Game Analysis: Atlanta Class Championship, Round 3

Hello everyone and welcome as we continue to analyze the games that lead to my victory of the Expert Section (which was also the top section) of the Atlanta Class Championship. In the previous two articles, we saw myself getting White against the top two players of the section, and taking both of them down. So what do I get as an opponent for my first game of the tournament as Black? The three seed! Of course!

In this game, we are going to see Black taking full advantage of the fact that White showed zero understanding of the opening. Black achieves a winning position, and keeps it until the one dark spot in the tournament hits for me. The time control was 40 moves in 90 minutes followed by sudden death in 30 minutes with a 30 second increment. Black, for the final 12 moves of the first time control, gets into severe time trouble, and we will see the entire position turn from a win for Black to a win for White in a mere matter of 12 moves. Even on move 40, Black has a shot at equality and salvaging half the point, but after move 40, Black is busted, and we will see the game conclude with correct technique by White, showing how to execute a winning Rook endgame with passed pawns for both sides.

Without further ado, let's take a look at the game.

Atlanta Class Championship, Round 3
W: Doruk Emir (2096)
B: Patrick McCartney (2018)
King's Indian Defense, Saemisch Variation

1.c4 Nf6 2.Nc3 g6 3.e4

This is a common idea in the English Opening. White has no good way to play for an advantage without directly transposing to the King's Indian Defense. There are two common ways to do it. One of them is to transpose to the Fianchetto Variation via 3.g3 Bg7 4.Bg2 O-O 5.Nf3 d6 6.O-O and following up Black's 6th move with 7.d4 as 7.d3 gives White nothing more than equality. The other approach is what happens this game, and playing 3.e4 has one major advantage over playing the fianchetto line. In the Fianchetto line, after 3.g3 Bg7 4.Bg2 O-O 5.Nf3, Black can also play 5...d5, leading to either the Grunfeld if 6.d4, or anti-Grunfeld positions after 6.cxd5 Nxd5 7.O-O, avoiding d4, which gives Black very few problems. By taking a more classical approach and playing either the Classical, Saemisch, Four Pawns, or any other line that involves an early e4, normally reached via 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6, White can specifically avoid the Grunfeld by playing 3.e4 before 4.d4. With 3.e4, the move ...d5 is taken out of the picture. White is basically saying that he has no objection to playing against a King's Indian, but no Grunfeld for you, sir! That said, those that know me know that I'm a King's Indian player, and despise the Grunfeld, and so it's no skin off my nose, but for those of you playing White, this is worth knowing.

Now you might ask, what happens after 1.c4 Nf6 2.Nc3 d5, as many Grunfeld players know that this move order is necessary due to 3.e4, as played in the game. I would suggest a line that Ulf Andersson played, namely 3.cxd5 Nxd5 4.Nf3 g6 (normally, his games would go 1.Nf3 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5, reaching the same position) 5.e4 Nxc3 6.dxc3 Qxd1+ 7.Kxd1 and Black has literally one move that leads to a near equal position, and that is the counter-intuitive 7...f6. Other moves lead to an advantage for White, including the "typical" move seen here, 7...Bg7. Those of you interested in learning this position are encouraged to pick up a copy of "How Ulf Beats Black" by Cyrus Lakdawala. Some old books on the English Opening may also cover this line.

3...d6 4.d4 Bg7 5.f3 O-O 6.Be3 e5

So we have the basic starting position of the 6...e5 variation of the Saemisch King's Indian. Black's idea is fairly simple. He wants to take on d4 to open up the diagonal for his Bishop on g7. White has two basic ideas here available to him. The first, and most common idea, is to lock the center with 7.d5. This move gains space for White, and slams shut the diagonal, and Black ends up with a bad Bishop on g7, similar to the bad Bishop on c8 in the French Defense, another opening that I play regularly, and so bad Bishops don't bother me, but again, you have to know how to deal with them to play this line against the Saemisch. If White does this, Black will usually play 7...Nh5 with ideas of either playing 8...f5, going for a Kingside Attack and chipping away at White's center, or there is an interesting line involving a sacrifice of the Queen for two Bishops and two pawns via 8.Qd2 Qh4+ 9.g3 Nxg3 10.Qf2 (10.Bf2? Nxf1 hits the Queen) Nxf1 11.Qxh4 Nxe3 followed by 12...Nxc4.

The second option is for White to allow the trade on d4 and play 7.Nge2, intending to recapture on d4 with the Knight. This leads to more of a Maroczy Bind type of position after Black trades on d4.

In the game however, White plays a bad move.

7.Qd2? exd4 8.Bxd4 Nc6

And here inlies the problem. The move 5.f3 has the downside of weakening the dark squares around White's King. Therefore, the absolute last thing that White can afford to do is give up his dark-squared Bishop for a Knight, leaving Black's dark-squared Bishop on what is now an open diagonal unopposed. Therefore, Black is developing with the gain of tempo. Had White played 7.Nge2 and recaptured on move 8 with the Knight, then 8...Nc6, while a fine move, does not gain time like it does in the game because White should have no objection to trading Knights if Black wishes to do so, but Black is spending time doing it, first developing the Knight, and then trading it off, falling behind in development while White develops his pieces and merely reacts when needed, like re-capturing when Black initiates a trade.

9.Be3

As explained, a sad necessity, losing time.

9...a6 10.Nge2 Rb8 11.Nd4 Bd7 12.Nc2

And so now White has lost time with the Bishop, and spends 3 moves to develop his Knight to c2? Yes, the Knight does often move 3 times in the Saemisch King's Indian, but usually to get to b3 or d3, not c2.

12...Ne5 13.Bg5 h6!

To see this move, Black must know about a very common tactical shot in the King's Indian Defense. When looking for ideas, always consider the possibility of the move ...Nxe4. This can lead to a number of possibilities, especially in lines where the White Queen is on d2, because the Knight directly attacks d2, and White must react to that. This often leads to one of two possibilities. The first is an attack on d4. With the Knight moved out of the way, the Bishop directly attacks d4, and so if that leads to more pieces attacking say, a Knight, on d4 than there are defending d4, this leads to the win of a pawn.

While that is not the case here, instead we have another common idea, this being the situation any time f3 is played, and this shows another downside to the Saemisch if it isn't followed up correctly. I should take a moment to state that this article is not a knock on the Saemisch Variation against the King's Indian, but it does show what happens if White subsequently has no clue what he's doing, which was the case here. In this case, we have a fork with the Queen, giving check to the White King, and that is why this move doesn't drop a pawn. White has three options, none of which are good for White. He can admit the loss of more time, and retreat with 14.Be3 or 14.Bf4, and maybe he thinks that the pawn advance weakens Black's structure. While that may be true in some cases, it is not here, but this may be the least of the evils for White. The second is to trade the Bishop for the Knight, which we already discussed. The third is what happens in the game.

14.Bxh6?

This allows Black to force White to give up the dark-squared Bishop for a Knight, which we mentioned earlier is usually really bad for White in this line.

14...Nxe4!

The fact that this hits the Queen gives Black the tempo needed to get in the check and capture the Bishop on h6. The only move that prevents this is 15.Qf4, but then after 15...Nxc3 16.bxc3 and White's pawn structure is a train wreck.

15.Nxe4 Qh4+ 16.Qf2 Qxh6

This move is stronger than trading Queens and taking back with the Bishop. Black's King is safe, White's is not. Black wants to keep the Queens on the board.

17.Be2 b5 18.c5 d5 19.Nc3 Rfe8 20.O-O

Not 20.Nxd5?? Nd3+, where the Queen falls.

20...c6 21.Rfe1 Qf4 22.Rad1 Nc4 23.Bxc4

Which way should Black take back? 23...bxc4, 23...Qxc4, or 23...Rxe1+ followed by one of the captures on c4?

23...bxc4

While not by any stretch the move that loses the winning advantage, this is the start of heading in the wrong direction. Whether Black trades on e1 first or not, he should recapture on c4 with the Queen rather than the pawn. The reason for this has nothing to do with Black's pieces, but rather White's. Black has a winning position and has two unopposed Bishops. His advantage is a long term one, and one that should not be rushed. The Knights must be contained, and by taking with the Queen, where are the Knights going? After 23...Qxc4, the c3-Knight is cannot move forward due to the Black pawns, and the c2-Knight remains passive as attacking the Queen with 24.Na3 forces White to react to the threat of ...b4 after the Black Queen moves, and trying to come out with 24.Ne3 hangs the pawn on c5. By taking with the b-pawn, White has the a4-square to get to the outpost on b6 for the c3-Knight and can cause Black some issues. In the game, it turns out to be enough to cause Black to run low on time. By the way, in the current position, White has 53 minutes to Black's 38 to make 17 more moves before each side is awarded with an extra 30 minutes.

24.Na4 Be5 25.g3 Qf6 26.Nb6 Bh3 27.b3 cxb3 28.axb3 Bc3 29.Re2 Rbd8 30.Na4

Black has 11 minutes to make 11 moves. Black has a fairly simple way to get a winning position, but the lack of time cost me and I fail to see it.

30...Bf5?

Black achieves a winning position with 30...Rxe2!. After 31.Qxe2 Rb8, White is stuck between a rock and a hard place. If White blocks the b-file with 32.Nb6, then 32...Bf5 33.Kg2 d4 forces 34.Ne1 due to the threat of ...d3, and White's pieces are in complete disarray. Otherwise, after 32.Nxc3 Qxc3, White loses material as 33.b4 is answered by 33...a5! and otherwise, both b3 and c5 are hanging.

31.Ne1?

White can shrink Black's advantage to a minimum after 31.Rxe8+ Rxe8 32.Nxc3 Qxc3 33.Nd4 as trying to grab a pawn via 33...Qxc5 only leads to equality as after 34.Nxf5 Qxf2+ 35.Kxf2 gxf5, Black's pawn structure compensates for the pawn loss.

31...Kg7

Once again, Black should trade Rooks on e2 and follow up with ...Rb8, just like he should have the previous move.

32.Ng2

White again should take the Bishop.

32...Bd4 33.Ne3 Bxe3?

All Black needs to do is move the King to avoid allowing a check by the White Knight and White is dead. A simple move like 33...Kf8 wins for Black.

34.Rxe3 Rxe3 35.Qxe3 Qe6?

What is Black doing? Well, it turns out, Black had under a minute to get to move 40, and went with trying to simplify the position. The problem is, this solves all of White's problems as pressure is removed, and now, the best Black can hope for is an equal endgame, and the next few moves we will see even more errors by Black in time trouble, and the resulting position at time control will be completely busted for Black.

36.Qd4+ Qf6 37.Nb6 Qxd4+ 38.Rxd4 Bc8?

What on earth is this? The only trump card that Black has left is the protected passed pawn on d5. He should play 38...a5, looking to get rid of his one weakness. If White plays 39.b4, he can trade. If 39.Ra4, then 39...d4 and White doesn't have time to pawn grab. Black might still have a slight edge this way.

The move played in the game is completely useless and outright bad for Black.

39.Nc4

Had Black left the Bishop on f5 and played 38...a5, this move would not be possible because Black could take the Knight. After 39...dxc4! 40.Rxd8 cxb3 41.Rd1 b2, the b1-square is covered and White loses his Rook. Here, with the Bishop on c8, White has this possibility.

39...Rf8 40.Nd6

Last chance for Black to keep an equal position, and it's move 40, seconds on the clock. What move do you play?

40...Kf6??

Black must play 40...Be6 here, giving him time, while he can, to get the Rook active via 41...Rb8. Now Black is busted. His position will crack, leading to White gaining a pawn as the Black pieces virtually can't move, and White gains a winning Rook ending which he never gives Black a chance at this point. It took 31 moves to get Black to resign, but it's winning the whole way for White. I encourage all of those with problems in their endgame play to analyze this endgame in depth. This is not an endgame article, and so I will only make a couple of high level points, but I encourage those with endgame issues to take at least an hour to go through the last 31 moves of the game.

41.Ra4 Ke5 42.f4+

Driving the Black King back before proceeding with his own attack.

42...Kf6 43.h4 Ke7 44.Kf2 Bd7

Now, due to a subsequent tactic, Black has to give up a pawn to just be able to move, unlike at move 40 where he could trade the a-pawn for the b-pawn.

45.Rxa6 Rb8 46.Ra7 Rxb3 47.Nxf7

Tactically maintaining the pawn advantage.

47...Rb2+ 48.Ke1 Kxf7 49.Rxd7+ Ke6 50.Rd6+ Kf5 51.Rxc6 Rg2 52.Rc8 Rxg3 53.Kd2 Ke4 54.c6 Kd4 55.c7 Rd3+ 56.Ke2 Re3+ 57.Kf2 Re7 58.Kf3 Kc4 59.Kg4 d4 60.Kg5 d3 61.Kxg6

White is just fast enough in the race to take all draw possibilities away from Black.

61...Kc3 62.f5 d2 63.Rd8 Rxc7 64.f6 Rc4 65.f7! Rf4 66.h5 Rf1 67.h6 Rg1+ 68.Kh5 Rf1 69.h7 Rh1+ 70.Kg4 Rg1+ 71.Kh3 1-0


What a disgusting way to lose a chess game. Often times, it's better to simply get blown away than to have a winning advantage from the get-go just to completely botch it in time trouble. The following items can be picked up from this game:

• Time management! Black had 11 minutes to make 11 moves, and this cost him. Earlier in the game, a number of moves should have been played faster. Instead, Black was constantly looking for the perfect move. I can recall a few of them I had the candidate in mind long before the move was made, and I spent all that time either looking for pipe dream moves for Black, or else needing to feel 100% positive that there is no counter-play for White. Yes, you do need to blunder-check, but don't spend for ever doing this. Moves with unnecessary time taken include 12...Ne5 (9 minutes), 16...Qxh6 (3 minutes), 17...b5 (5 minutes), 22...Nc4 (9 minutes), 27...cxb3 (7 minutes), and 29...Rbd8 (9 minutes). Some moves do require time to be taken, like the 6 minutes I took for 21...Qf4 as this move impacts the entire idea of what Black is going to do, and I should have spent more time on move 23 than the 1 minute I took there, but of the 42 minutes taken for the 6 moves mentioned, I should easily have been able to preserve 20 minutes of that, giving me the time I needed for moves 23 and 30 thru 40.
• A word of advice - if you don't already do so, take down the time at every move. Don't try to calculate time taken. Simply write the time remaining. So on my score sheet, I have a 27 beside 26...Bh3 and a 20 beside 27...cxb3, and that's how you figure out that you spent 7 minutes. Don't try to calculate that you took 7 minutes during the game. Simply write the time left at each move, and use that information afterwards when analyzing to determine where you need to better manage your time. Look for long times spent on moves that aren't that complicated, and also look for critical decisions in the game that perhaps you didn't spend enough time on. The only move where I feel I made the latter mistake is the recapture on move 23, but there were numerous times that I spent way more time than I should have, and I only know that by taking down the time at every move. I also take down my opponent's time at each move as well. This can tell you where he spent too much time, but it also gives you the information of whether one player has a major time advantage during a critical point in the game.
• In the opening, move order matters. The way White handled the opening gave Black extra tempos because White failed to develop his King's Knight, and Black got a winning position less than 10 moves later.
• When you have the Bishop pair versus a pair of Knights, it's not about rapid fire. It's all about containing the Knights, which capturing with the Queen instead of the pawn on move 23 would have done.
• Endgame knowledge is vital. If White didn't understand his Rook endings, Black might have been able to pull a draw due to the passed d-pawn, but with best play, it wasn't enough to offset the pawn deficit.

So this loss was a major setback, especially given that I should have won the game. There were 14 players in the section. Going into the round, there were only two players with 2 points, and that was the two featured in this game. There were only two players with 1.5. The first one had a half point bye for round three, and the other one drew. Combine that with the players with one point that won their third round, and you had five players with 2 points and nobody with 2.5. Therefore, Emir was in a prime position to win, and any hope for the other 5 of us had to come via two wins. There was virtually no way to mathematically win outright with a win and a draw. So my work was cut out for the final two rounds, which we will look at in the next two articles.

Till next time, good luck in your games.