White can choose to play the Saemisch system of the King’s Indian by playing 5.f3. The Saemisch is a solid and flexible system for White. Often he can castle on either side of the board. He can play purely positional or launch a sharp kingside attack at the right time by advancing the g-pawn to g4 and the h-pawn to h5 and then opening the h-file.
———————————————————
After 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O, Black’s normal means of counter play are similar to most King’s Indians. He will choose to challenge White’s center with …c7-c5 or …e7-e5 and try to open a file in the center. We will now look at a side variation of the Saemisch that promises good chances for an advantage and illustrates many of the common features well.

6.d5?! may transpose to other main lines later on, but is not most accurate. Try to develop a new piece now and wait to see how Black will try to challenge your center before you make such a committing move.

6.Bd3 is usually not chosen this early in the Saemisch, but is absolutely playable. Following both 6…c5 7.d5 e6 8.Nge2 a6 9.a4 Qc7, the position would resemble a Benoni, while 6…e5 7.d5 c6 8.Be3 transposes to the 6.Be3 line.

6.Nge2 at first looks like an awkward move since it moves this knight in the way of the f1-bishop. But White will often embark this knight on an interesting route. The most common alternatives are 6.Be3 and 6.Bg5, although 6.Bd3 e5 7.d5 is also playable. Following 6.Be3, play could continue with 6…e5 7.d5 Nh5 8.Qd2 f5 or 6…Nc6 7.Qd2 a6 8.Nge2 Rb8. We are looking at the 6.Be3 Saemisch from Black’s point of view in another challenge. 
Developing the dark-squared bishop to g5 with 6.Bg5 instead can lead to similar positions if Black challenges this bishop with …h7-h6 soon and White retreats this bishop from g5 to e3. A separate main line without …h6 is 6…c5 7.d5 e6 8.Qd2 exd5 9.cxd5 a6 10.a4 with an unclear position. Some of the ideas in this line are borrowed from the Benoni.

Black has to make a decision how he will go about challenging White’s pawn center. 6…c5 tries to open the long a1-h8 diagonal for Black’s bishop on g7 and prepares counter play on the queenside. If Black instead advances the e-pawn with 6…e5, White often responds with 7.Bg5. Then both 7…c6 8.Qd2 Nbd7 9.d5 and 7…Nbd7 8.d5 h6 9.Be3 Nh7 lead to complex positions. 
If Black instead develops with 6…Nc6, he can meet 7.Be3 with either 7…a6 8.Qd2 Rb8 9.h4 h5 or 7…e5 8.Qd2 Nd7 9.d5 Ne7 10.g4 f5. Both would lead to unclear play.
———————————————————
White has a nice pawn chain in the center. But the White pawn on f3 took away the natural square of the kingside knight. White will have to be a little resourceful to give this horse a good post later on.

7.d5 changes the pawn structure and keeps the center closed. Now the position has similar characteristics to most Benonis. White could have also played 7.Be3 instead. Black does not have to protect the c5-pawn and often continues with 7…Nc6 next. Then 8.d5 Ne5 9.Ng3 e6 or 8.Qd2 b6 9.d5 Ne5 is unclear. Note that exchanging pawns on c5 with 7.dxc5?! dxc5 gives Black greater control of the dark center squares.

Black continues to challenges White’s pawn center with 7…e6. This is necessary in these types of positions as Black will otherwise not find enough space to move his pieces in. Black tried a different plan in Akhsharumova-I. Ivanov (1988) following 7…a6 8.a4 e5. But White got the better chances with 9.Bg5 h6 10.Be3 Nh7 11.h4 f5 12.h5.
———————————————————
Usually the player who has greater pawn control in the center will have more good options for his pieces.

Developing the dark-squared bishop to g5 with 8.Bg5 is a fine idea now. The main idea is to tempt Black into playing …h7-h6 and only then move this bishop to e3. While your move is playable, we prefer getting a step closer to castling kingside and finding a good post for the e2-knight.

8.Bd2? gets this bishop in the way of the queen and does not improve White’s position. This bishop has several better options available.

8.Be3 develops this bishop to a fine square. However, if you want to develop this bishop now, White usually prefers to go to g5 first. The main idea is to tempt Black into playing …h7-h6 and only then move this bishop to e3. While your move is playable, we prefer getting a step closer to castling kingside.

8.g3?! is a bit awkward after playing f2-f3 previously. White’s pawn structure on the kingside is now weakened more. Developing the light-squared bishop to g2 would not be a good idea, as the bishop would be very passive there. Try to develop this bishop in a different manner.

8.Ng3 begins the odyssey of this knight. On g3, the kingside horse doesn’t look much better placed than on e2, as Black’s g6-pawn take away two forward options. But this move gets the horse out of the way of the f1-bishop and thus allows White greater control of the f1-a6 diagonal. White will later try to improve the position of this horse. 
White’s other fine option now is 8.Bg5, when play usually continues with 8…h6 9.Be3 exd5. Note that taking on e6 now is not very promising. After 8.dxe6?! Bxe6 9.Ng3 Nc6, Black would be ahead in development.

8…exd5 immediately clarifies matters in the center. As Black has to play this move in the near future, there are no other independent ideas at this point.
———————————————————
An important decision has to be made.

9.cxd5 is White’s best recapture and changes the dynamics of the pawn structure. Black now has a three against two pawn majority on the queenside, while White has an extra center pawn on e4. Had White captured with the tame 9.exd5?! instead, he wouldn’t have any chances for an edge at all. Following 9…a6 10.a4 Ne8 11.Bd3 Nd7 12.f4 Qh4, Black was already gaining the upper hand in Soffer-Djuric (1992). Note that exchanging knights with 9.Nxd5 Nxd5 only helps Black, as he has less space and will be happy to see one of his pieces leave his crowded quarters.

9…Nbd7 gets the queenside knight into action. This knight is now in the way of the c8-bishop, but that bishop has no reasonable options to choose from along the h3-c8 diagonal anyway. If Black chose 9…a6 instead, play should transpose later if you find the correct reply. Note that also 9…h5 is playable. Then both 10.Be2 Nbd7 11.0-0 a6 and 10.Bg5 Qa5 were featured in the Spassky-J.Polgar 1993 match.
———————————————————
The g3-knight will later often embark on a surprising journey to the other side of the board.

10.Bd3 develops this piece to a reasonable square. But after 10…Ne5, Black’s knight may capture this bishop under favorable circumstances later on. We prefer not giving Black this option.

10.Be2 develops this bishop to a modest post where it can’t be bothered by Black’s pieces. This is clearly superior than 10.Bb5? a6, 10.Bc4? Ne5 or even 10.Bd3?! Ne5. If White tried to advance on the kingside with 10.h4 and h4-h5 to follow, Black does best to stop this idea by playing 10…h5. White would then find it more difficult to safely castle on the kingside later on. 
White could also develop the dark-squared bishop to e3, f4 or g5 now. But we will soon see that it is best for White to delay this a little more and wait for Black to reveal his plans first.

10…a6 tries to get Black’s pawn majority on the queenside into motion. If Black advanced with 10…h5 instead, White should reply in the same manner as he will a bit later in our chosen line.
———————————————————
White is in no rush to castle kingside.

Castling kingside with 11.0-0 is not a well timed move as Black could now expand on the queenside with 11…b5. White should try to make it much harder for Black to get his queenside pawn majority into motion.

11.a4 is an almost automatic response in these types of positions when Black tries to expand on the queenside with …a7-a6 and …b7-b5. Now Black will have a much harder time to get his queenside pawn majority into motion and find counter play on the queenside.

11…h5 advances on the kingside and tries to mix things up. In the game Antonsen-Mortenson (1993), Black consistently pursued the plan of advancing with …b7-b5 instead. Following 11…Rb8 12.0-0 Ne8 13.Be3 Nc7 14.f4 b5, Black had fine counter play. However, it is not hard to improve on White’s tame play in this game. Instead 12.Bf4 Ne8 13.Qd2 makes it much harder for Black to get …b7-b5 in. 
Black has also tried 11…Re8 at this juncture. Then 12.Bg5 h6 13.Bf4 Qe7 14.0-0 b6 lead to a complex battle in Balashov-Pospelov (1994).
———————————————————
Since there are few immediate tactics in this line, both sides are mostly creating strategic plans.

12.Bg5!? pins Black’s f6-knight and stops the advance of Black’s h5-pawn to h4. This bishop is well placed on g5 now that Black can no longer challenge it with …h7-h6. If White instead castles kingside with 12.0-0, play could continue with 12…Nh7 13.Be3 Re8 14.Qd2 f5 and about even chances.

12…Qe8 unpins the f6-knight and still hopes to create active play on the kingside. But this move is probably not Black’s best option. The game Korchnoi-Xie Jun (1999) instead continued with 12…Rb8 13.Qd2 Qc7 14.Bh6 c4 15.Bxg7 Kxg7 16.a5 and Black stood fine. White should probably try to improve on this game with 13.0-0 Qc7 14.Qb3. Black will then have a much harder time to play …c5-c4 and …b7-b5. 
A similar plan by Black was tried in Jussupov-Kotronias (1997), where Black played 12…Qe7 instead. Then 13.Qd2 Re8 14.0-0 Rb8 15.Rfc1 Nh7 16.Bh6 Bh8 gave White only a small edge.
———————————————————
The player with the better piece coordination can generally outmaneuver his opponent over time. One should aim for harmony between one’s pieces.

Castling kingside with 13.0-0?! now is inaccurate. After 13…Nh7, White couldn’t continue in the same active manner as in our line.

13.f4? advances the f-pawn. This move creates several weaknesses, most notably the e4-pawn and the g4-square. It also leaves the g5-bishop with hardly any good retreat squares. This is quite awkward and does not lead to harmonious piece coordination amongst White’s units.

13.Qd2 demonstrates how to support a bishop that was developed to a more advanced square along the same diagonal. This is much superior to 13.Qc1?! since the queen would then be in the way of the a1-rook in many lines. Also castling kingside with 13.0-0?! would be inaccurate. After 13…Nh7, White couldn’t then continue in the same active manner as in our line.

13…Nh7 gets out of the way of the g7-bishop and the f7-pawn and also eyes the g5-bishop. Black will usually get this knight to a more active square later on after he tries to achieve other positional goals.
———————————————————
White’s light-squared bishop on e2 is the inferior one with this type of pawn structure. This is due to White’s pawns on d5, e4 and f3 that greatly limit its scope.

14.Bh6 moves the dark-squared bishop out of reach of Black’s h7-knight and plans to exchange this bishop for Black’s g7-bishop. This will make the dark squares around Black’s king more vulnerable.

14…Qe5 tries to maintain piece control over the dark squares. Perhaps Black should consider the more aggressive 14…h4 15.Nf1 f5. Then 16.Bxg7 Kxg7 17.exf5 gxf5 18.Qg5+ Qg6 19.Qxh4 Qxg2 20.Qg3+ gives White a somewhat better endgame.
———————————————————
The defects of Black’s position will soon be exposed.

Castling queenside with 15.0-0-0?! is not very safe in this position now that White has played a2-a4. Black could quickly open up the queenside with the pawn sacrifice 15…b5.

Castling kingside with 15.0-0?! now is inaccurate. Black could then reach a fine endgame with 15…Qd4+ 16.Qxd4 Bxd4+ 17.Kh1 Re8. You should keep the queens on the board.

15.f4? leaves White down an important center pawn following 15…Bxh6 16.fxe5 Bxd2+ 17.Kxd2 Nxe5.

15.Bxg7 is a fine move that doesn’t bind White’s queen to the protection of this bishop any longer. You should always look at the most forcing continuation and see if you can find a favorable line of play. This way your opponent will have fewer responses he can consider. This will make it easier for you to calculate several moves ahead. 
Castling kingside with 15.0-0?! would have been inaccurate. Black could then reach a fine endgame with 15…Qd4+ 16.Qxd4 Bxd4+ 17.Kh1 Re8. White should keep the queens on the board.

15…Qxg7 gives Black’s king an extra defender and vacates the e5-square for Black’s d7-knight. If Black recaptured with 15…Kxg7 instead, White could reach a good position with 16.f4. Then all a) 16…Qf6? 17.Bxh5, b) 16…Qd4 17.Qc2 and Rd1 to come and c) 16…Qe7 17.0-0 give White a clear edge.
———————————————————
White has to make sure he doesn’t end up in an endgame with his inferior light-squared bishop against Black’s light-squared bishop or a Black knight.

16.Nf1! is the only move that justifies White’s previous play. This knight is now aiming for the ideal c4-square, where it will also attack Black’s weak d6-pawn. Since the g3-knight is doing nothing useful now, it makes sense to move this horse over to the queenside. White could have also moved the c3-knight to c4 (via d1 and e3). But that would have reduced White’s control over the important b5-square. White instead castled kingside with 16.0-0?! in the game Disconzida Silva-Leitao (1996), but Black then traded queens with 16…Qd4+ 17.Qxd4 cxd4. After 18.Nb1 a5! Black soon got a knight to c5 and already had a clear edge since most of his pawns will be on dark squares, while most of White’s pawns are on light squares. Since both sides only have a light-squared bishop left, this is very important. Had Black played 16…Qd4, White could have avoided queen trade with 17.Qc2 and then chase back Black’s queen with Rd1 to follow. 
Following 16.Nf1!, the game Ivanchuk-Kramnik (1996) continued with 16…f5 17.exf5 gxf5 18.Ne3 Kh8 19.Nc4 Qf6 and Black’s queen protected the d6-pawn.

This is better than 19…Qxg2 20.0-0-0 Qg6 21.Rhg1 Qf6 22.f4, and White has a very strong initiative for the pawn. After 19…Qf6, White then castled kingside on move 20! Castling this late can sometimes be done when the position is closed and other positional goals need to be pursued first. Trading queens on d4 with 20…Qd4+ 21.Qxd4 cxd4 was no longer advantageous for Black due to 22.Nb1 Rf6 23.Nbd2. Black instead tried to generate counter play on the queenside with 20…Rb8. Black soon played …b7-b5, but White’s c4-knight then moved to a5 and soon reached the ideal c6-forepost. Black didn’t last much longer as White penetrated Black’s position via the queenside. It is also important to notice that Black’s attempt to trade off White’s strong c4-knight with 20…Ne5 would have been futile. Then 21.Nb6! keeps this knight on the board. After 21…Rb8, White could push back the e5-knight with 22.f4 and then following 22…Nd7 retreat the b6-knight to c4 and reach similar positions as in our main line. 
It is noteworthy that Black never got any active play on the kingside. All he did there was expose his own king and get two isolated pawns on f5 and h5. White’s strategy is fairly simple in this line. He tries to slow down Black’s counter play on the queenside with a2-a4 and Black’s plans on the kingside by playing Bg5 at the right time. Note that White never moved this bishop to the normal e3-post. Once White traded the dark-squared bishops, he needs to make sure to avoid queen trade.

The c4-square is often the perfect place for a White knight in this pawn formation, especially if White can meet a later …b7-b5 challenge by moving the c4-knight to a5 and then c6. Black has many attempts to improve on this line. He can try 6…e5 or 6…Nc6 early on or he can play 11…Rb8 or 12…Rb8 much later. White can follow a different plan by playing Be3 at move 6 or 7.