
Barsky: You’d won three games in a row, and now you had match point…
Karjakin: Yes, and here Peter visibly pulled himself together. The loss upset but didn’t surprise me. In that game I sacrificed the a4-pawn and, it seemed, I got excellent compensation for it. But after that I think I let things slip – I needed to maintain the tension. Instead I won back the pawn and I thought I’d make a draw, but in fact he had a strong knight against my bishop, with all pawns on one side – objectively it seems it was an unpleasant position. Peter won pretty convincingly.
Svidler: I played the second game pretty well. By and large I got nothing out of the opening, but the structure was fixed; I had a queen, a rook and a knight against a queen, rook and bishop, with five pawns each. I had a very clear idea of what I needed to focus on and what I should strive towards.
In general, that wasn’t my first game against Sergey which had followed roughly the same scenario. He very calmly parted with his a4-pawn – in Sochi we played just such a game.
I won the a4-pawn and was terribly satisfied, then I took a closer look and realised: if the black pieces get space to operate that extra pawn won’t be felt at all. I think that initially he didn’t have the slightest problem, but at some point he underestimated how unpleasant the ending would turn out to be when queens were exchanged.
Svidler: I converted that game quite convincingly into a win and in the break I asked myself what I should nevertheless do after 1.Nf3 Nf6 2.c4 c5 3.e3. My confident win in the third game wasn’t, of course, a verdict on what White did in the opening, but overall, it seems to me, I had a better idea of precisely what was going on in that system. And the moment I was given some kind of initiative I never let it slip. No doubt at some point I could have played still more energetically coming out of the opening but, as the well-known joke goes: “It turned out pretty well as it is!” [that’s the punchline of a joke where two friends at a funeral talk about how the dead man could have got an even better hand in a card game than the one that was so good it gave him a heart attack...]
Karjakin: In the remaining rapid games I needed to immediately go 1.e4 – to play systems well-known to me. 1.Nf3 is a good move, but I don’t have such a good feel for it that when utterly exhausted I can rely on being able to grasp unfamiliar positions. That was too optimistic. Svidler allowed d4-d5; I knew a little about that position, but I completely mixed up plans. I acted badly and I was lacking one tempo in order to finish development. And then I also blundered, although to be honest even without a blunder the position was already unpleasant.
Barsky: That was a familiar situation: once again you had to win on demand with the black pieces!
Karjakin: Yes! Strangely enough, I won pretty convincingly and I played quite a high-quality game. I’d prepared that variation for the third classical game; the plans there aren’t complicated, but he didn’t manage to arrange his setup normally. He got too clever, and completely lost track. It was roughly the same situation as in the previous game, when I completely lost my way with White. He lost a pawn, then another and then resigned.
Svidler: Then came the next attempt to make a draw with White. I think now that it was wrong to go 1.e4. On the other hand, in a rapid game it’s also wrong to put all the blame on a relatively unsuccessful opening choice, since in a ten-minute game at some point you nevertheless need to calculate variations. And the moment I tried to calculate variations I blundered a crucial pawn.
There followed 14.b4? Nb6
Barsky: You didn’t see the move 14…Nb6, right?
Svidler: No, I saw 14…Nb6 and I thought I’d go 15.а5. But when we got to that point I grasped that on 15.а5 there would follow 15…Nxc4!, which in advance I hadn’t considered such a calamity. And everything simply gets lost. After that I flailed about a bit, but in the end Black had two extra pawns and a position without a single weakness.
But after all, you could still have played on a bit? In the previous game Sergey played on for 15 moves a piece down…
Yes, Sergey surprised me a little by continuing to play with a pawn for a piece, but “each to their own...”. Experience has shown that he in particular occasionally manages to save such positions. There’s probably no-one else in the world who manages, but for him that happens. Therefore he has memorable examples before his eyes of why it’s never too late to resign.
Karjakin: That was a key game. You have to understand our condition: eight decisive games, the most crazy of matches, crazy tension, and then blitz. I think whoever played that would blunder, so it’s difficult to commentate on the last two games seriously. You can simply “give the diagnosis” that my nerves turned out to be stronger. Thanks to that I won the match.
Svidler: Then there was that Marshall… A lot of events occurred in that game, but my brain cut out to such an extent that in the end I left my b8-rook en prise. What was curious was something else: as it turned out, we’d both rechecked the line that morning! So he felt I would play the Marshall, and precisely that line of the Marshall, which is really impressive, since the line isn’t fashionable at all and I’d never played that way in my life.
But although Sergey revised the variations in the morning he couldn’t recall the very first move; we discussed it together:
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 0-0 8.c3 d5 9.exd5 Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d3 Bd6 13.Re1 Bf5 14.Qf3 Re8 15.Rxe8+ Qxe8 16.Nd2 Qe1+ 17.Nf1 Bg6
In that position, where White plays 18.g3, he thought for roughly two and a half of his five minutes and went 18.Bc2. I realised the move hadn’t been considered in my file, which means it’s bad, since all the normal moves had been looked at. The file is serious, and for the last day and a half I’d actively revised it, but it’s tough to grasp. I sank into thought and as a result my initial plan was to go 18…b4 19.c4 Nc3 – to an empty square! The fact that I was planning to provoke c3-c4 and then go Nd5-c3, but at the same time didn’t spend a second on the move 18…Nxc3, no doubt says a lot about me as a person! Joking aside: it’s actually a very clear illustration of something I already knew about myself – it’s much more comfortable for me psychologically to sacrifice material than to pick it up. The move Nc3 to an empty square really appealed to me, but to an occupied square it didn’t. There’s some kind of banality in taking the pawn! Overall, of course, that’s a good demonstration of the extent to which we were no longer ourselves by that point. I went 18…b4, he replied 19.c4. Then I realised that 19…Nc3 doesn’t work due to 20.bxc3 Qxc3 21.Qxc6 Rd8 22.Bg5, and I had to switch to Plan B. But Plan B wasn’t up to much either.
19…b3 20.Bd1 Nb4 21.Bd2 Qe5 22.Bc3 Qc5 23.Bxb4 Qxb4 24.Bxb3 Qb6 25.Re1 Bc5 26.Ba4 Rd8 27.Rd1 Qxb2 28.Bxc6
I’d ended up two pawns down, but with some kind of positional compensation. Here I landed the blow 28…Bh5.
When I put my hand down I realised that after 29.g4 I’d have to return the bishop to where it had been; meaning I’d lose two tempi just to give him some luft for his king.
But when Sergey threw up his hands and went for 29.Rb1 it was such a relief that I just didn’t look at alternatives to the move 29…Qxb1. There was a quick win with 29…Bxf3 30.Rxb2 Bxc6, and I’m a piece up. Another fast win was 29…Qxa2 – you get a version of what happened in the game, just he won’t have the a-pawn. After that I would somehow push my passed pawn to the queening square and there wouldn’t be any problems converting.
A dubious move – it wasn’t necessary to exchange that good bishop for a pretty bad one, but I liked that I was picking up a pawn with check. The ensuing position was already quite hard to convert, since White has a pawn for the exchange and the knight will be very strong on d5. I had to constantly check that it didn’t start jumping around towards my king, while I could never manage to give more than one check myself. However, at that point I had 30 seconds to his six…
31.Kxf2 Qb6+ 32.Ne3 Qxc6 33.Nd5 Qd6 34.g3 h6 35.Qe2 Rb8 36.Kg2 Kh8 37.h4 Qa3 38.Kh3 Qc1 39.Nf4 Qb2 40.Qe7 Qb7 41.Qe5 Qd7+ 42.Kh2 Kg8??
43.Qxb8+ Black resigned.
At the moment when I blundered the rook I had about 35 seconds to his six. In actual fact, at the 5+3 time control it’s not so easy to flag someone, particularly Sergey, who plays very fast and never makes bad mistakes i.e. the fact that he’s down on time doesn’t particularly affect the level of his play – his reflexes are in order. But, of course, the decision to improve the position of your king when your rook is hanging isn’t optimal.
Svidler: Funnily enough, in the final game I was quite close to a comeback. When I calmed down a little and returned to what had happened on the board I found a moment where, in my view, I was doing very well.
I’d just gone 21.g3, and he replied 21...d5. And here instead of 22.Qxd5 it was very strong to play 22.Qxb7! – particularly from a practical point of view. I think that move would have offered me very serious winning chances, particularly given the fact that I had more time by that point. From then on he played the game very well. And then I was also a little unlucky when I found an amazing trick.
28.d5!? Rxd5 29.Nh6+! But, unfortunately, there’s 29…Kf8 30.Rf1+ Nf4+! If he had to go to the h-file, then after Rh1 White would have some hopes. But from then on, of course, there was no great hope of things coming together for me.
Barsky: Well, the spectators got a lot of pleasure from your games, but how was it for you chess players?
Svidler: While the tournament was going on I tried not to write anything, but when it was over I let myself go a little on social networks. In Gladiator there’s a scene where the main hero, turning to the stands, says: “Are you not entertained?”
So on the inside it doesn’t look quite the same, but the
public got a memorable spectacle. For better or worse, we gave a wide audience
of chess fans exactly what they needed - pretty good chess
accompanied by crude blunders.
Both players were asked whether they thought the almost month-long tournament was too much:
Svidler: That’s the format; it’s impossible to make it shorter. I still recall the times when tiebreaks were played on the same day as the second game. That’s something you wouldn’t wish on your enemy! I’m not even talking about the emotional drain, but it’s simply impossible to readjust your brain to the new time control, though back then I was younger and my nervous system was more flexible. You’ve only just made a comeback in classical chess and then you’re given 25 minutes and then, god help you, 10. I clearly recall not being the only person who suffered serious problems with that – I talked about it with my colleagues. That’s without even mentioning that a game can last seven hours and then after that you have to sit down and play another six games.
So there’s no leeway for shortening the schedule. I think 128 participants is correct; it’s a qualifying event and you need to give everyone a chance. Therefore the tournament is very long and extremely tiring; with each round the significance of a mistake grows while the number of mistakes also grows, but what can you do?
Karjakin: To be honest yes, it is too long. I recall getting to the World Cup semifinal in 2009, when I lost to Boris Gelfand, who went on to win the tournament. And back then I said at the press conference that we need more rest days.
Barsky: So the format is normal – 128 players, but there should be some extra rest days?
Karjakin: Yes, they need to be added.
Barsky: Recently Magnus Carlsen suggested deciding the World Championship title using a knockout tournament. It would be very interesting to hear your opinion on that.
Karjakin: Of course I wouldn’t be against being named World Champion after the end of this tournament! But objectively it doesn’t deserve such a high status, because a great deal is decided on tiebreaks. If we play for the classical World Championship title then its fate should be decided in classical chess i.e. tiebreaks are possible, but let’s say after twelve games, not two. For all my respect for the World Cup and joy at winning I don’t think that would be right.
Barsky: So the current system – with a Candidates Tournament and a match – is normal?
Yes, absolutely normal. It works, there are sponsors and, frankly, I don’t see any point in changing something that works.
Both players summed up how the tournament had gone for them overall:
Svidler: As for me personally, much to my regret I didn’t withstand the nerves when I was an inch away from triumph. That is, however, a crucial component of our sport, so I can’t blame anyone other than myself, and when Sergey was given a chance the same thing didn’t happen to him. He absolutely deserved his win. The semifinal and final ended up being extremely tough for him, but by the skin of his teeth he extracted everything he could. I think he’ll have some criticism of the chess content, as we all do, but when it comes to fighting spirit, will to win, focus and a healthy and strong nervous system there’s nothing but cause for admiration. So I have no criticism of the format. It’s just a pity I turned out to be less resilient in the decisive moment.
Karjakin: It was one of my greatest successes, not in terms of quality of play, but in how I managed to make so many comebacks. That always means a lot! Naturally, winning the World Cup is very prestigious.
By the way, in terms of comebacks I remember that when I won the World Rapid Cup in Odessa in 2010 I had to win on demand four times. And besides that there was Armageddon in which I had to avoid defeat with Black. So actually it was five times. But after all, that was still rapid chess, without the same tension or range of emotions. Here, meanwhile, there really was euphoria.
Peter Svidler is striving to put the World Cup disappointment behind him by returning to his regular routine. Fortunately, that includes playing chess24 users live, with Peter's latest Banter Blitz session scheduled for this Friday:
We respect your privacy and data protection guidelines. Some components of our site require cookies or local storage that handles personal information.