Game Analysis: Know Your Endgames!

Hello everyone. Here I'm going to be covering the first of three articles on games from the South Carolina Championship. They won't be covered in order as I felt this game from the third round perfectly illustrates how critical it is to understand and know your endgames. Usually, if I'm not in severe time trouble, I can normally execute drawn positions to a draw and won positions to a win if I have gotten the game down to an endgame. However, what we are going to see here is a game where there really isn't much to say about it prior to the endgame, but we will be seeing a number of errors made by both sides once the endgame is reached. Various endgame topics can be seen in this game alone, including things such as the King being inside versus outside the box of an opposing passed pawn, the importance of King activity and how the King can be used as an active piece in the endgame, the importance of counting moves and finding the quickest way to execute what you are trying to achieve, how to handle pawn majorities and the fact that smaller majorities are better (i.e. a 2-on-1 majority is better than a 3-on-2 majority which in turn is better than a 4-on-3 majority), and the importance of calculating to the end, unlike in middle games where multiple sources will tell you not to try to count all the way to the very end of the line as you will land in severe time trouble more often than not that way, but in an endgame, calculating to a conclusive point, where you can definitively say that the position is winning for one side or the position is drawn, can often be critical.

So without further ado, let's take a look at the feature game.


South Carolina Championship, Round 3
W: Patrick McCartney (2018)
B: Gene Nix (1855)
Italian Game

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.d3

One advantage to the Slow Italian as opposed to the old traditional Italian where White's pawn goes to d4 instead of d3 is that while it normally arises via 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.c3 Nf6 5.d3, it can also be reached if Black plays a Two Knights Defense because after 3...Nf6 4.d3, while Black does have other options, there really is nothing better than 4...Bc5, leading back to the traditional Italian Game and pretty much avoiding all of that Two Knights Defense theory. Just something to keep in mind if you are an e4 player.

4...Bc5 5.c3 O-O 6.O-O d6 7.b4 Bb6 8.a4 a6 9.a5 Ba7 10.Na3 h6 11.Re1 Re8 12.Ra2 Be6 13.h3 d5 14.exd5 Bxd5 15.Qb3 Bxc4 16.Qxc4 Qd5 17.Rae2 Rad8 18.Qxd5 Nxd5 19.Bd2 Nf6 20.Nc4 Rxd3 21.Nfxe5 Nxe5 22.Rxe5 Rxe5 23.Rxe5 Kf8 24.Kf1 Rd5 25.Re1 Rf5 26.Be3 Bxe3 27.Nxe3 Re5 28.Rd1 Ke7 29.Rd4 c5 30.Rc4 Kd6

So here we have our first real position of interest. Right now, the position is dead equal, but it is important to note all the features of the position, and which ones favor White and which ones favor Black as that is usually what will determine needs to be done here. You should ask yourself the following questions:

• Who has the better placed pieces?
• Who has the better pawns?
• Who has the better King?

Now in some ways, that can be a loaded question. For example, what makes one side's pawns better or worse? That said, let's look at the answers one by one.

First the pieces. This goes to White. His Knight is better placed than Black's and keeps the Rook from getting active. The Rook has no useful lateral or vertical move. Until he rearranges his pieces, he constantly has to look out for Nf5, forking the King and g-pawn, and Rxc5, if he retreats the Rook backwards in order to get it onto the d-file. Advancing the g-pawn for Black weakens the Knight and the pawn on f7. White, on the other hand, can easily play g4 as the Knight remains protected, it covers the f5 square that the White Knight wants to go to, and f2 is covered by his own King. White's biggest weakness is the pawns on the Queenside, which for now the Rook covers, and so White has the better placed pieces for now.

Now the pawns. This goes to Black. There is nothing special about the Kingside pawns other than the fact that Black has to keep a closer monitoring of the f-pawn and g-pawn as his King is not there to protect them, unlike White. However, White's Queenside pawns are farther advanced, and when there is no majority and no passed pawn, advanced pawns can be more of a liability in an endgame than an asset because they are easier to access and the Black King can get at the White pawns a lot easier than the White King can get at the Black pawns that are much farther back into Black's territory.

As for the Kings, that should be fairly obvious. While Black's King may not be covering his Kingside pawns, this is by no means the most important factor. The most important factor is that the King is active and centralized and is acting more like a normal piece and not hiding back behind his own pawns as the material is reduced and mate is highly unlikely at this point. It's all about getting a passed pawn and promoting it.

So now that we considered all of this, what should be White's next move?

31.Rf4?

This move is terrible. He has moved his Rook away from the c-pawn that it was guarding. It removed all pressure off the c5-pawn. It is doing nothing special on f4 as Black is under no obligation to move the Knight. And the threat of the Knight going to c4 is a cheap, one-move threat. The main reason I played this was not for the cheapo threat, but I was looking to relocate the Knight and get the Knight closer to Black's pawns, but it allows Black to be active, and with the better placed King, this favors Black. The only moves White should be considering at all are 31.g4 and 31.Ke2, and the latter is what White should have done here. He needs to get his King into the game. After 31.Ke2 cxb4 32.cxb4 Nd5 33.Rd4 (otherwise Black can force an isolated pawn in a King and pawn endgame) Kc6 34.Rxd5 Rxd5 35.Nxd5 Kxd5 36.Kd3 and despite Black being on his fourth rank, it isn't enough and the position is a dead draw with correct play.

31...Kc6 32.Nc4?

One bad move is followed by another. Relatively best was 32.g4, trying to make something out of the Kingside while Black still has trouble getting to the weak c3 pawn and a trade on b4 moves the weakness to b4 where the Rook on f4 continues to guard.

32...Re4!

Now Black is winning.

33.Rxe4 Nxe4 34.Ne5+ Kb5 35.Nxf7

Black to Move and Win

35...Nxc3?

Clearly, with the location of the King's, Knights, and pawns, this is a foot race. Black should be figuring out the fastest way to get all of White's Queenside pawns and maintain having his King and Knight in ideal spots.

The correct way to do this is via 35...cxb4!, which White is then forced to recapture with 36.cxb4, and then 36...Kxd4 and the next time that Black has time to make a free move, he will take the a5-pawn with his King. That is three moves for Black. All other ways of doing it takes four or more moves, and in some cases, the Knight ends up on c3 instead of the more idea e4. From a timing perspective, this is vital.

After 35...cxb4 36.cxb4 Kxb4 37.f4 (37.Nd8 Nc5!, stopping both Nxb7 and Ne6) 37...Kxa5 38.Ke2 b5 39.Kd3 Nc5+ 40.Kc2 b4 41.Nd6 Ka4 42.Nc4 Ne4 43.g4 a5 44.Kb2 Kb5 45.Ne3 a4 46.Nd5 Nd2 47.g5 hxg5 48.fxg5 Kc5 49.Nc7 Nc4+ 50.Kb1 b3 51.Na6+ Kd4 and Black wins. Aside from reacting to threats, like 37.Nd8, it took only three moves to scoop up the three Pawn. With the game move, it will take a minimum of four moves, and the Knight needs time to relocate to a more centralized post.

36.bxc5

The correct move, elongating Black's process to scoop up the Pawns.

36...Ne4?

Black should have taken on c5 via the King with 36...Kxc5 and a clean victory. Now, with the White pawn on c5 and a tactic, White is able to stir up trouble.

37.Nd6+!

The only move that gives White a chance, but he is reliant on one more blunder by Black. Despite being a pawn up for the moment, all other moves lose more easily as they are too slow. White must go into the King and pawn endgame and hope for one more error by Black.

37...Nxd6 38.cxd6 Kc6 39.f4

By computer standards, this is not the best move for White, but no matter what, White has to be reliant on an error by Black, and so therefore, sometimes the best computer move is the worst move over the board because it does not open the room for error. By playing 39.f4, White is hoping that Black makes what would normally look like a very natural move, but turns out is a horrible mistake.

Now Black has to find the correct move. After failing to play the best move and taking the cleanest approach to victory on move 36, one and only one more wins for Black. Can you find it?

Black to Move and Win

39...Kxd6??

This natural looking move outright loses! There is one move the wins, one move that draws, and all other moves win the game for White.

The drawing line is 39...b5, to which White should not take en passant. Instead, 40.d7 Kxd7 41.Ke2 Ke6 42.Ke3 Kd5 43.Kd3 b4 44.g4 g6 45.f5 gxf5 46.gxf5 b3 47.Kc3 Ke5 48.Kxb3 Kxf5 49.Kc4 Kf4 50.Kc5 Ke5 (Trying to go running to grab the h-pawn loses for Black as White gets his pawn to a8 long before Black gets his to h1.) 51.h4 h5 and White cannot win this as the Black King will get to f8 just in time, which is the drawing square against a Rook pawn. 52.Kb6 Kd6 53.Kxa6 Kc6 54.Ka7 Kc7 55.a6 Kc8 56.Kb6 Kb8 57.Kc6 Ka7 58.Kd6 Kxa6 59.Ke6 Kb6 60.Kf6 Kc6 61.Kg5 Kd6 62.Kxh5 Ke7 63.Kg6 Kf8 and the Black King gets there just in time to draw.

The winning move, however, is pushing the b-pawn half as far. After 39...b6!, Black wins via 40.d7 (40.axb6 Kxb6 41.Ke2 Kc6 42.Kd3 Kxd6 43.g4 Kd5 44.g5 hxg5 45.fxg5 Ke5 46.Kc4 g6 47.Kb4 Kf5 48.Ka5 Kxg5 is also winning for Black) 40...Kxd7 41.axb6 Kc6 42.b7 Kxb7 43.Ke2 Kc6 44.Kd3 Kd5 45.g4 a5 46.h4 a4 47.h5 a3 48.Kc2 Ke4 49.g5 Kxf4 50.gxh6 gxh6 51.Kb3 Kg5 52.Kxa3 Kxh5 53.Kb3 Kg4 and Black wins as the White King can't get to the drawing square, namely f1.

40.g4 Kc5

Now it's White's turn to find the win. Which pawn should White push? The f-pawn, g-pawn, or h-pawn? Be careful, only one of them actually wins!

41.f5??

This is only good enough for a draw with correct play.

The winning move is 41.g5!!. After 41...Kc6, (41...hxg5 42.fxg5 g6 43.h4 Kd6 44.Kf2 Ke6 45.Kg3 Kf7 46.Kg4 Kg7 47.h5 gxh5+ 48.Kxh5 Kh7 49.g6+ Kg7 50.Kg5 Kg8 51.Kf6 Kf8 52.g7+ Kg8 53.Kg6 b6 54.axb6 also wins for White) White wins via 42.Ke2 hxg5 43.fxg5 g6 44.h4 Kd5 45.Kf3 Ke6 46.Ke4 Kf7 47.Kf4 Ke6 48.Kf3 Kf7 49.Kg4 Kf8 50.h5 Kg7 51.h6+ Kg8 52.Kf4 Kf8 53.Ke4 Kg8 54.Kd5 Kf8 55.Ke6 Kg8 56.Kf6 Kh7 57.Kf7.

41.h4 fails for the same reason as 41.f5, simply inverting moves 41 and 43.

41...b5 42.axb6 Kxb6 43.h4

Last chance for Black. Do you advance the pawn? Or do you move the King towards the Kingside? One draws. The other loses. What's your move?

43...Kc5??

The wrong move! Black must advance the pawn in order to draw. After 43...a5!, White has nothing better than mutual promotion. After 44.g5 hxg5 45.hxg5 a4 46.f6 gxf6 47.gxf6 a3 48.f7 a2 49.f8=Q a1=Q+, it's an obvious draw. If White tries to use the trick used in the game, matters are worse. After 44.g5 hxg5, if White plays 45.f6??, then 45...gxf6 46.h5 a4 47.h6 a3 48.h7 a2 49.h8=Q a1=Q+, Black is winning with the extra two pawns. Therefore, the only other option is 44.Ke2, getting out of the check at the time of promotion. Here, Black promotes with check and one pawn rather than two via 44...a4 45.Kd2 a3 46.Kc2 Kc5 47.g5 hxg5 48.f6 gxf6 49.h5 a2 50.Kb2 a1=Q+ (This move is important in order to invoke the check when Black promotes the g-pawn.) 51.Kxa1 g4 52.h6 g3 53.h7 g2 54.h8=Q g1=Q+ and while this is not a blatant win like the case with two pawns, it is Black that has the only shot at winning.

By not advancing the pawn first, White gets to make the one necessary pawn advance to win by one move, and advancement of the passed a-pawn can be ignored as it will only get to a2.

44.g5 hxg5 45.f6!

45.hxg5?? would be a blunder as Black can then draw with 45...Kd5 and the King is inside the box. After 46.f6 gxf6 47.gxf6 and the Black King is in the box, and even if White tries to advance with 47.g6, which actually loses, the Black King is still in reach with 47...Ke6.

The move played in the game creates a passed h-pawn that is out of the reach of the Black King.

45...gxf6 46.h5! a5 47.h6 1-0

Black resigned as after 47...a4 48.h7 a3 49.h8=Q a2 50.Qxf6 and the pawn is stopped.


A lot can be learned from this endgame. Everything from how critical an active King is in an endgame, to the fact that advanced pawns that are not passed and not part of a majority can actually be viewed as a weakness, particularly if they are on the opposite side of that player's King, to the idea of sacrificing multiple pawns in order to create a passer on the outside, away from the reach of the opposing King, to the importance of piece placement when very few pieces remain. This is why studying endgames is one of the most important aspects of chess. And even when you think you know your endgames, which generally speaking, the endgame tends to be a strength of mine in most cases, you still probably haven't mastered it as you see duds like this one. Yes, White won, but it sure wasn't pretty!


That concludes this article on "Know Your Endgames". In the next couple of articles, we will be looking at two other games from the South Carolina Championships. Until then, good luck in your tournament games.