Endgame Analysis: The Art of Drawing

Hello everyone and welcome to the latest installment of the Endgame Analysis series. In this one, we are going to talk about the art of drawing, and recognizing what patterns might be out there without just aimlessly playing moves in an inferior position. Back in July 2017, I wrote an article on miracle draws (to view, click here), but in that case, it was items like perpetual check, stalemating tricks, and creating chaos in the position. Other than the stalemate, this are all mostly tricks that will happen in the middle game. What I want to talk about here is a late middle game and endgame situation where it is about recognizing resources in the position and knowing ahead of time what you must do to hold the position, keeping in mind that it will not succeed every time, but often times, lost positions can be saved by simply making the situation complicated for your opponent, and the only way to complicate matters is to actually know certain patterns. Let's look at the game, starting with the position after Black's 33rd move:

Patrick McCartney (2031) - Michael Kliber (1995)
Position After 33...Rxa2

Before we go any further, I must bring up something that I have noticed to be a major problem for some amateurs. I call it chess maturity. If you are going to succeed in these positions, you have to have some level of chess maturity, and the first step to chess maturity is recognizing and admitting that you are worse, and that a win is pretty much out of the question. This is regardless of what happened previously. In this case, White had two far superior moves at move 21 that would have been virtually winning for White, and a move at move 23 that would have kept at least a small advantage. This is neither here nor there. We are at White's 34th move, and White is clearly worse. Black has a Bishop and three Pawns for a Rook, and it's not like White has any compensation to show for it. The Knight on e1 is stuck playing a defensive role of g2, and either the Rook on f1 or the Rook on e5 is likely going to have to babysit the Knight for a while to avoid the loss of the piece. In actuality, if perfect play were made by both sides, Black would win this position. Therefore, if you are sitting down at the board playing White here, and the first thing you say is that you think you have winning chances, you need to find a different game. Chess is not for you.

However, we have to keep in mind that while Black is winning with perfect play, humans are not perfect. So what we want to talk about here is how to induce imperfection in the opponent. Sometimes, this might mean playing a riskier move, or sacrificing material to achieve certain patterns on the board. This is where the human element of understanding the endgame is better than that of a computer (outside of tablebases, but those don't exist above 7 pieces and the position we have here is 15 pieces). A computer is going to say that move X leads to a -2.09 position, move Y leads to a -2.31 position, and move Z leads to a -2.54 position, and so therefore, move X must be White's best move. The problem is, move X leads to routine play that is easy for Black while move Y or move Z adds the element of human complication, and so this is why you might find on message boards of sites like chess.com having comments from high rated players stating that computers are not great at openings or endgames, and that their real superiority is middle game play and calculation of deep tactics, which in those two areas, the computer blows away the human brain.

So with all of that said, let's take a look at the position, and see if we can figure out features of the position that White must pay very close attention to. What are some features that White must recognize in order to give Black the greatest challenge? Let's list some of them out here:

• First things first. What is the top priority for White? Stopping and eliminating the c-pawn! Now you might be asking why the c-pawn is any more important than the a-pawn. For starters, the c-pawn is farther advanced than the a-pawn, but there are other factors that will be mentioned in the other bullets. For now, we need to stop the c-pawn.
• While "general concepts" say that the player down material wants to trade pawns and the player up material wants to trade pieces, that is not always true. Typically, the side up material in these types of positions want to have at least one heavy piece on the board, often referred to as "The Conductor" of the attack, and here, Black would like to have the Rook to get behind the passed pawn, or it's second preference if that can't be achieved is to cover the 2nd and 1st ranks to tie White down and aid the c-pawn to promotion. Therefore, in addition to eliminating the c-pawn, another item on White's checklist should be to try to trade off a set of Rooks.
• While the first two bullets are immediate, you have to look long term as well. The first "long term" item that should come to mind is the situation with the opponent's Bishops and Rook Pawns. Black has one Bishop, and both Rook Pawns. A well-known drawing technique is that a wrong color Bishop and Rook Pawn against a King is a draw if the King can get in front of the Pawn. The way to determine if the Bishop is the wrong color is to compare the Bishop with the promotion square of the Pawn. If the color complex that the Bishop occupies and the color square of the promotion square are the same, you have the right color Bishop. If they are different, you have the wrong color Bishop. Looking at the current position, we see that Black has his Light-Squared Bishop. The Rook Pawn that promotes on a dark square is the a-pawn. With Black having three extra pawns compared to White, White doesn't have time to go chasing after all of them, but one thing to keep in the back of your head is that if we succeed in achieving the second bullet, the Rook trade, and we eliminate the Black Knight, then as long as we can block the a-pawn, we don't need to go chasing after it. This is why we go for the c-pawn, and try to defend the 3-on-2 on the Kingside, and not even try to capture the a-pawn. All we want to do is block it for now.
• Another thing to look at is pawn patterns and King routes. Many know that Rook Pawns are the exception to many endgame rules. However, Knight Pawns are also tricky, particularly when talking the topic of blocked pawns. Let's say you put a White Pawn on d4 and a Black Pawn on d5. The White King is on d3, and the Black King is on h8, trapped there so it can't help the d5-Pawn. How can White use the King to get at the d5-Pawn? He can go to his left via d3-c3-b4-c5-d5, or he can go to his right via d3-e3-f4-e5-d5. He has room to go around either side of the pawn. However, let's pretend these pawns are on b4 and b5 with the White King on b3 and the Black King once again blocked off on h8. Now how can White get the pawn? He only has one way! He must go to his right because there is no room on the edge of the board to get around the blocked Knight Pawns. He has to go b3-c3-d4-c5-b5. So keep this in the back of your mind about blocked Knight Pawns.
• Observe for backwards pawns and whether the pawns are on the same or opposite color of the opposing Bishop. Here we see Black having all of his Pawns on the Kingside on light squares, namely f7, g6, and h5. What if we could entice Black to trade h-pawns, and then get our pawn to g5? f6 is a dark square, and so the Bishop can't help in the backward Pawn's advancement. The Knight or the King would have to do that. Now let's say we achieve the g5 vs g6-f7 Pawn structure. The Bishop can't get to g8 or h7, and if the Black Knight is eliminated, and White gets his Knight to f6, it's his to occupy. Black can take for ever to get his King around to the g5-pawn, and since the Bishop can never disrupt the h7-square, when the Black King wraps around, we move the Knight to h7, guarding g5. This is the critical idea of why the Knight Pawn forces the King to wrap all the way around the long way. If these were f-pawns, White could come down the other side to harass the Knight. Here, the Knight has a safety net.

Ok, so what does this all mean we need to do as White. Here's the checklist:

1. Eliminate the c-pawn.
2. Trade a set of Rooks.
3. Entice the trade of h-pawns out of Black.
4. Get the g-pawn to g5 to block the Black Pawns despite the 2-on-1.
5. Eliminate the Black Knight, even at the cost of the Exchange.
6. Get the White King over to the Queenside to block the a-pawn.
7. Get the Knight to f6 to put a stranglehold on the Kingside.

Now one thing to keep in mind is that Black gets moves too, and everything that Black does could alter the checklist. That said, in this endgame that we are about to look at, Black walks right into White's plan.


34.Rf2

White starts off by attempting to trade the Rooks right off the bat.


34...Ra1

Black, of course, correctly declines the trade. Now here is where the mentioning of computer moves not being best in an endgame comes into play. White can try to play the "computer move" of 35.Kh2, but it does nothing to execute White's checklist given above. Instead, White goes for the 3rd move given by the computer, which is evaluated at -2.45 instead of -2.12 (after 35.Kh2).

35.g4

White takes the opportunity to offer the trade of Rook Pawns while it also attacks the Knight, making by-passing with 35...h4 impossible.

35...hxg4

Already Black errors. This is not a blunder, and Black is still technically winning, but he is inching closer to that drawing zone that White is looking to achieve, and is playing right into White's hand. A stronger idea is 35...Nd4, putting the Knight in a better spot. If White plays a benign move like 36.Kh2, then he can trade pawns on g4 and it would then be Black's move as White wasted time with the King move. The other option for White is to take on h5, which Black should continue to ignore. After 36.gxh5 Nf3+! 37.Rxf3 Bxf3 38.hxg6, once again, Black should not take, and play 38...f5! with a completely winning position in the R+B vs R+N endgame. The g-pawn will fall, putting Black up two pawns, and White is not in position to execute the checklist given above. Remember, we said the starting position is winning for Black, and that our plan is a manipulative way to try to reach the draw, not a forced sequence that guarantees success. With correct play from the starting position above, Black should win!

36.hxg4 Nh6

Once again, 36...Nd4 was superior.

37.Rf4

The e5-Rook has to guard the Knight, and so the Rook on f2 has to do the job of guarding the g-pawn and going after the c-pawn.

37...c3

38.Kf2

Again, one of our checklist items. Get the King over to the Queenside to stop the Rook Pawn. This can also contribute to the chase down of the c-pawn. At this point, Black is still better, but compared to say, 35...Nd4, his advantage has gone down from almost -3 to about -1.4.

38...Ra2+ 39.Ke3 Bc8 40.g5!?

This move is interesting. From a computer's perspective, and in something like correspondence chess, this move would be outright stupid and White should continue with the suffering after 40.Rc5 Bxg4, but in over the board play, tricks can be pulled. It should also be noted that while there is a 15 second increment per move here, White has under 2 minutes and Black has under 3 minutes, so neither side has time to do a full analysis on the position.

40...Ng4+ 41.Rxg4 Bxg4 42.Kd3

Opportunity Number 2 to put White away.

42...a5?!

Once again, Black doesn't completely blow it, but far stronger was 42...c2!, answering 43.Nxc2 with 43...Bf5+ and 43.Rc5 with 43...Ra5!! 44.Rc6 c1=N+!! 45.Rxc1 Rxg5 with a completely winning, 3-Pawn up Rook and Bishop vs Rook and Knight ending. Black should be able to play from here on out in his sleep.

43.Kxc3

So what have we achieved thus far? We eliminated the c-pawn, we traded h-pawns, we got the desired 2-on-1 pawn structure on the Kingside, we've eliminated the Black Knight, and we've gotten our King in range with the Black a-pawn. Black is still winning here, but there are only two items left on the checklist. Trade the set of Rooks, and get the Knight to the Kingside. Black must stop at least one of these in order to have any shot at winning.

43...Re2

So what's the first thing Black does? Offers the Rook trade! Better is 43...Bf5. That said, this still isn't a draw yet for White. White slips one more trick on Black.

44.Nd3

Taking on e2 is not best as the Knight is close to being dominated. White would rather execute the trade on e5, or if Black just sits there, White will trade next move now that his Knight is out.

44...Rxe5

And Black just complies. Again, last chance to play 44...Ra2 and keep the Rooks on. This is about to be a nightmare position for Black.

45.Nxe5 Be6

The necessary square for the Bishop to keep the White Knight out of d7 and g4, both of which lead to the Knight heading to it's target, f6.

46.Kb2 a4 47.Ka3 Bb3 48.Kb4

White should immediately play 48.Ng4 or 48.Nd7.

48...Kf8 49.Ng4 Ke7 50.Nf6

Mission Accomplished! Just look at what White has pulled off. White is down two Pawns, but yet, he has narrowed down Black's attempts at any type of win down to one thing, which we will see takes Black 41 moves to realize. Observe the position and note the following items:

• White has the desired Wrong Color Bishop and Rook Pawn scenario on the Queenside with the King on a3 that can toggle between a3, b2, and b4 (provided the Black King can't get to b2.
• On the other side, the Knight can't be touched because it's on a dark square against a Light-Squared Bishop, and the King must come all the way around to attack the g-pawn. At that point in time, the White Knight can go to h7, guarding the pawn, and for the King to harass the Knight, he has to come all the way back around via f5-e6-e7-f8-g7, at which point the Knight simply goes back to f6, pretty much at any point during the walk by the Black King as long as g5 is not attacked, so as early as the moment the King goes to e6. The h7-square is totally safe as there is no way for the White Bishop to get behind it's own pawns on f7 and g6.
• White has two ways to toggle, and so Zugzwang is impossible. If the Black King is on the Kingside, harassing the g-pawn, White can toggle the King. If the Black King comes running to the Queenside, threatening to enter at b2 if the White King moves, then White can toggle the Knight harmlessly between f6 and h7

Therefore, there is only one thing left that Black can do, and that is to advance the f-pawn while the King is attacking the g-pawn, forcing the Knight off of f6 and onto h7. We will see Black tries this 41 moves later. It should be noted that during this stretch of 41 moves, White was constantly analyzing the score sheet, looking for three-fold repetition scenarios, but with all the triangulating by Black, any position that possibly occurred three times that I could find occurred with opposing sides to move. A position that occurs 3 times, but with one side to move in 2 of them and the other side to move in the 3rd is not three-fold repetition. It must be the same position with the same player to move, both sides having the same legal options. During that time, I heard Black comment that it's not 50 yet, thinking I was counting moves, and I think that might be what drove Black to the advancement of the f-pawn later on. Not recognizing that it is literally his only try, but recognizing that he was nearing 50 moves. He probably thought at first that he could win by triangulating the King at first.

All of this said, White is not completely out of the woods yet, but Black has a LOT of work to do now to win compared to the earlier scenarios where White could resign in 3 to 5 moves, and with Black's inability to execute the simpler tasks earlier, White should expect more of the same inferior play by Black at this point.

50...Ke6 51.Kc3 Kf5 52.Nh7 Be6 53.Kb4 Bb3 54.Kc3 Ke5 55.Kb4 Kd4 56.Ka3 Kc5 57.Nf6 Kd4 58.Nh7 Ke3 59.Nf6 Kf3 60.Kb2 Ke3 61.Ka3 Bd1 62.Kb4 Kd2 63.Ka3 Kc1 64.Nh7 Bd3 65.Nf6 Kd2 66.Nh7 Ke2 67.Nf6 Kf3 68.Kb4 Kg3 69.Ka3 Kf4 70.Nh7 Ke5 71.Nf6 Ke6 72.Kb4 Kf5 73.Nh7 Ke4 74.Nf6+ Kd4 75.Ka3 Kc3 76.Nh7 Kd2 77.Nf6 Kc1 78.Nh7 Kb1 79.Nf6 Bd1 80.Nh7 Kc1 81.Nf6 Kd2 82.Nh7 Ke3 83.Nf6 Kf3 84.Kb4 Kg3 85.Ka3 Kh4 86.Nh7 Bc2 87.Kb4 Bb3 88.Ka3 Kg4 89.Kb4 Kh5 90.Ka3 f5

This is the only road now to victory for Black, and it requires very delicate play by Black, and with 2 minutes left on each clock with 15 second increment per move, and with what Black has shown thus far in this endgame, that is never going to happen.

91.gxf6 Kh6 92.f7

Both moves lose for White with correct play. Both involve surrendering the a-pawn and both involve dominating the Knight. Knowing your endgame domination tricks is critical here. After 92.Nf8, Black has to start with a couple of only moves. 92...g5 (again, Black cannot let White sacrifice the Knight for the g-pawn as otherwise, we have the Wrong Color Bishop and Rook Pawn scenario, which we already know is a draw) 93.Nd7 Kg6 (again forced as otherwise Ne5, with or without f7, depending on Black's alternative move, will draw the game) 94.Kb4 Be6 95.Nc5 (Or 95.Ne5 Kxf6 96.Nd3 Bf5 and now 97.Kc1 g4 or 97.Nf2 Ke5 as the Knight is dominated) 95...a3 96.Kxa3 Bf5 97.Nb3 Kxf6 98.Nd2 g4 99.Kb4 Ke5 100.Kc3 Be4 101.Nf1 Kf4 102.Kd2 Bb7 103.Ne3 g3 104.Ke2 Be4 and White is in Zugzwang. If he moves his Knight anywhere where it can't just be immediately captured, such as 105.Nc4 or 105.Nf1, then 105...Bd3+ wins immediately, while if 105.Kd2, then 105...Kf3 is winning. Therefore, White went for the alternative route, testing Black against a Knight that is more centralized than going to f8.

92...Bxf7 93.Nf6 Kg5 94.Nh7?

More testing is 94.Ne4+ in which Black's road to victory is 94...Kf4 95.Nc5 Ke3 96.Nd7 g5 97.Ne5 Be6 98.Kxa4 Ke4 99.Ng6 g4 100.Nh4 g3 101.Kb5 Bh3 102.Kc5 Kf4 103.Kd4 Kg4 104.Ng3 g2 105.Ne5+ Kg3, winning.

94...Kh6?

Once again, domination of the Knight wins the game for Black after 94...Kf5! 95.Kxa4 Bg8 96.Nf8 g5 97.Nd7 g4 and the pawn can't be stopped. The King is already ideally placed.

95.Nf6 Bb3

Black had the opportunity again to play 95...Kg5 and if 96.Nh7, then 96...Kf5 and if 96.Ne4+, then 96...Kf4, as shown above.

96.Ng4+ Kh5 97.Nf6+ Kh4

97...Kg5 is better, elbowing out the Knight.

98.Nh7 Kg3

What Black does here is completely mind boggling. He must use the King and Bishop to distract the Knight and guide the g-pawn to promotion, completely abandoning the a-pawn and using the fact that the White King is out in Timbuktu to his advantage. Black just proceeds to move around now like a chicken with his head chopped off until he makes a complete bone head move on move 106, throwing away any remote hope at victory.

99.Ng5 Kf2 100.Nh7 Bc2 101.Ng5 Ke3 102.Ne6 Ke4 103.Ng5+ Kd4 104.Ne6+ Ke5 105.Ng5 Kf6 106.Nf3 g5??

Remember what we said earlier? All White has to do to draw is sacrifice the Knight for the g-pawn. First chance White gets, what happens?

107.Nxg5!!

Game Over! The White King will remain standing!

107...Kxg5 108.Kb2 Bb3 109.Ka1 Kf4 110.Kb2 Ke3 111.Ka1 Kd4 112.Kb2 Kc4 113.Ka1 Kb4 114.Kb2 Bd1 115.Ka1 Ka3 116.Kb1 1/2-1/2

The White King is never abandoning the corner, and the position is a theoretical draw.


Whether it be spectators at the end of a round in a larger tournament with my game being one of the last ones done, or it be at a small event or a club where maybe the pairing sheet is taken up and they are simply waiting for that last game to end, I have had many people, director or spectator, after asking about the result of a game, proceed to say "You Drew That?????" in a shock-type tone. Not saying that happened here, but that has happened many times. I also have the reputation at the club I'm in for pulling off a lot of BS draws. All of this is not by accident. If you know the various draw techniques, such as the various patterns of stalemate cages, piece configurations that are drawn, and known book draw positions like Philidor's Draw or the Short-Side Defense (both being scenarios of R+P vs R), then you know what to aim for when there are more pieces and pawns in an inferior situation. Now trust me, I lose many games that are lost because the opponent plays the right moves. There is nothing you can do about that. But just hearing people at the club mumble to themselves that they realize who they are playing and that they have to avoid the numerous cheap shot drawing tricks is just music to my ears, and that music just gets louder each time that my opponent fails to win!

So how can you do the same thing and get that same reputation at your club that I have? Remember the following items:

• First and Foremost! RECOGNIZE AND ADMIT that your position is worse. ACKNOWLEDGE that a win is out of the question, and know well in advance that what you are playing for is a draw. If you cannot get over the fact that pipe dream scenarios of winning are just not going to happen, you will never succeed in drawing lost positions. Sure, if your opponent blindly hangs his Queen, then maybe you have a victory coming and must adjust, but barring something like that, don't go into it with the mentality that you are playing for a win because you will lose quickly that way.
• Know ALL of your basic endgames with minimal pieces. That includes Bishop and Knight vs Lone King, Two Bishops vs Lone King, Rook and Pawn vs Rook (including all draw and winning techniques, such as Lucena's Position, Philidor's Draw, the Short Side Defense, the Long Side Defense, and the Vancura Position), Wrong Color Bishop and Rook Pawn, etc. Knowing these ahead of time can be a major aid to knowing what you need to accomplish when there are more pieces and pawns on the board. You might be wondering why I mention winning scenarios like Bishop and Knight or Lucena's position. You need to know these so that you know how to impose the most difficulty on your opponent in the Bishop and Knight ending just in case he doesn't know it, and so that means don't voluntarily walk into the corner that is the color of the opposing Bishop, and knowing what positions to avoid in say, Rook and Pawn vs Rook endings. If you are too lazy to study endgames, you will never get better.
• Visualize these scenarios based on the material that is on the board. In the game above, White recognized the best shot he had and achieved it. This doesn't mean this will always work as White was already lost in the first place, but the more problems you create for your opponent, the more likely it is that he will mess up.
• Know when to resign and when not to. A position like the game above, there were ways for Black to go wrong, and he did! However, something like King and Queen vs Lone King, or King, Queen, Rook, Knight, and Five Pawns vs King, Bishop, and 3 Pawns, give it up! There are no tricks. You are lost. Resign! There is a difference between imposing legitimate problems to your opponent, and simply wasting your opponent's time.
• Study a book that has lots of problems on endgames. For lower rated players, that could be finding a general problems book that includes at least a full chapter on endgame positions. For more advanced players, such as those over 1900, I would suggest going through Jacob Aagaard's book "Grandmaster Preparation: Endgame Play". I have read four of his books in that series thus far (Calculation, Positional Play, Strategic Play, and Endgame Play), and out of those four, Endgame Play was by far the most beneficial.

I will conclude this article by mentioning that those that want another example of the Wrong Color Bishop and Rook Pawn trick is encouraged to go through my Round 5 game from The Potomac Open back in the summer of 2018. You can get to it by clicking here.

Also, those of you that are interested in the other endgame articles I've written, they can be found here by clicking the topics below:
Corresponding Squares
Opposite Color Bishops
One Pawn Up Rook Ending
Bishop vs Pawn and Same Color Bishop Endings


Til next time, good luck in all of your games.

Karthik's corner - The swinging pendulum

The French Connection - Volume 20

Hello everyone and welcome to the twentieth edition of The French Connection. Today, we are going to cover a number of things that we have not seen in a while (or at all yet) in the series. First off, we haven't had coverage of a GM game since Volume 10. We haven't seen a game that White has won since Volume 11. Lastly, we haven't seen a game with the Rubinstein Variation yet in the entire series. Well, this time, we'll be seeing all three! Before we get to the game, a little background on the player playing White in this game.

Rashid Nezhmetdinov (12/15/1912 - 06/03/1974) is from the former USSR, and spent the majority of his career within the circuit, rarely playing in international events, and of those that he did, they were predominantly minor events, the lone exception being Bucharest 1954. Therefore, he was not very well known by areas of the Western world, such as the United States, but those local to the area in the USSR all knew him as a very fierce attacker, though unlike many amateurs, including a few at our club itself, that show a complete lack of patience and want a bloody game every time, Nezhmetdinov was also able to grind out late middlegame and endgame positions that required positional technique, and that is what we will be seeing here. His opening repertoire consisted mostly of playing 1.e4 as White, and playing 1...e5 in response to 1.e4 along with the King's Indian Defense in response to 1.d4. Every GM has a most notable game. For instance, with Fischer, it was his game as Black in a Grunfeld against Byrne in New York in 1956. For Kasparov, it would be his game as White in a Pirc against Topalov at Wijk aan Zee in 1999. Well, for Nezhmetdinov, it was most certain his game as Black in a King's Indian Defense (played through an Old Indian move order) against Polugaevsky in Sochi in 1958, and that game can be viewed here. The most notable moment in the game is the Queen sacrifice on move 24 and how the White King is hunted down similar to Fischer's game from 1956, though the follow-up is not quite as long as Fischer's was. Those of you with a bloodbath style of play, I recommend looking at his games.

There's another thing that Nezhmetdinov was known for. Novelties. They may not have always been the best move on the board, but it was enough to trick his opponents, and unlike many amateurs, he knew exactly when to execute. If his opponent plays the best moves, it was not a complete waste of time for Nezhmetdinov, and his position was still playable, which is one key factor to successfully playing novelties and traps. Don't rely on them to always work, and if your opponent doesn't fall for it, you have to make sure that it doesn't impose self-inflicting damage to your own position. A prime example of this was his game against Stalberg in Bucharest 1954. It was also a French, in this case a McCutcheon, where Black played a slightly inferior 12th move (12...Qc7 instead of 12...Bd7 or 12...Qa5), and Nezhmetdinov had expected this, and played the slightly odd 13.dxc5. The position at the end was still drawn with best play at move 43, but Nezhmetdinov had tried a tricky waiting move at move 43 and Black failed to defend correct and Nezhmetdinov once again won. That game can be seen here.

But today we will be looking at one of his games that required good endgame technique, and also features one of his novelties in the opening. With that, let's take a look at the feature game.


W: Nezhmetdinov
B: Sedov
RSFSR Championship, Kuybyshev 1947

1.e4 e6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Nd7 5.Nf3 Be7 6.Bd3 Ngf6

Now the normal moves here would be 7.Nxf6, 7.Ng3, or 7.Qe2. However, White plays an unusual move here that violates "general principles", but I think we have seen enough times already that you can almost throw general principles right out the window in the French Defense as the opening is a completely different beast compared to just about any other opening in all of chess, and that even goes for tame variations like the Rubinstein.

7.Ne5!?

White moves his Knight twice in the opening in an unprovoked situation (unlike say, the Bishop in the Ruy Lopez when it is actually attacked by Black's a-pawn), which is violation number one if you are following general opening principles, and with White having a space advantage, he is inviting Black to trade down to an endgame, which is violation number two, trading down pieces when you have a space advantage to alleviate the cramp for Black. All of that said, this may very well be an exception to the rule as Black is probably best off just ignoring White's offer and castling here. That said, Black takes up White's offer, and once Black decides to take the Knight, the rest of the sequence is a fairly forcing sequence.

7...Nxe5 8.dxe5 Nxe4 9.Bxe4 Qxd1+ 10.Kxd1 Bc5 11.Ke2 f5 12.Bf3

So now, we've traded a bunch of pieces including the Queens, the position is symmetrical, the material count is equal, and White should pack it in, offer a draw to Black and call it a day, right?

Uhm, no! As mentioned at move 7, Black should probably have ignored the offer by White and castle instead because while this position may look dull with the pawn symmetry, this is not equal, despite the fact that computers will say it is with the move Black plays next in this game, which is actually not best. Each side has one developed piece. However, the equality ends there. White undeveloped Bishop has far more scope than Black's. Black's Bishop can only go to d7, which impedes the open file on the board for Black's Rooks, which means the Bishop needs to move again. That's one extra move for White. Secondly, Black's King is still on the back rank. To connect the Rooks will take an extra move by Black, either lifting his King to the 7th rank, or else castling, but it doesn't make much sense to castle with the reduced material, but either way, there's a second extra move the Black needs to make, so it's like as if White has a two move advantage in the position. This two move advantage is not astronomical given the reduced material, and White only has a slight advantage, but the point is that the position is not equal, and Black can't be playing like it's equal. With that said, Black's next move does not help his cause.

12...Bd4

Black's top priority should be catching up in development. This is best done via 12...Bd7. Everyone talks about Queens capturing b-pawns as often being a bad idea. The same can be the case with reduced material as long as your opponent has at least one Rook on the board, and here, 13.Bxb7 would be bad for White after 13...Bb5+ 14.Ke1 Rb8 15.Bf3 Bd4 with a better position for Black and he will regain his pawn. Black is also ok after 14.Kf3. At first glance this appears to lose for White because 14...Rb8 traps the Bishop, but White has tactics to get out of it via 15.a4! Bd7 16.Ba6 Be4 17.Bb5 Bxb5 (17...Bxe5? 18.Re1!) 18.axb5 Rxb5 19.Rd1 Bxe5 20.Rxa7 Kf7 21.Rd7+ Kf6.

13.Bf4 Bd7

And now it's Black's turn to realize that 13...Bxb2 is a bad move because after 14.Rab1 Bc3 15.Bxb7 Bxb7 16.Rxb7 Ba5 17.Be3 Bb6 18.Bxb6, neither recapture by Black is good. 18...axb6 19.Rxc7 leaves White a pawn up as 19...Rxa2?? is answered by 20.Rc8+, and 18...cxb6 19.Rd1 gives White a winning advantage. For example, 19...Rc8 is answered by 20.c4 and again, Black can't capture on c4 due to the skewer on the back rank, and otherwise, White's Rooks are coming in.

14.c3 Bc5 15.b4

Now we see the problem with Black's 12th move. White gains time on the Bishop, and with no b-pawn hanging on the second rank, White is ready to take on b7.

15...Be7

15...Bb5+ just transposes back to the game after White's 17th move.

16.Bxb7 Bb5+ 17.Ke1

White still has the advantage with this move, but it may have been even better to play 17.Kf3, again due to the a4-trick. If 17...Rab8, then 18.a4! If not for that move, White's Bishop would be trapped. White probably either missed this idea, or else maybe thought the indirect Bishop trade didn't favor him. That said, the move played in the game is not bad by any means, just the disconnecting of the Rooks can be a bit of a nuisance for White.

17...Rd8

Now, with the b-pawn advanced to b4 and the f3-square available for the White Bishop, there is no point in attacking the Bishop with 17...Rb8 as all it would do is drive the White Bishop to a better square, and then there is nothing better for Black than to relocate the Rook to the d-file anyway, and so the move played in the game makes more sense.

18.a4 Bc4 19.Be3 a6 20.f4 Rd3

Note: Black has castling rights, White does not

21.Kf2!

This is stronger than trying to hold the c-pawn, which turns out is not really a threat. 21.Bd4 is answered by 21...Bxb4 and even worse is 21.Bd2 as after 21...Kf7, it's actually Black that has the advantage with ...Rhd8 coming.

21...Kf7

If 21...Rxc3, then 22.Rhc1 Rxc1 23.Rxc1 followed by 24.Rxc7 continues to leave White a pawn up, and here with material reduced even further, making it be even closer to completely winning for White.

22.Bf3 Rhd8 23.Rhc1 c5

With no real way to improve on the position and still being down a pawn, Black goes into full throttle desperation mode, and sets up a nasty trap for White.

24.b5!

The trap was 24.bxc5?? Rxe3! and Black wins a piece as 25.Kxe3 Bxc5 is mate.

24...axb5 25.axb5 R3d7

25...Bxb5 can be answered by 26.Be2 Rd2 27.Bxd2 Rxd2 28.Re1 and after either 29.Ke3 or 29.Kf1 on the next move, Black does not get enough compensation for the exchange.

26.Be2 Bxe2 27.Kxe2 c4 28.Ra4 Rc8 29.b6 Bc5 30.Ra7 Rxa7 31.bxa7 1-0

Black resigned because after 31...Bxe3 32.Kxe3 Ra8 33.Ra1, the c-pawn is about to fall and White wins.


A number of things can be picked up from this game:

• Opening priciples don't always apply in the French Defense. This is why it is encouraged by many higher rated players on chess forums, such as chess.com, for players to first learn the Ruy Lopez and Queen's Gambit, because they are the openings that pretty much follow priciples to the letter. After you learn the basics of opening play, and are ready to expand the repertoire, the French is an excellent place to start because of the limited number of variations to learn unlike other openings such as the Sicilian.
• When your opponent tries to provoke you into doing something, there is likely a reason behind it. Before you automatically bite, calculate and figure out the specifics of the situation. In the game we looked at, White provoked Black into a trade down, which under normal circumstances, Black should do because of his lack of space, but in this case, we saw that the resulting position after the forcing sequence saw Black with the more difficult task of getting the undeveloped Bishop out and that his King was still on the back rank, allowing White to connect Rooks faster, and so here, it would have been best for Black to just castle on move 7.
• When your opponent poses a threat, or what appears to be a threat, always try to see if you can ignore it first. For example, the threat on b7 on move 12 should have been ignored by Black and he should have played 12...Bd7 since 13.Bxb7 lead to at minimum an ok position for Black. In essence, Black's 12th move is what put him behind the 8-ball. The b2-pawn couldn't be taken by Black on move 13 due to tactics and infiltration by White. The same can be said about taking the c-pawn on Blacks 21st move. Not only does the player that can take these Pawns have to realize that they are poisonous, but the player who owns the Pawn has to also realize that their opponent can't take them. If they don't realize this, they will spend useless time defending something that doesn't need to be defended. This requires one very important skill, and that is the ability to calculate. If calculation is a weakness of yours, I highly recommend reading "The Inner Game of Chess" by Andrew Soltis.

Well, that concludes this edition of The French Connection. Good luck in all of your French Games, Black or White.

The French Connection - Volume 19

Hello and welcome to the nineteenth edition of The French Connection. In this article, we are going to be talking about the move order in the Winawer variation, specifically the differences between two moves at move 4, and I will be showing a game that illustrates the major difference of one line in the notes, and we will see White trying a novelty of his own on move 6, but in the end, Black ends up prevailing.

One thing to keep in mind about the game is that it is a Correspondence event, and so things like playing sidelines that your opponent doesn't know or trying to push your opponent out of time are both non-factors. With that said, let's take a look at the game.


2017 Electronic Knights Semifinals (Correspondence)
W: Cameron Leslie (1953)
B: Patrick McCartney (1958)


1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5

This is the starting position of the main line of the Winawer. White has other options at move 4, but we won't be talking about those here in this article. Here, Black has two main moves, and both have their pros and cons. These moves are 4...c5 and 4...Ne7. In some cases, these moves can lead to the exact same thing. For example, the Poisoned Pawn variation can be reached via 4...c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Qg4 Qc7 or 4...Ne7 5.a3 Bxc3+ 6.bxc3 c5 7.Qg4 Qc7. The same can be said about any other line that can arise from the position after Black's 6th move, such as the 7...O-O line, the aggressive 7.h4, and the positional lines, 7.a4 and 7.Nf3.

However, there are deviations from the main position after Black's 6th move, and which move Black plays on move 4 decides which options each side has.

Let's start with 4...c5. This is the more common move of the two. Here, besides 5.a3, White has three legitimate options. They are 5.Qg4, 5.Nf3, and 5.Bd2. Let's go through each of them. We will look at each, and then look at Black's main deviation after 5.a3.

A) After 5.Qg4, Black's best move is 5...Ne7!

Now 6.Qxg7 is not good because Black still maintains the Dark-Squared Bishop. After 6...Rg8 7.Qh6 (Or 7.Qxh7 cxd4 8.a3 Qa5 9.axb4 Qxa1 10.Nce2 Nbc6 and White lacks compensation) cxd4 8.a3 Ba5 9.b4 Bc7 10.Nb5 a6 is better for Black. Therefore, White needs to play 6.dxc5, and after 6...Nbc6 7.Bd2 (7.Qxg7 is still bad for the same reason except Black advances the d-pawn instead of a capture) Ng6 8.Nf3 Bxc5 9.Bd3 Qc7 10.O-O-O a6 11.h4 Bd7 12.Bxg6 hxg6 13.Qf4 O-O-O and Black has completely equalized because 14.Qxf7 is not good due to 14...Nxe5 15.Nxe5 Qxe5. Note that 12.h5 Ngxe5 13.Nxe5 Nxe5 14.Qxg7 O-O-O 15.Bxh7 Nc4 is also good for Black.

B) After 5.Nf3, Black can play 5...Ne7, which will likely transpose to the main line with 7.Nf3, but he can also play 5...cxd4 when 6.Qxd4 Nc6 7.Qg4 Nge7 8.Qxg7 Ng6 with ...Be7 to follow and Black is fine.

C) The more strategic approach with 5.Bd2 is far more popular:

The idea here is that White wants to play 6.Nb5. This avoids the wrecked pawn structure, it leads to a trade of Dark-Squared Bishops (which is White's bad Bishop) instead of Bishop for Knight and wrecked pawns, and White sees that d6 is weakened by the move 4...c5. Therefore, Black must develop the King's Knight, either with 5...Ne7 or 5...Nh6 so that 6.Nb5 can be answered by 6...Bxd2+ 7.Qxd2 O-O, avoiding a check by the Knight on d6 and forcing the Black King to remain in the center. The downside to this is that White is moving his already developed Knight, and the line as a whole is very slow for White. A positional battle ensues, but the lack of speed in White's development allows Black to equalize.

D) This now leads to 5.a3, and after 5...Bxc3+ 6.bxc3, we have the following position:

Now 6...Ne7 takes us back to the main line. However, Black has one other alternative. He can play 6...Qa5, known as the Portisch-Hook Variation. The idea behind it is that White must guard c3, and after 7.Bd2, Black plans to play 7...Qa4. This eyes the c2-pawn instead of the c3-pawn, and prevents White from advancing a4 himself, which is a common idea in the main line as White can get get the Dark-Squared Bishop active via Ba3, and let's not forget that this Bishop is uncontested because Black gave up his Dark-Squared Bishop for the Knight. After 7...Qa4, White has two main ideas. He can play 8.Qb1, with threats of winning the Queen via Bb5+, against which Black plays 8...c4 and will usually castle Queenside and attempt to break through on the Kingside. The other main option is 8.Qg4, forcing Black into a decision similar to that of the MacCutchen Variation. Do you surrender castling rights with 8...Kf8? Or do you weaken the dark squares on the Kingside with 8...g6? Both are options and have their own theory. We saw the 8...Kf8 option in The French Connection - Volume 18.


So now this leads us to 4...Ne7. We already mentioned how the main line can arise from this move. Let's take a look at what the differences are between this move and 4...c5.

A) First of all, if White tries to play 5.Qg4, Black is not forced to transpose via 5...c5, and can instead play 5...Nf5. After 6.Bd3 h5 7.Qf4 Qh4 8.Bxf5 Qxf4 9.Bxf4 Bxc3+ 10.bxc3 exf5 11.h4 Be6, the position is equal.

B) After 5.Nf3, the game will almost always transpose to the main line with 7.Nf3 as Black doesn't have the extra option that he had after 4...c5 of taking on d4.

C) After 5.Bd2, Black can take advantage of the fact that he has not advanced the c-pawn yet, and this is what we will be seeing in the game.

D) After 5.a3, the main difference is that Black lacks the option of playing the Portisch-Hook Variation, and he also lacks the option of playing another well-known side line and that is 6...Qc7, intending to answer 7.Qg4 with 7...f5, guarding g7 with the Queen rather than having to castle into it or play 7...Kf8 to guard it, or go in the wild mess in the Poisoned Pawn and allowing White to take on g7.


So we can conclude that 4...Ne7 makes 5.Qg4 harmless, and the game will show the major pluses for Black in the 5.Bd2 variation. The only downside to this move for Black is that it removes the side options against 5.Nf3 and 5.a3. If Black fully intends to play the main line (4...c5 5.a3 Bxc3+ 6.bxc3 Ne7), this may actually be the better way to do it as it makes two White sidelines inferior, and the other main sideline will merely transpose.

4...Ne7 5.Bd2 b6!

Here is the main advantage to 4...Ne7 against the 5.Bd2 variation. Black has no weakened d6, and so there is no real threat of Nc3-b5-d6. Black's plan is simple here. He intends to take the Knight on c3, to avoid any Nb5, and then play ...Ba6, forcing a trade of Light-Squared Bishops, the typical problem piece for Black in the French Defense.

It is well known that White must act aggressively on the Kingside in the Winawer and playing something like 6.Nf3 Bxc3 7.Bxc3 Ba6 8.Bxa6 Nxa6 9.Qe2 Qc8 10.O-O c5 11.Rfd1 Nb8 12.a4 Nbc6 13.dxc5 bxc5 14.Qb5 c4 15.Qc5 O-O simply favors Black.

The main line runs 6.Qg4 O-O 7.O-O-O Bxc3 8.Bxc3 Ba6 9.Bxa6 Nxa6 10.h4 c5 and Black has equalized. In the game, White tries to play a novelty, avoiding the trade of Light-Squared Bishops.

6.Nce2

The problem I see with this move is that it plugs up White's development.

6...Bxd2+ 7.Qxd2 Ba6

Black goes here anyway with the idea of trading Bishops the moment White moves the e2-Knight.

8.h4 c5 9.h5 h6

With Black's Dark-Squared Bishop gone, this move is critical, avoiding 10.h6 by White and severely weakening the dark squares around the Black King.

10.f4 Nbc6 11.Nf3 Nf5 12.c3

If White could have his way, he would like to advance g4, driving the Knight back, and then break open the Kingside with an eventual f5 or g5. So the decision for Black is critical. First thing to recognize is that White's advancement of the f- and h-pawns have lead to some weaknesses in the White position, namely the squares e4, g3, and g4. The other question Black must ask is how weak his light squares will be if he allows White's Bishop to be uncontested. Any pin on the Knight on c6 can be answered by the Rook or Queen guarding the Knight, and otherwise, there aren't any real weaknesses for Black in the light squares. Black's issues in the French Winawer usually lie in the dark squares around the Black King. Here, with White's dark-squared Bishop gone, that's not an issue either. Therefore, Black proceeds to eliminate the Knight on e2 that is covering g3, which is Black's main route to e4 for his Knight, and e4 is a much more stable square for the Knight than f5, again because of an upcoming g4 by White if Black does nothing.

12...Bxe2 13.Bxe2 Ng3 14.Bb5

White takes a very risky approach, attempting to keep pieces on the board. Better might have been to admit that he has no advantage and play 14.Rh3 Nxe2 15.Qxe2 cxd4 16.Nxd4 Nxd4 17.cxd4 O-O 18.O-O-O Rc8+ with an equal position.

14...O-O

Black uses tactics to get his King out of the center. The fact that he is hitting the Rook on h1 and that moving the Knight to e4 hits the Queen is important, otherwise Black would lose after move played in the game.

15.Rh3

15.Bxc6 can be answered by 15...Rc8 as there is no way to avoid giving the piece back. Any harassment of the Rook, such as 16.Bb7, Black can simply hit the Bishop with a move like 16...Rc7. If the Bishop flees, the Rook on h1 hangs. Also, if the Knight is attacked by the Rook, then Black plays ...Ne4 and the Queen cannot guard the Bishop.

15...Ne4 16.Qe2 Rc8 17.O-O-O cxd4

Obviously it makes far more sense for Black to open the c-file rather than close it with 17...c4.

18.Bxc6

Possibly better is 18.Nxd4 Nxd4 19.Rxd4, after which Black should reply with 19...f6, opening up the f-file for his other Rook. Black would have the initiative, but with proper defense, White can probably maintain the balance with careful play.

18...Rxc6 19.Nxd4 Rc4 20.Re3 Qc7

21.Rxe4?

Yes, the Black Knight is strong and very annoying for White, but for now, it can be worked around. White must defend via a counter-attack, and should go after the Black King. After 21.g4! Rb8 (After 21...b5, White can play 22.Rxe4 and do as he did in the game as now he gets an extra pawn, which makes a big difference here) 22.g5 hxg5 23.h6! gxf4 and only now should White sacrifice the exchange and after 24.Rxe4! dxe4 25.Nb5 Qc5 26.h7+! Kh8 (White is winning after 26...Kxh7?? 27.Qh5 Kg8 28.Rh1 Kf8 29.Qh8+ Ke7 30.Qxb8 Qe3+ 31.Kb1 Qd3+ 32.Ka1 Qd7 33.Qh8 Qxb5 34.Rd1 Qe8 35.Qh4+ f6 36.exf6+ gxf6 37.Qh7+) 27.Nd6 and now Black is forced to take the draw via 27...Rxc3+ 28.bxc3 Qxc3+ 29.Kb1 (29.Qc2 Qa1+ 30.Kd2 Qd4+ 31.Kc1 is a perpetual) Qb4+ and now 30.Ka1 Qc3+, 30.Kc1 Qa3+, and 30.Kc2 f3 31.Nxf7+ Kxh7 32.Qh2+ Kg6 33.Nd6 Qa4+ all lead to perpetual check while 30.Qb2? Qxb2+ 31.Kxb2 e3 32.Rh1 Rf8 33.Kc2 f6 34.Kd3 fxe5 35.Ke4 b5! is better for Black since 36.Nxb5 is answered by 36...Rf7 and in the long run, the h-pawn is dead.

So we see that White has to play accurately to hold on to half of the point, but it is better than immediately sacrificing the exchange.

21...dxe4 22.Nb5 Qc6 23.Nd6 Rc5 24.Nxe4 Rd5 25.Re1 b5 26.g4

Now the attack is too slow.

26...b4

27.c4

27.g5 doesn't work now. After 27...hxg5 28.h6, Black can simply respond with 28...bxc3 and his attack is faster.

27...Rfd8 28.g5 Rd4 29.b3 Kf8

With the Black Rook no longer on f8, Black can escape the g-file rather than take on g5 and not give White the option of trying to advance the h-pawn to tear open the position.

30.Kb1 a5 31.gxh6 gxh6 32.Nd6 a4 33.Qe3

33...R8xd6!

After 33...Qc5 34.bxa4 Rd1+ 35.Kc2 Qxe3 36.Rxe3 Rf1, Black has no advantage. Also note that taking with the other Rook is wrong because Black wants his battery on the d-file to be lead with the Rook, not the Queen.

34.exd6 Qxd6 35.f5

Trying to pry open the Black King and find a perpetual, but it isn't there. Black can safely take the pawn.

35...exf5 36.Qe8+ Kg7 37.bxa4

After 37.Rg1+, both 37...Rg4 and 37...Kf6 38.Qh8+ Ke7 39.Re1+ Re4 are winning for Black.

37...Rxc4 38.Qe2

Black must now keep two things in mind. The first is not to let his King get mated. The second is that he must keep at least one set of pieces on the board. If all the heavy pieces are traded off, the White a-pawn is out of reach.

38...Qd4

Pretty much blocking all checks either directly, as in the a1-h8 diagonal, or indirectly, as in the g-file since the Black Queen covers g1, and 39.Qg2+ can be answered by 39...Kf6 and White has no checks. Black is willing to trade Queens, but the Rooks must remain on the board.

39.Qb2 Rc3

Black could probably still win by immediately trading Queens, but Black activates his Rook to the maximum before trading.

40.Qe2

Or 40.Rc1 Qd3+ 41.Ka1 Qe3 intending 42...f4 is also winning for Black while 40.Qg2+ Kh7 is nothing for White.

40...Qd3+ 0-1

White Resigned because after 41.Qxd3 Rxd3, his position is hopeless. The King is cut off from the third rank and the Black Rook can get behind the a-pawn by going at any point to the protected a3 square, and the King is going to go to f6 where it is save and pretty much makes it virtually impossible for White to win a pawn on the Kingside.


So we looked at the pros and cons of 4...c5 versus 4...Ne7, and the game showed the major advantages for Black in the latter line against 5.Bd2, and shows that 5.Bd2 is less effective against 4...Ne7 than 4...c5. This does not make 4...Ne7 better than 4...c5, and both are fully playable, but this article illustrates how various move orders as early as moves 4 to 6 can lead to vastly different options for both sides, and if you are going to play the Winawer from either side, you need to know these nuances and figure out for yourself which move order works best for you if you are Black, and if you are White, you have to understand the differences between 4...c5 and 4...Ne7 because either one can be played against you.


This concludes this edition of The French Connection. Til next time, good luck in all of your French games, Black or White.

Karthik's corner - Puzzle Corner

King and pawn endgames are usually considered the most basic end games. King and pawn endgames are statistically most often occurring type of endgames in tournament play. Though these look simple, there are so many hidden ideas that even experienced grandmasters miss out on. This position is from an Averbakh study.

White to play. Can white win? Provide your analysis in the comments.

Note: Board orientation. White pawns move from bottom to top.

Karthik's corner - Playing for initiative - Does it always help?