It’s all even halfway through the 2024 World Chess Championship between reigning champion Ding Liren and challenger Gukesh D, as they fight for chess’s highest title over 14 games in Singapore. The champion Ding Liren, coming off a nightmare run of form in 2024, sensationally won the first game with the black pieces, before the 18-year-old phenom Gukesh leveled the scores in Game 3. The other games have featured both exciting and pedestrian draws as the play has lurched between sublime and ridiculous.
Let’s recap Games 1-7 of the match, notating moves which actually occurred during the games in bold, and hypothetical variations which did not occur but could have in italics. If you missed out, check out our preview piece before we get started!
Game 1: Gukesh 0 – 1 Ding Liren
Gukesh would start with the white pieces in Game 1 after the drawing of lots at the pre-match press conference, where Ding Liren revealed that he had only started preparing for the match three weeks ago. Chess world champions: they’re just like us!!
Gukesh opened the first game with 1.e4, and Ding responded with 1…e6, the French Defense. It was an interesting choice, as Ding had last played that opening in Game 7 of last year’s world championship match against Ian Nepomniachtchi, where he successfully navigated the opening and a complicated middlegame only to sensationally freeze at the board and fail to make his moves in time in the run-up to the time control. There was no repeat of those troubles this time, as the players gave us some mainline theory before Ding stopped to think for more than 27 minutes over his seventh move, pawn to a5:
Ding’s plan was to advance his a-pawn, position his queen behind it (she came to a5 a few moves later), and push the pawn all the way to a3. In the modern game, we call this an “AlphaZero-style” pawn, in homage to the strategy often employed by that chess engine (most famously in this game) of placing a pawn on h6 (for white) or a3 (for black), where at the expense of investing several precious moves it would hamper the opponent’s defensive efforts and ideally pay long-term dividends. Indeed, in Ding’s case his a-pawn would become a long-term thorn in white’s side, as it supported a later knight invasion to b2, and remained a constant threat to promote to a queen in the endgame if the tactics worked out right.
In turn, Gukesh showed some opening preparation in blowing the kingside open with pawn to g4! early on, and with white having advanced f- and g-pawns, Ding was always going to have to keep one eye on securing his king’s safety, restricting his options elsewhere.
We had an interesting and evenly-balanced position out of the opening, and it was in the middlegame that Ding started to turn the screws.
Queen to c4! on move 20 and Ding’s pieces started to swarm white’s half of the board. Gukesh’s attempt to whip up some counterplay with pawn to f5? was a small slip that weakened the g5-square, and so Ding’s queen to d3! offered an exchange of queens that wasn’t advantageous to white, and after queen back to e1 (necessary to guard the backward c3-pawn), black’s bishop to g5 attacked the c1-rook.
Rook c2 by white and now black steps on the accelerator with rook to c4! and white is all sorts of discombobulated:
White’s pieces are disconnected and cramped (look at that sad rook on f1), while black’s are dominant—the pawn on a3 is paying dividends as that horsey on b2 has become a monster—and after pawn to h4 and bishop to f4, black had a tight grip over the dark squares around the white king as tactical options abounded.
White’s queen had to abandon its twin defenses of the c3 and e5 pawns by moving to b1 to defend the rook (otherwise the simple tactic rook takes knight on d4 followed by queen takes rook on c2), and so Ding picked up both of those pawns and was very much in the driver’s seat.
Gukesh suddenly found himself down on the clock, after taking a lot of time to think his way through this complicated stretch he had under one minute to make his last eight moves before the players received an extra half-hour on their clocks after move 40 (he ultimately made it with one second to spare!).
Ding looked to have a decisive advantage which he just needed to convert by navigating the tactical complications, getting his king to safety, and liquidating into a two-pawns-up endgame. After successfully castling and then a trade of rooks on f1, Ding needed to show one more flash of brilliance to ensure the win:
The white bishop now attacks the knight for a second time and pins it to the queen, but the temporary pawn sacrifice pawn to e5! by Ding solved all of his problems: bishop takes pawn on e5 is met by black’s queen takes pawn on g4 with check!, which gets her out of the bishop’s firing line, regains the pawn, buys an extra move to save the knight, and weakens the enemy king even further.
Ding would pick up the a2 pawn with check a few moves later, and with his king safe in the corner of the board and his minor pieces active in the center, it was simply a won endgame where he had three extra pawns, and so Gukesh resigned on move 42.
A boil over in Game 1! After the players’ vast differences so far in 2024, a Ding win with the black pieces in Game 1 was certainly not the result foreseen by many. A perfect start for him with a superb Carlsen-style win, while for Gukesh it’s a long match, and the opening loss would force him to grow up fast in world championship matchplay terms.
Game 2: Ding Liren 1/2 – 1/2 Gukesh
Having sent a message in Game 1 of the match with a spectacular win with the black pieces, Ding Liren decided to throw a curveball with his first move in Game 2 with white, opening up with 1.e4, rather than the 1.d4 on which he built his hugely successful career. These two main opening moves set up entirely different structures (you can think of it as like facing a left-handed vs. a right-handed baseball pitcher), and although modern players can play all types of structures and Ding has played 1.e4 intermittently, he wouldn’t have the same depth of instinct as he has in the queen’s pawn opening.
Before we proceed, a moment of your time, if we may, to discuss the site of the match at the Resorts World Sentosa in Singapore. The players are playing in front of a large crowd, although in a soundproof room behind two-way glass, with the spectators peering through like they’re waiting for the players to slip up and confess to a crime. They’re sitting on office/gaming-style chairs, dressed smartly in nicely-fitting suits, alone in that room with just each other (and an arbiter), making for a very intimate setting. At the board they’ve just got their scoresheets, some water, a spare queen, and their national flags. As always in high-level tournaments, they are playing with wooden pieces in the classical Staunton design—no novelty Simpsons pieces at the world chess championship! Gukesh is a leg-shaker, while they both take time at different points to stand up and pace around the room while their opponent is thinking. The live broadcast has totally unnecessary but amusing low-angle camera shots from tiny cameras right down in the board, so the players look like giants when moving the pieces. The live feed is beamed all over the world, with the moves relayed to analysis boards commanded by broadcasters ranging from amateurs to former world champions, with the whole chess world following the action on the board. As mentioned in the preview, there’s far more interest in this match in India than in China, and one charming detail is the ChessBase India broadcast’s choice to always display Gukesh’s pieces at the bottom of the screen (putting the viewer in his shoes), rather than the conventional method of always displaying the white pieces closest to the viewer.
Back to Game 2, and we got ourselves an Italian Four Knights game, and although there were some intricate details for each side to consider in the opening (the players followed a line previously played by Ding’s second Richard Rapport, as they did in Game 1 the day before), including Ding springing a surprise by pushing his pawn to a5 on move 9 (a move the engine only finds at a lower depth), the game ultimately failed to spring into action. Both sets of bishops and the queens came off the board by move 13, and although the position was by no means dull—there are always tactical possibilities at every junction with four knights and four rooks on the board—after a little bit of shuffling the players eventually found a draw by three-fold repetition, which they did not hesitate to take, after 23 moves.
An inauspicious Game 2, with both players probably happy with the result—Ding keeping his lead, and Gukesh having steadied after dropping the first game. Onto Game 3!
Game 3: Gukesh 1 – 0 Ding Liren
After his defeat against Ding Liren’s French Defense in Game 1, Gukesh decided to open with 1.d4 in his second try with the white pieces. The players gave us a Queen’s Gambit Declined, with Gukesh opting for a structure which invited black’s light-squared into the game (7.h3 was a lesser-known move which came as somewhat of a surprise).
The players traded queens off the board by move 10, and with Gukesh having flung his g-pawn up the board for the second time in two white games, Ding played a move which was to have decisive consequences down the line:
There was nothing wrong with bishop to c2 at this stage, but in time that piece would find itself trapped and would end up causing Ding all sorts of problems. It initially looks like after Gukesh’s reply bishop to f4 that the bishop can capture the c3 pawn, but after knight to e2 the bishop will be captured, and white would regain the pawn with an advantage—hence it was stranded all alone on c2 for the time being.
But after a trade of pawns on the g-file and a shuffling of the pieces on the kingside, just as Ding looked to have equalized and perhaps even start to tilt the arm wrestle toward his side, a momentary lapse left his bishop trapped:
Rook to h5? was one attacking move too many which suffered from a crucial detail in the complications that followed, and left a path for Gukesh to surround and trap the c2-bishop. Starting with pawn to e4!, cutting off the bishop’s escape, after a trade of pawns on e4, knight to e6 (trying to bring another defender to the party), white’s rook to c1 attacking the bishop, knight to d4 looks like it’s there in time to defend. But: after bishop f2 (attacking that knight) and bishop g7 (defending it) Gukesh still had to find the one winning continuation—meaning it’s time for another installment of everyone’s favourite game: with white to play, in the following position, Guess The Move!
Here Gukesh found it: knight to e2! and white has a decisive advantage. The d4-knight is now twice-attacked, but is the only defender of the c2-bishop. If knight takes knight on e2, simply bishop recaptures on e2 and now it’s attacking the h5-rook in addition to the attack on the c2 bishop. Black has no choice but to give up a piece for a second pawn, and so there was nothing better than knight takes on b3, and finally rook takes bishop on c2 and white is winning.
Well, “winning” here is a relative term: a super-Grandmaster should be able to convert this advantage into a win with their best play, and could do it in their sleep against any lower-level player, but often in complicated positions such as this the advantage can slip if a sub-optimal continuation is chosen at a critical juncture.
That is to say, Gukesh was in the driver’s seat, but Ding should still force Gukesh to prove that he could convert the win. Gukesh looked well on the way to doing so, but Ding made a practical error in letting his clock burn down too far too early, leaving himself with only six and a half minutes to make his final 10 moves (and just over a minute for his final seven), in a phase of the game where he wanted to be posing Gukesh difficult questions. If he could force off the right pieces and liquidate into an opposite-colored bishop endgame, for example, it would be very possible that he could rescue the position and come away with a half-point.
Ultimately, though, Ding lost track of his clock and, in the following position, lost on time as he knocked over the pieces as he was attempting to make his 37th move with time expiring:
Ding! Get it together, man!!! It was a disappointing way for the game to end, even given that the position looked lost—it’s always better to make your opponent beat you, rather than doing it yourself. Losing on time just ain’t the way to go out in a world championship game. (Hat tip to The Guardian: this is thought to be the first time a player has lost on time in the world championship since Nigel Short against Garry Kasparov in 1993.)
Gukesh had leveled the match and must have been feeling much better about his chances, both in the way he turned the screws in an intricate position and having exorcised the demons of his Game 1 loss. Ding, on the other hand – well, who knows with this guy, at this stage! He’s given us easily the most entertaining world championship play in recent times, and has shown that he’s anything but predictable from one day to the next.
Game 4: Ding Liren 1/2 – 1/2 Gukesh
After the rest day, Ding Liren came armed for Game 4 with somewhat of a surprise, springing 1.Nf3 with the white pieces. Ding and Gukesh found an uncommon setup early on, with white playing a reversed-Queen’s Indian setup. (As world No. 21 Anish Giri noted on his stream, the term “Indian” in chess opening taxonomy denotes a structure where a player has only advanced their pawns one square, because in the old Indian Chess the rules did not allow for pawns to move with an initial double-hop.)
The game only threatened to spring into life a few times after the interesting theoretical battle, with Ding sidestepping one potential landmine by not playing the tempting pawn to f4? on move 16:
That wouldn’t have been pleasant for white, since after knight to c4! either the black knight is camped on a very nice outpost, or, if white kills it off with bishop takes knight, after black recaptures with the pawn white has to choose between taking that bishop with its queen but eventually losing the d2-pawn, or allowing the h7-bishop to come to d3, attacking the f1-rook and upsetting the harmony of its pieces.
As it happened, Ding saw these problems and proceeded with knight to f3, inviting a trade of knights, and most of the air had gone out of the position. The players eventually found a draw by threefold repetition with just one pair of rooks and three pawns each on the board on move 42. A bit of a ho-hum affair which saw the players finding their range as the match rolls on.
Game 5: Gukesh 1/2 – 1/2 Ding Liren
Gukesh opened up Game 5 with another try with 1.e4, and Ding Liren again gave us 1…e6, the French Defense. The players decided to give us the Exchange Variation with the queens traded off the board by move 7, perhaps the single-most derided opening in the whole tree, due to its dullness and drawishness:
Well, maybe the players would have something up their sleeves in the nuances that would follow, and sure enough, Gukesh tried to make a fight by playing pawn to g4! for the third time in three games with the white pieces. (Shakhriyar Mamedyarov was the modern pioneer of aggressively advancing the g-pawn that is usually stationed directly in front of your castled king, an unthinkable option in days gone by, after Garry Kasparov had won many memorable games in the 1980s and 1990s by pushing his h-pawn and checkmating fools on the kingside.)
After a shuffle of pieces and a few trades, the position sprung to life when Gukesh lost concentration for a moment and blundered on move 23:
If rook takes bishop on e5 and an exchange of rooks on that square, the game keeps rolling along. Instead Gukesh’s pawn takes bishop on e5? was a blunder, as after Ding’s knight to d3! the knight was attacking both the e1-rook and the white pawn now on e5.
White had to choose between either allowing that pawn to fall, or, as Gukesh did in the game, taking the knight on d3 with the bishop, allowing black to station a nicely-advanced pawn on d3, where it could be protected by its light-squared bishop. Soon after black played knight to c4, and with a trade of horses on that square, although material remained equal, black had a protected passed pawn and could push for a win. Ding is back!
However, as soon as he had taken the reins, Ding decided to pull them rather than to crack the whip, as he seemed to immediately start playing for a draw. Players often go into a game with the black pieces with no designs on winning and simply happy to take a draw, but ready to switch plans if a winning opportunity presents itself. It looked like even with a winning position Ding wasn’t in the mood for a fight, however, and starting in the following position, he inexplicably decided to give up both of his advanced pawns and all of his advantage:
After bishop takes on f3 and king f2, bishop to h5 would at least have caused Gukesh to calculate his way out of trouble, as now the bishop would cordon the king off from coming to gobble up black’s d-pawn as it prepared to march down to the first rank and promotion. Instead, bishop back to c6 led to a strangely toothless liquidation, with rook takes on c4 taking care of one pawn, and then with rook d8, rook d4, and a trade of rooks on d4 the white king could come and gobble up the passed d3-pawn a few moves later. With that, there was no more life in the position, and the players found a draw by repetition on move 40. Weird!
A very strange ending to a game in which Ding had all the advantage and no risk. Perhaps he’s still a bit spooked by losing on time in Game 3, or respects Gukesh’s calculation ability too much—again, who knows with this dude! There’s no telling what will come in the games ahead.
Game 6: Ding Liren 1/2 – 1/2 Gukesh
With Ding Liren’s third game with the white pieces, he decided to try a third opening move, finally starting with his preferred 1.d4, and after 1…Nf6 2.Bf4 we had another much-derided opening: the London System. Again, it’s known for its boring play (though its ubiquitous enough these days that players just have to deal with it), but Ding had played it for the first time in world championship history in Game 6 last year, and gone on to ensnare Ian Nepomniachtchi in a beautiful mating trap to win the game, so he might have been hoping for something similar this time around!
The players appeared to follow a line they had both prepared deeply, with Gukesh only taking his first significant think on move 16, while Ding had spent less than seven minutes total before mulling for more than 42 minutes (!) over his 20th move, queen takes on c6:
This opened the way for the game to finish in a draw by repetition, and that’s what looked like happening after black’s queen takes pawn on e5, and then the queens did a little a dance in the center of the board between the fifth and seventh ranks. However, Gukesh didn’t feel like letting Ding off the hook that easily, and so queen to h4!? came as a surprise, as it kept the game going just as everyone had started pulling up stumps. Suddenly it was all to play for:
It seemed like a decent practical decision by Gukesh, who may have suspected that as with Game 5 before Ding wasn’t interested in playing on, and so may be able to wear his older opponent down in a long game before the rest day. However, he still had to negotiate the position on the board: white looks for choice here, with its passed c-pawn, its pieces dominating the center files (with that e1-rook coming to d1 on the next move), and white’s queen in a better central position for now.
A few moves later, after Gukesh had spooked Ding into accepting an exchange of queens on his terms, the challenger had the chance to find an incredibly bold counter-attacking plan:
After white’s pawn to h4, instead of replying with pawn to e3 as he did in the game, perhaps Gukesh could have considered king to g7!, preparing a king’s march, turning le roi from fragile goods to be protected into another attacker! There’s no good way to stop the king from getting to that h-pawn, since the white king is too far away and the rooks need to keep an eye on black’s passed e-pawn, which will start rolling as soon as it gets the chance (hence rook to d7, check and picking up the a7-pawn wouldn’t immediately work). White is holding things together with the threat of its passed c-pawn, but in the spirit of this game perhaps Gukesh could have probed Ding’s defenses until he found a crack.
As it was, we didn’t get that interesting continuation, as in short order white traded its h-pawn for black’s f-pawn, with the players finding a draw by repetition after 4 hours 15 minutes of game time shortly thereafter. It was a great battle with some long calculations throughout, and heading into the rest day, the match remained up for grabs.
Game 7: Gukesh 1/2 – 1/2 Ding Liren
Game 7 saw one hell of a battle end in a dramatic draw, as Gukesh started with a little more move-1 trickery: 1.Nf3 clearly came as a surprise to Ding Liren, who had evidently been banking on 1.e4 and another theoretical battle in the French Defense.
It turned out that we instead got a theoretical battle in another line, with Gukesh’s 7.Re1 in the symmetrical Grünfeld truly one for the opening theory nerds:
There are at least 12 moves which have been played more often in this position than this one, a waiting move which allows black to capture the pawn and prepares a push of white’s e-pawn with the intention of taking over control of the central squares. Unfortunately for Ding, it appears to have been a good move as well as a novelty, and so he was immediately out of book and having to come up with a refutation over the board, and spent 28 minutes on his reply, pawn takes pawn on c4.
Gukesh forged a slightly-better position in the middlegame—unsurprisingly, given his opening novelty, with his positional play forcing a trade of minor pieces which left him with the bishop pair and one of Ding’s knights stuck way out of the action. Ding saw that he needed a way to complicate the position, and so flung his queen into the fray with the pawn-grab queen takes on a2?! which could have gone either way:
Here the threat of pawn to c3 loomed a resource for Ding in several lines that may have followed, but after rook a1, queen b3?, rook a3, and queen to b1, check, simply king g2! by white, and black’s queen looked a bit silly all alone on the first rank, with white ready to take over.
The players traded queens and a two more pairs of minor pieces as they entered the endgame, where Gukesh still had a slight edge but Ding had every chance of holding with determined play. On move 40, with only seven seconds left on his clock, disaster struck, as the world champion blundered:
King to e5? by Ding wasn’t the best: the computer tells us that knight to c8, re-routing it to the d6-square (where it would blockade the d-pawn and attack the b-pawn) was best (duh!). This slip as the clock ticked down allowed Gukesh to quickly find rook to h4!, grabbing black’s h-pawn at the expense of his passed d-pawn, creating a monster protected passed pawn on the eastern side of the board which threatened to prove decisive.
A few moves later, and Gukesh perhaps fatefully passed on the opportunity to trade his bishop for black’s knight and go into a rook endgame in which he would have a better king and that all-powerful passed pawn. It was perhaps understandable that he passed on that line, given how draw-ish rook endgames are with best play, but this appears to have been the time to prove that they can be won—another example in this match that at the highest level and under this sort of pressure it’s often the practical decisions as much as the chess ones that make the difference.
And so with one loose endgame move from Gukesh, Ding found an incredible drawing resource:
Bishop to d1?, instead of the more direct bishop to e2 (or king to e2) attacking the rook, gave Ding the tempo he needed to get back in the game, with pawn to f4!!, where after a pawn trade on f4 the black rook threatened to swing across to h3, harassing the white king with checks in complications which threatened to envelop it in a mating net, pick up the d1-bishop, or (the last-bad scenario) trap it in a perpetual check loop. What a find!
Such are the slim margins at this level that the winning chances for Gukesh had suddenly gone, and although there was still a tricky tightrope to walk for both sides, the players eventually played down to nearly bare kings, and reached a draw after an incredible 72-move game. Whoa!
And so, the match is all tied up at the halfway point! We’ve had another rollercoaster seven-game stretch, with just about everything we had hoped for: wins with both colors, fighting draws, blunders, saves, opening novelties, Gukesh starting slow and then seeming to grow into the flow of the match, Ding Liren simultaneously regaining and losing his championship form.
It’s going to be a barn-burner all the way home, and there’s no telling how this one will turn out! Check back soon for the Defector recap of the second half of the match.