Chess Strategy/Minor pieces

Differences between the minor pieces
Understanding the relationships between minor pieces is crucial to chess. It's more important than most other things, including space, pawn structure, and sometimes even material. It's also one of the most misunderstood. For example, for many years the position in Spassky--Fischer, Reykjavík 1972 would have been classified as "great advantage for White". After all, he had the two bishops. No one questioned the authority of two bishops until the modern players like Botvinnik, Tal, Smyslov, and Bronstein realized that chess is too dynamic to say things like "the bishop pair is always good".

Rules of the minor pieces
First, here are some rules of thumb regarding the minor pieces. They will not work in all situations, as discussed above, but they will provide a good general guide.

Rules of knights

 * 1. Knights need advanced support points to be effective. A support point, or outpost, is a square supported by a pawn that no enemy pawn can attack, or if it can, it cannot attack it without incurring some weakness. Remember that the knight must affect the area of the board where the play is happening, otherwise it's useless! Here are some rules governing a knight's activity depending on the rank it sits on.
 * A. Knights on the first or second rank are purely defensive and are usually on their way to a better square.
 * B. Knights on the third rank are useful defenders and are ready to jump to the fifth.
 * C. A knight on the fourth rank is equal to a bishop and can withdraw for defense or venture even further for attack.
 * D. A knight on the fifth is a superb piece, constituting a powerful attacker and is often better than a bishop.
 * E. A knight on the sixth is often cause for the opponent's resignation. It creates disharmony in the enemy camp and the Exchange is often sacrificed to get rid of it. (See Botvinnik--Flohr, Moscow 1936 for an example of such a knight)
 * F. A knight on the seventh or eighth often diminishes in value, because it controls fewer squares than a knight on the sixth.
 * 2. Knights thrive in closed positions. This is for two reasons: one, because there are many support points provided by central pawns, and two, their ability to police squares of both colors and jump over other pawns means they are relatively unaffected by the pawns.
 * 3. The best blockader of a passed pawn is a knight. This is because knights can remain active while blocking pawns, due to their ability to jump over other pieces and pawns.
 * 4. Knights are usually superior to bishops in endings with pawns on one side of the board. This is because pawns on one side of the board make light of a bishop's long-range powers, while a knight does not need these long-range powers and can use its ability to jump on squares of both colors.
 * 5. The way to play against knights is to take away their advanced support points. This means that, by taking away central squares from knights, you can turn them into useless slabs of meat.
 * 6. It's usually a good idea to post knights on opposite color of the opponent's single bishop, other things being equal. As result of doing this, the bishop can't dislodge the knight from the secured outpost.

Rules of bishops
The bad bishop is one that is blocked by its own pawns. The good bishop is one that is not blocked by its own pawns. These names are too static, however, either piece can be an active bishop, depending on the situation. Here are the laws governing bishops:
 * 1. If you have a bad bishop you should correct it in one of the following ways:
 * A. Trade it for an enemy piece of equal or greater value. This makes sense. If your bad bishop is doing nothing good for your army, you want to trade it off for a good piece of your opponent's. This means that you have not lost much, but your opponent has lost a lot!
 * B. Make it good by moving the central pawns off its color square. Your bishop will no longer be bad if the pawns are no longer there.
 * C. Make it active by getting it outside of the pawn chain. The bishop can still be bad but active, if it is outside of the pawn chain's cramping effect, it will become active.
 * 2. Bishops are strongest in open positions. The long-range power of a bishop shows itself best when no pawns block its diagonals. In this way a bishop can sweep from one end to the board to another quite easily.
 * 3. Bishops are better than knights in endings with pawns on both sides of the board. Again, the bishop shows its long-range powers best by supporting its own pawns to queen while stopping the enemy pawns at long distance.
 * 4. Endings of bishops of opposite color are very easily drawn. This is because of the bishops inability to patrol squares of both colors. The defending side can be three pawns down and still hope to draw by arranging a blockade on his strong squares!
 * 5. When your opponent has a bishop, you should place your pawns on the same color squares as the bishop. However, if you have a bishop yourself, then you should try to keep the pawns on different colored squares than your bishop, no matter if your opponent has a bishop or not. This rule was formulated by the third World Champion, Jose Capablanca. It makes sense--you should try to give your bishop activity and restrict your opponent's bishop using your pawns.

Good bishops vs. bad bishops
As said earlier, "bad" bishops cannot be pinned down to inactivity, and "good" bishops cannot be pinned down to activity. Of course, these are pretty good general rules. Alatortsev—Levenfish, Leningrad 1937 was a good example of this.

White has a bad bishop. It is blocked by a long string of pawns from b3 to f3. However, this means that all of his diagonally-attacking pieces are on light squares, leaving his dark squares weak. By contrast, Black's pawns perfectly complement his bishop—they cover the dark squares while the bishop covers the light squares. '''24...Kf7 25.Kf2 Kf6 26.Ke2 Rh5! 27.Rh1 Ke5 28.Kd3''' It would be very embarrassing to allow the Black king to penetrate on the weak dark squares. This is a direct consequence of White's bad bishop—neither the bishop nor the pawns can cover his dark squares. 28...h6 29.h3? Weakening g3. The king now looks to the kingside to penetrate. 29...Rg5 30.Rh2 Rg3 Or the rook. 31.h4 Rg8 32.Ke2 g5! Black breaks through. Now the g-file is opened, and White loses material. 33.hxg5 hxg5 34.Kf2 g4 35.Rh5+ Kd4 The Black king invades. However, White accelerates the disintegration of his own position, but there is really nothing left to do. '''36.Rd1+? Kc3 37.Rh7 gxf3 38.Bf1''' 38.Bxf3? walks into 38...Rxf3+ 39.Kxf3 Bg4+ and 40...Bxd1. '''38...Kc2! 39.Rd3 Bh3! 40.Rxf3 Rxf3+ 41.Kxf3 Bxf1 42.Rxf7 Rf8''' Up a piece, with a passed pawn on f4 and an active king, Black is winning easily. He won on move 49. Black's dominance of the dark squares allowed the king to penetrate into White's position decisively.

However, despite what this game shows, bad bishops are not always bad. Bad bishops protect their pawns. Or, as in the following game, Botvinnik—Kan, Leningrad 1939, they can be excellent outposts.

White apparently had a bad bishop after 26.e4. However, the bad bishop on d5 exerts power in both directions. By contrast, the Black bishop on a6 is "much better", but it is much less active. White has a plan now of putting pressure on b6, followed by the exchange of the base on b6. Black's forces are not well-placed enough to combat this, and thus cannot regroup in time to protect the c-pawn. 26...Bc8 27.Qa4 Bd7 28.Qa7 Be8 29.Rb1 Rd6 Black has covered the b-pawn. However, once it's exchanged, Black cannot defend the c-pawn and will lose a pawn. 30.a4 Kh7 31.a5 bxa5 32.Qxa5 Notice that White did not play to win the b-pawn, just to exchange it. Why? The result is that the rest of the chain, on c5, becomes weak. 32...Ra6 33.Qxc5 Ra2 34.Qe3! White returns the queen in time to prevent Black's threats from materializing. Black was aiming at 34...Qg5 35.g4 Qd2, when nothing can prevent mate. 34...Qa6 35.Rb8 Qa4 36.Kh2 Ra3 37.Qc5 Ra2 38.Ra8 Now Black is forced to give up even more material to prolong the game. 38...Qxa8 39.Bxa8 Rxa8 40.Qxe5 Bc6 41.Qc7 Black resigned.

Botvinnik used his mastery of bishops and chains to first win a pawn, and then a game. It turns out that the bad bishop on d5 was not so bad after all!

The bishop pair
The bishop, by itself, is still a powerful piece. However, its value is more than doubled when another bishop is added to it. Think about it this way: The main drawback of a bishop is that it can only move on half of the squares on the board. However, when two bishops are sweeping the board, nothing is safe from their malevolent eyes. The two bishops can usually outweigh other positional considerations, such as pawn structure. The following game is Botvinnik—Smyslov, Moscow 1948. 1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 d6 6.Bg5 The ever-popular Richter-Rauzer Variation. 6...e6 7.Be2 Be7 8.O-O O-O 9.N4b5 a6 10.Bxf6 gxf6 Starting here, White has a plan of wrecking Black's pawn structure. This gives him a static advantage: it will last into the endgame. However, later he becomes so lost in this plan that he forgets about dynamic considerations. 11.Nd4 Kh8 12.Kh1 Rg8 A good open file for the rook. Black has good kingside chances because of his two bishop and open g-file. 13.f4 Bd7 14.Bf3 Rc8 So here Black has two bishops, but unfortunately they are not very active at the moment (this is characteristic of the Richter-Rauzer). However, Black's position is very solid. White's typical plan here would be to gain a support point on d5 with the moves f4-f5, pressuring e6. If the pawn moves, the d5 square is suddenly available to White's knights, while the bishops remain inactive. However, White begins to play for a superior pawn structure and forgets the effect this will have on the minor pieces. '''15.Nxc6? bxc6!''' Black takes away d5 from the knight and begins to play for a central break. 16.Ne2 White intends to play 17.c4, putting the permanent kibbutz on ...d5. However, there came 16...d5!. White tried to get a better pawn structure with 17.f5?! This allows Black to open up the position even further for his bishops. 17...Qc7 18.c4 dxc4 19.Qd4 c5 20.Qxc4 Bd6 21.g3 Bb5 Now both bishops begin to activate. Soon they will dominate the entire board. 22.Qc2 exf5! Finishing the self-destruction of his own pawn formation. However, the e-file and long diagonal have opened up, and White now feels the full brunt of the bishop pair. 23.exf5 Rce8 24.Rf2 Re3 Penetrating down to the third rank. 25.Bg2 Qe7 26.Ng1 Bd3 The bishops, especially the one on d3 (remember, this is the one that was "blocked" by the c-pawn) slice into White's kingside. Two rooks, on the e-file and g-file, bear down at their targets. 27.Qd2 c4 28.Rf3 Re8 29.Rd1 Bc5 White's pieces are being eaten alive. Note that his pieces are too passive to be making use of the weak pawns on a6 and f6. 30.b3 Re1 31.bxc4 Bxc4 32.Bf1 Rxd1 33.Qxd1 Rd8 34.Qc2 Bd5 White can resign here. He is about to lose a lot of material. 35.Qc3 Bd4! Taking time to guard his pawn. '''36.Qd3 Qe3! 37.Qxe3 Bxe3 38.Bg2 Bxf3 39.Bxf3 Rd2 40.Ne2 Rxa2''', and White resigned. So here, although White had a clear structural advantage, he forgot to take into account the destructive power of two bishops. In fact, the activity of these two pieces was so great, it forced White to take up a passive, defensive position. With White tied up, Black could force the exchange of queens, win material, and then win the endgame.

Play against the bishop pair
When do the bishop pair suffer? In many cases. Closed positions, for example, block the activity of the bishops. Sometimes, if the knights are simply better-placed than the bishops, the knights will do better. Here's an early example.

This is the game Brinkman—Nimzowitsch, Denmark 1922. White has the bishop pair. However, since the position is closed, this is not an important factor. In fact, Black has all of the play in this position. He will play to create pawn blockades and conquer support points for the knights. With this in mind, Nimzowitsch played 20...b5! "Sacrifice for the blockade," he wrote. "Black gives up a pawn in order to create the possibility of exchanging the opponent's light-squared bishop." The idea is that after the light-squared bishop is exchanged, White will have no light-squared guardians, and be in danger. 21.Bxb5 Rab8 22.Be2 Nb6 23.Kd1? Better was to accept the sacrifice with 23.Bxh5 Nc4 24.Qc2 Nxa3! 25.Qd2 Nc4!, with a draw by repetition. Instead, White erroneously believes that his bishop pair will win the game for him. 23...Nc4 24.Bxc4 Rxc4 Now the strength of his other knight grows. The remaining bishop is very bad. 25.Rg5 Ng7 26.h5 Nf5 27.hxg6 fxg6 27.Rxf5 The knight was about equal to the rook, so White does not hesitate to take it off. 28...exf5 29.Bxa5 Rb3! Material is level, but White's light squares are critically weak. 30.Ke2 Qb7 31.Bb4 Qa6! and White resigns. The threat is a discovered check, and it is unstoppable.

So it was Nimzowitsch, one of the greatest chess thinkers, who formulated the plan to beat the bishop pair. Do one of these three things:
 * 1. Arrange a pawn blockade so as to limit the scope of the opponent's bishops.
 * 2. Conquer support squares for the knights so you can make your knights at least equal to the bishops.
 * 3. Exchange off one of the bishops (preferably the good one, if there is a bad one) so as to diminish the value of the remaining bishop.

Knights vs. bishops
This is usually one of the most interesting conflicts in chess. Both pieces are worth about three pawns, yet they are completely different from one another. Bishops are long-range artillery, while knights are slow hoppers. Knights can tour the whole board, but half the real estate is off-limits to a bishop. Bishops can "lose" moves in an endgame, but knights can jump over pieces. The list is endless.

When the knight is superior to the bishop
White has a spatial advantage because of his binding pawns on c4 and e4. The bishop on g2 can always free itself with an e4-e5 advance, and he threatens to fork on c7. This would lead us to the conclusion that White is better. But that would be incorrect! It is Black's move now, and he needs to think of a plan—fast.

Give up? I'll walk you through it. We realize that at least one of the knights must be eliminated. But which one? Our eyes are drawn to 16...Nxd5 and 16...Bxb5. After 16...Bxb5, the c-file has been opened up, so that factor must be taken into account. Also, it creates an imbalanced position—bishop versus knight. If the c5-knight can be driven away somehow, the pressure on c7 will become quite unbearable. Bearing this in mind, Black played '''16...Nxd5! 17.exd5''' 17.cxd5 hangs the knight, and 17.Rxd5 loses the e-pawn after 17...Bc6. But now the bishop on g2 has been blocked out of the game permanently, because there is no freeing pawn advance. 17...Bxb5 18.cxd5 So far, Black has exchanged pieces (good for the cramped side) and given up a bishop for a knight. White has a half-open c-file. True, the knight on c5 plugs it up, but it can be driven away by the b-pawn or captured by the bishop. Black must eliminate both "public enemies". '''18...a3! 19.Bd4?''' White has no hope for active play now, since he has a bad bishop and Black has a beautiful knight. Better was to sacrifice the Exchange for active play with 19.b4! Bxa1 20.Rxa1. True, Black is up the Exchange, but on the other hand, White has the bishop pair, more space, threats down the c-file, and weak dark squares to play with. So he has adequate compensation. 19...Bxd4 Public enemy #1 has been eliminated. 20.Rxd4 axb2 And next comes enemy #2. 21.Qxb2 b6! This move cements the knight on c5 (preventing any sacrifices of the Exchange for a pawn), fixes the pawn on b5 (thereby limiting the scope of the White bishop) and prevents White from getting counterplay with a possible b5-b6 break. 22.Rd2 Qf5 Black has secured a better minor piece and better pawn structure. He can now think about another plan. One thing to note is that Black does not have to proceed with speed. There is no chance that White's bishop will ever become good, and the White pawns will always be weak. Black's plan is to take control of the open e-file. 23.Re2 Nd3 24.Rxe8+ Rxe8 25.Qc2 Qe5! Seizing full control of the e-file and getting out of the pin. 26.Rd1 Nc5 27.h4 Qe2 Penetrating. 28.Qd2 Qxb5 Threatening now to bring his rook down to the second. 29.Re1 Rxe1 30.Qxe1 Qb2 Without rooks Black's advantage is highlighted. White has no counterplay, and although he controls the e-file, it is meaningless because there are no penetration points. 31.Qe8+ Kg7 32.Qc8 Qxf2 33.Qxc7 immediately fails to 33...Nxe4, winning. 31.f4 Kf8 Preventing the queen from penetrating. 32.f5 Qe5 33.Qf1 Qxf5 34.Qxf5 gxf5 35.Bf3 Ne4 The heavy pieces are gone, so there is no longer any need to keep the c-file blocked. 36.Bxh5 Nc3 37.Bf3 Nxa2, White resigned.

When looking at this game, the reader should realize that Black's minor piece advantage was permanent and lasting. Black built up a plan based on this, and the rewards were great. Also note that once Black won material, he did not start playing "lazy" moves.

When the bishop is superior to the knight
Of course, the opposite can happen. Just as often, the bishop is superior to the knight. The game above made it look as if bishops are easy creatures to contain. Nothing could be further from the truth. Just remember Botvinnik—Smyslov, and you'll correct your mistake.

Black has many weak points and an extra pawn, but most importantly White has a bishop for a knight. The long-range powers of a bishop manifest themselves most greatly in this type of position. By switching attacks between a7, c5, and e6, White will keep Black moving. Eventually something will crack: because of White's greater mobility, he can move faster than Black. 22...Nb5 23.Be3 Rc8 24.Ra6 Nd6 25.Qe5 Nf7 In desperation, Black sacrifices back a pawn to make the transition into some kind of tenable endgame. However, White's pressure does not let up. His bishop is simply too active. 26.Qxe6 Qxe6 27.Rxe6 a5 28.Bd2 Black's correct method of thinking is to exchange the pawns so that it's a more manageable two vs. two or two vs. one situation on the kingside. This follows the rule that knights are better in endgame situations with pawns on one side of the board. 28...Ra8 29.Rc6 a4 30.bxa4 Rxa4 31.Rxc5 So now White enjoys his positional advantage, plus the advantage of a pawn. 31...Nd6 32.Rc6 Nc4 33.Bg5 From g5 the bishop controls many important squares, but due to its long-range power it can still eye the queenside. 33...Kf7 34.Kf2 Ke8 39.Rc7 Nd6 36.Ke2 Rc4 37.Rxc4 Nxc4 38.Kd3 Ne5+ 39.Kd4 Nc6+ 40.Kc5 Kd7 41.Bf6! Limiting the mobility of Black's knight as much as possible. This is a favor that the knight cannot return. 41...Na7 42.Kd5 Nc8 43.Bg5 Nb6+ 44.Ke5 Nc4+ 45.Kf6 Na3 46.Kg7 Black resigns, because the kingside pawns will be swallowed up.

So in this game, White was down a pawn—down a pawn, but still managed to win! This was in part due to Black's weak pawns, but mostly because of the activity of the White bishop. Leaping across the board in a single bound, the slow-moving knight was no match for the bishop.

Knights and their support points
The last game we will see shows support points in action. In a normal Ruy Lopez, both sides fight to get their own advanced support points. The game is Kavalek—Karpov, Caracas 1970. 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 d6 8.c3 O-O 9.h3 Na5 The books say that "a knight on the rim is dim", but it plans to chase the bishop to c2, where it will be even dimmer. 10.Bc2 c5 11.d4 Qc7 This is known as the Chigorin Variation of the Ruy Lopez and is the blueprint beginning of tens of thousands of grandmaster games. Black retains a strong-point on e5 while keeping his position flexible. White has a variety of plans, but the one chosen by White in this game will be revealed later. 12.Nbd2 Note that this is not the final square for the knight: it is a stopping point. The knight usually goes to f1 and then to either e3 or g3. 12...Nc6 13.dxc5 dxc5 This idea was discovered by Fischer. White captures on c5 and leaves Black with two holes, on d5 and f5. White will play to occupy them with knights. 14.Nf1! Steering towards d5 and f5 by going to e3. 14...Be6 15.Ne3 Rad8 16.Qe2 c4! Black's move lays claim to d3. Now he plans to maneuver a knight there. He intends ...Nf6-d7-c5-d3. 17.Nf5 Here White has a knight on the fifth rank. Black can do one of three things, two of which are listed below.
 * 1. He can chase it away with g7-g6. However, this is a weakening move, and Black's kingside dark squares are weak. Black's king is now a target.
 * 2. He can take it with ...Bxe6. But, after 18.exf5 Rfe8 19.Ng5! h6 20.Ne4 Nxe4 21.Bxe4 White has activated his bishop on the long, light diagonal and enjoys a stable plus due to his bishop pair and spatial advantage.

Or, he can choose the move Karpov made, 17...Rfe8, which refuses to panic. 18.N3h4 Another exception to the "a knight on the rim is dim" rule. White cements his control over f5. 18...Kh8 19.Nxe7 Qxe7 20.Qf3 Nd7 At last Black gets to execute his own plan. 21.Nf5 Qf8 22.Be3 Black's position is clearly inferior: he has less space. Still, his next move is a blunder, though a natural move. '''22...Nc5? 23.Nxg7!''' A sharp blow. After 23...Kxg7 (23...Qxg7 hangs the knight to 24.Bxc5) 24.Bh6+! Kxh6 25.Qf6+ leads to mate. White eventually won.

This is an example of the efforts both players. Both players tried to create an outpost for their knights. It also went to show how great knights could be once they landed on their dream squares.

Playing against the knights
Playing against the knights has been seen in all of the examples on this page. However, there are certain points to keep in mind:
 * 1. Play to deny the opponent's knights support points. Kick them out of their advanced squares with pawns (pieces if possible), unless the pawn moves weaken your position.
 * 2. Open the board for your bishop(s). Once this is done, the bishop will evidently be superior due to its long-range abilities.