Carlsen's 9/9 Run In The Zagreb SuperUnited Blitz

Introduction

Did you see how Carlsen won nine games consecutively against several of the world's best players on Day 1 of the Zagreb SuperUnited Blitz?

I had the pleasure of watching it live on July 8, and it was honestly the best event I have watched in my entire life. (And I have seen some pretty incredible Age Of Empires and poker streams over the years).

In this post, I'll help you relive the magic of that historic day by sharing each of Carlsen's 9 wins, with some brief commentary on the critical moments. I've auto-analyzed all the games with Lichess's 'Request A Computer Analysis' feature to make the games easier to follow.

By the way, I will be analyzing these games in full, in video format, in my upcoming course 'My Best Chess Lessons (July 2023)', which is exclusively for those with my Illingworth Chess Academy Membership - Library.

Round 1 - Caruana vs. Carlsen

Carlsen had a bit of a slow start to the Rapid (getting outplayed on the Black side of a French by Caruana in Round 1 of the rapid section). But in this game, Carlsen brought back some of his old magic, winning a 'drawn' queen endgame:

https://lichess.org/study/uubzR67B/KAqmetOr#186

Carlsen had been grinding this queen endgame for a long time already - it's well known that queen endgames are typically a comfortable draw when all the pawns are on one side of the board, even with a pawn less.

However, Caruana faltered with 94.Kg1?, not realizing that after 94...Kh5 95.Qd4 (to stop ...Kg4xg3) 95...Qe2!, White is in zugzwang, as most queen moves allow Black to either win the g3-pawn with ...Kg4, or trade into a winning pawn endgame with ...Qe1-e4. That's why Caruana gave up the second pawn on g3, but Carlsen won the endgame with two extra pawns very comfortably from there.

Round 2 - Carlsen vs. Duda

Duda has previously been a somewhat difficult opponent for Carlsen - for instance, Duda ended Carlsen's 125-game unbeaten streak (still a record, by the way) in Norway Chess 2020, and also knocked out Carlsen in the semi-final of the 2021 FIDE World Cup (going on to win the whole event by beating Karjakin in the final).

However, in this case, Carlsen was able to navigate the minefield of complications better than his opponent, in a game that could easily have gone either way:

https://lichess.org/study/uubzR67B/KCIMFxJ5#71

In a losing position, Carlsen found a great practical try in his last move, 36.c5!, which forces Duda, playing Black, to be very precise.

He had to find 36...Qb5!, not just keeping the b3 and a5-pawns defended, but also setting up a decisive counterattack on White's king with ...Qf1. White can cover this with 37.Rac4, but then 37...a4!, connecting the passed pawns, is far too strong.

Instead, Duda blundered with the natural 36...Qc6?, missing that after 37.Qxb3!, the d1-rook is under attack, meaning that the intended 37...Nxc5 now just loses to 38.Qxd1. Duda tried to keep things tense with 37...Rd5, but Carlsen was able to break the pin in time and consolidate his extra exchange to win the game.

Round 3 - Gukesh vs. Carlsen

This was a very interesting game, featuring a topical variation of the Taimanov Sicilian with 7.g4 (which is the main reason I stopped playing the Taimanov as Black, but that's another story). In any case, Carlsen was outplaying his opponent in an endgame, but Gukesh defended well and eventually, they reached this rook endgame:

https://lichess.org/study/uubzR67B/XCzUB6HQ#88

However, those of you who have played through some of Carlsen's games will know that he often wins by grinding down his opponents from equal or marginally better endgames - and this game was no exception. Gukesh was down to seconds on his clock, with a pawn less, and cracked here with 45.b5? axb5 46.axb5, seeking counterplay with his passed b-pawn.

What he missed is that Black can now transition to a winning pawn endgame with 46...Rc2! 47.Rxc2 dxc2 48.Kxc2 Kc5, as after 49.Kd3 Kxb5 50.Ke4 Kc6 51.Ke5 Kd7, the White king's advance is kept at bay, and the g5-pawn eventually is lost due to opposition. After 52.Kd5 Ke7 53.Ke5 g6!, White resigned, as Black has the opposition, and will play something along the lines of 54.Kf4 Kd6 55.Ke4 Ke6 56.Kf4 Kd5 and further outflanking to win the g5-pawn outright.

Instead, White could in fact draw by sitting on the position, as the f2-rook is stuck defending the f7-pawn. After something like 45.Kd1 Kd4, White can now happily play 46.b5, as the idea of trading into a winning pawn endgame is no longer available, and in the time it takes to stop the b5-pawn, White can take the kingside pawns. Also note that we don't have to fear ...Ke3, as we'll always have a Re7 check from behind to keep the Black king at bay (much like in the Philidor position).

Carlsen was on 3/3, but Firouzja (the world no.1 on the FIDE blitz ratings) had also started the event strongly, and the blitz was quickly becoming a two-horse race between the two blitz giants.

Round 4 - Carlsen vs. Saric

This was an exhilarating game to watch, where Carlsen had smoothly outplayed Saric for most of the game, then suddenly allowed several open goals, which Saric kept missing while both players surfed the two-second increment.

This position may be the most extreme example:

https://lichess.org/study/uubzR67B/ngttxWJU#79

Carlsen has just played 49.Qf4? - do you see why that's a blunder?

White should have started with 49.Rd3! and only then Qf4, as here 49.Qf4 loses the exchange to the 49...Ne6 fork, basically winning on the spot.

Because that was a relatively simple example, I'll also show you the second 'open goal' Black missed in the time scramble.

https://lichess.org/study/uubzR67B/ngttxWJU#95

This one comes from White playing the 'aggressive' 48.Qb8?, attacking the e8-bishop. However, Black could ignore this with 48...Ne3!, threatening mate on g2. White can defend this via. an X-ray with 49.Qa8, but then 49...Qc1! threatens ...Nf1-g3 for a crushing queen+knight attack on the White king. (The queen and knight are a great 'tag team' when attacking the king, because of how differently each piece moves). Note that 50.Rxe8 would already allow forced mate after 50...Nf1 51.Kg1 Ng3 52.Kf2 Qd2 53.Kg1 Qe1 54.Kh2 Qh1#, but if White can't take the bishop, he's simply lost.

That was the luckiest break Carlsen had in the tournament. After this game, he turned it up another notch.

Round 5 - Anand vs. Carlsen

Anand had recently come into form with a strong performance in a recent rapid teams tournament, but he has historically struggled against Carlsen since their 2013 and 2014 World Championship Matches. (To be fair, that speaks more to Carlsen being the best player in the world than to Anand specifically).

This game resembled some of the games of those World Championship Matches, however, with Carlsen outplaying Anand from an equal endgame. This was the critical moment:

https://lichess.org/study/uubzR67B/A6jOQEES#54

So far, both players have played very accurately, without any major mistakes, but White has to be a bit careful, due to the open position of his king. Yes, the queens are off the board, but there's still two rooks, a bishop and a king that can theoretically set up an attack.

Instead of 28.Bxa7, exchanging the queenside pawns for a likely draw, Anand thought it didn't matter which way he captured, and went for 28.Rxa7?, perhaps assuming that Black had to take on a4 and swap rooks for a dead draw.

Carlsen shattered that illusion with 28...Bd4!, threatening to take or otherwise deflect the e3-bishop and then mate White's king with ...Rb1. After 29.Rac7 Bxe3 30.fxe3 Rxg2, White resigned, as he's losing nearly all his pawns, and if that's not bad enough, he also has to contend with ...Rbb2, threatening ...Rg1 mate.

Round 6 - Carlsen vs. Nepomniachtchi

This game saw the rematch of Carlsen and Nepo (they played for the World Championship title in late 2021, a match Carlsen won very decisively after the first 5 games were drawn).

This game also went quite smoothly for Carlsen, who played a creative Catalan gambit en route to outplaying his opponent positionally. It's not easy to single out one position for a puzzle, but this was an interesting moment:

https://lichess.org/study/uubzR67B/Tnzsyey1#28

At this point, there's a lot of tension in the position. I think a lot of players' first instinct would be to 'save' the knight with 15.Nc3?, but that runs into ...Nb3, forking a1 and d4 to win the crucial central pawn.

Instead, Carlsen plays 15.Bf4! (not the only good move, to be fair), with the idea that, by threatening the c7-pawn while developing, we encourage Black to resolve the tension in our favour. That's indeed what happened, as the natural 15...Bxb5?! was a mistake after 16.axb5 Rxb5 17.Nc2, as Black's rook is stuck defending the a5-knight (making ...Rxb2 impossible). But if Black moves the a5-knight, the c4-pawn falls. Meanwhile, White still enjoys the long-term advantages of his bishop pair and big space advantage.

This is maybe the most complex example from our selection, but it shows how the thought process and understanding of Carlsen differs from the rest, in how he very naturally appreciates this kind of long-term compensation.

In any event, it was after this game that Carlsen realized he may actually have a shot at the 'impossible' dream of a 9/9 start for the day's play.

Round 7 - Firouzja vs. Carlsen

We were all getting hyped up for this match in the chat, as Firouzja was on a strong 4.5/6 score himself entering the game, and looked like the one person left with a decent shot of stopping Carlsen's golden run. Also, the fight for the no.1 spot on the FIDE blitz ratings was very much on :)

This game got off to a very creative start, with Carlsen meeting the Fantasy Caro-Kann with 3...Qa5!? - a move neither Firouzja nor I had seen before. Carlsen quickly gained a very promising position from the opening (despite missing a shot on move 8), as you can see below:

https://lichess.org/study/uubzR67B/DYWvJLtS#27

Out of interest, how would you deal with White's idea of Bxe6?

Carlsen missed a very nice shot in 14...Bxg3! (which I was screaming for live, but it's easier to sacrifice someone else's pieces). The point is that 15.hxg3 Qxg3 16.Kh1 Ne5 brings the queen and two knights into the attack, and White has no development to deal with the rooks also coming in after ...0-0, ...Rd8 and ...Nfg4 in some order.

If you didn't fully understand it, you can take comfort in knowing Carlsen didn't either, as he played the 'simple' 14...0-0-0, ignoring the threat to e6. After 15.Bxe6 Rhe8 16.Bh3 Nxe4, Black had a massive lead in development, while I quipped in the live chat that the completely undeveloped White queenside looked like something from a Greco-NN game from the early 1600s. Indeed, Firouzja soon got crushed in the 'Romantic' spirit by Carlsen, taking Carlsen to 7/7 and just a couple of points from the no.1 spot on the FIDE blitz ratings.

Round 8 - Carlsen vs. Rapport

GM Richard Rapport was Ding Liren's main second for the 2023 World Championship Match vs. Nepomniachtchi. Many were questioning the decision at the time for Ding to work with an overly 'creative' top GM as opposed to someone more solid, but given that Ding won the match (and set Nepo a lot of problems in the opening), I have a feeling Rapport's seconding services for World Championship Matches will be in high demand for a long time to come.

In any case, this game saw Carlsen nursing an edge against Rapport's Alekhine Defence, when sparks suddenly started flying:

https://lichess.org/study/uubzR67B/DCB5a5bf#41

At first glance, it may seem that White is simply winning, due to the trapped h2-bishop, but Rapport found a great blow to save the bishop. Do you see it?

The answer lies in 21...Bxg3!. 22.Qxg3 obviously loses to 22...Qxg3 23.fxg3 Rxe3, but the more subtle point is that 22.fxg3 is also not possible, due to 22...Qb6!, when moving the king to the 2nd rank (to save the e3-bishop) instead loses the a1-rook to the ...Qxb2 fork.

To his credit, Carlsen composed himself and played the calm 22.Kf1, simply playing on with a pawn less. This tenacity was rewarded when Rapport went astray several moves later, allowing Carlsen to first equalize and then eventually win a queen endgame that was a beautiful example of the principle 'The speed of your passed pawns is more important than how many you have' for queen endgames.

This left one match-up that many would dismiss as a write-off, but it was anything but...

Round 9 - Lupulescu vs. Carlsen

This game, against the local representative (Romanian GM Constantin Lupulescu), proved quite a tough one for Carlsen, as he was playing Black against a solid opponent trying to 'kill' any chances for Black to win the game.

To his credit, Lupulescu had been defending a worse endgame quite well for a long time, but doing so took him down to surfing the two-second increment when he reached this position:

https://lichess.org/study/uubzR67B/YuWuwRDV#76

How would you defend here as White?

The correct defence is to disrupt the pin on the g2-pawn with 39.Ra1! - not an easy move to find with mere seconds on the clock. Black can win a second pawn with the intermezzo 39...Rxh3 40.Kxh3 Rxa1, but then 41.Kg4 Ra2 42.Kf3 allows White to hold a draw, as he can put his rook behind the a-pawn and then keep a fortress behind the doubled f-pawns.

Instead, White played the more natural 39.Kh1?, unpinning the g2-pawn, but after 39...Rff2, White's g1-rook was now stuck defending the g2-pawn. After 40.Rd3 (to stop a later ...f3 and ...Rh2 mate) 40...a4 41.Rc3 Kf6 42.Rd3 (it's not a good sign when you are limited to shuffling back and forth in a dynamic double rook endgame) 42...a3 43.Rc3 f3!, Carlsen broke through (44.gxf3 Rh2 is mate) and after 44.Rxf3 Rxf3 45.gxf3 Rb2, White could not stop the queening of the a-pawn with ...a2 and ...Rb1, so White resigned.

The Aftermath

After White resigned, Carlsen showed his joy for this historic 9/9 result, which we may never seen again in chess history. (The closest comparison is Fischer's 19/22 in a 1970 Herceg Novi blitz tournament against the other best players in the world, but the top players then were considerably weaker than the top players now). There's also the well-known 'doing a Caruana', based on Caruana's 7/7 start at the 2014 Sinquefield Cup, but that's only 7 games, not 9 (although you could argue a winning streak in classical is even more impressive).

In any case, we can see some patterns in Carlsen's nine wins:

- Playing 'shortcuts' in the opening to get the opponent out of book and build an early lead on the clock;
- Playing actively in the middlegame to keep the pressure on the opponent;
- Most significantly, fantastic endgame technique to convert marginally better endgames into victory when the opponents make mistakes in time pressure.

Now that you've gone through these games, I am confident you will be much more comfortable playing technical and dynamic endgames in your own games, and perhaps even win much more often from such positions against your usual opponents.

Looking For More Grandmaster Educational Content?

If so, you've come to the right place.

I've noticed, in my 20+ years of experience in the chess world (first as a player, then as a coach and content creator) that there are three main reasons why you aren't improving your chess at the pace you would like:

1. Lack Of Awareness (not knowing what you need to know to win more chess games/avoid unnecessary losses);

2. Learning Things Incorrectly (usually from learning from too many people at once, leading to confusion when things are explained differently or even contradicted entirely);

3. Using Your Resources Ineffectively (aka wasting time on irrelevant things that won't help you win more games against your opponents).

My Solution

My new membership, the Illingworth Chess Academy Membership (Library), solves these three problems, through my 20 courses (and growing) where I show you how to win your chess games, using themes and thinking techniques that come up over and over again in your chess games.

These include:

- Tactical motifs;
- Calculation methods;
- Simple techniques for making good positional/strategic decisions;
- Understanding the key ideas and plans for specific position types

I've kept it simple, because the simpler something is, the faster you'll understand it, the faster you'll apply it correctly in your own games, and the faster you'll see the wins rack up and your rating rise to where you think it 'should' be (but isn't yet).

By the way, if you want to save 34% on the membership, you can opt for the yearly option here.

Still On The Fence? Here's My Guarantee

I've made the Illingworth Chess Academy Membership - Library risk-free for you, with a 28-day Money Back Guarantee.

You can try out the membership for 28 days, and if you're not satisfied with it, let us know within 28 days of purchase, and we'll give you a full, 100% refund.

See you inside :)