The Hungarian Attack, beginning with 5.Nge2, is a less explored system that can cause Black trouble if he isn’t familiar with White’s basic ideas. Typically White will continue with moves such as Ng3, Bg5 and h4-h5 to attack Black’s castled king early on. In this challenge, White never gets very far with this plan as Black keeps him busy in the center. We are also looking at this system from White’s point of view in another challenge.
———————————————————
After the opening moves 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6, White has just played 5.Nge2. This move looks a bit awkward since it gets the knight in the way of the f1-bishop. From e2, this horse also doesn’t seem to have many good forward options. But White’s plan should by no means be underestimated. We will now follow the game Yusupov-Shaked (1997), which leaves the beaten path of opening theory very early on.

5…c5 is probably not Black’s best way to challenge White’s pawn center, but is playable. After both 6.dxc5 dxc5 7.Qxd8+ Kxd8 8.e5 Nfd7 and 6.d5 b5 7.cxb5 a6, White can hope for a small edge. Very good, we are instead looking to complete kingside development.

5…0-0 completes kingside development before attempting to find any counter play in the center or on the queenside. This is the most popular move order although Black can also play 5…Nc6, 5…Nbd7, 5…e5, 5…c6, 5…c5 and 5…a6 at this point. 5…Nc6 develops the queenside knight to an aggressive post. After 6.d5 Nb8 7.Ng3 0-0, play was unclear in Szabo-Westerinen (1967). We slightly prefer moving this knight to d7 instead. Following 5…Nbd7 6.Ng3 c6 7.Be2 a6 8.Be3 h5, chances were even in Serper-Nikolaidis (1993). 
5…e5 6.d5 has done very well in practice for Black after both 6…Na6 7.Ng3 h5 and 6…c6 7.Ng3 0-0. Challenging White’s center with 5…c5 instead is also playable, but gives White more options. Both 6.dxc5 dxc5 7.Qxd8+ Kxd8 and 6.d5 b5 7.cxb5 a6 lead to very different types of positions. Black could also play 5…c6 or 5…a6 now. Both moves typically try to prepare advancing with …b7-b5 a little later.

6.Bg5 develops the dark-squared bishop to an aggressive post. This is often played later in the Hungarian Attack. The typical Hungarian Attack continues with 6.Ng3. Then 6…c6 7.Be2 Nbd7 8.Bg5 h6 was unclear in I.Sokolov-Van Wely (1994). If White chose 6.f3 instead, play would transpose to a Saemisch.
———————————————————
Black will choose an active means to stop White’s plans on the kingside.

6…a6 sometimes prepares queenside expansion with …b7-b5 to come. This is a reasonable idea and led to unclear play in Chernin-Lautier (1989) after 7.Qd2 Nbd7 8.Ng3 c6 9.a4 Qa5 (1989). While this is a fine idea, Shaked has chosen to challenge White’s kingside plans right away.

6…Bg4? loses valuable time after 7.f3. White could then play a Saemisch with an extra tempo.

6…h6 is a double-edged move to play in most King’s Indians. This move gains time by attacking the g5-bishop, but it creates a potential target in front of Black’s monarch. Black most promising alternatives are probably 6…a6, 6…c6 and 6…Nbd7. After 6…a6 7.Qd2 Nbd7 8.Ng3 c6, play was unclear in Chernin-Lautier (1989). 
Instead 6…c6 is also a fine idea, but 7.Qd2 b5?! 8.cxb5 cxb5 9.a3 Nc6 was better for White in Smyslov-Sherbakov (1955). Black probably does better to prepare the …b7-b5 pawn advance with 7…a6 instead. 6…Nbd7 develops the queenside knight to its best post. Then 7.Qd2 c6 8.Ng3 a6 leads to the same position as in the lines featuring an earlier …a6 or …c6.

7.Bf4 is an ambitious move that tries to stop Black from playing …e7-e5 for some time. As we will soon see, this is not without risk. The more conventional 7.Be3 didn’t promise White anything special after 7…Ng4 8.Bc1 c5 8.d5 e6 in Larsen-Gligoric (1973). Perhaps the slightly obscure 7.Bh4 g5 8.Bg3 c5 9.d5 deserves some attention.
———————————————————
Tal Shaked won the World Junior Championships in 1997, but has since retired from active tournament chess.

7…e5?? loses a pawn to 8.dxe5 dxe5 9.Qxd8 Rxd8 10.Bxe5. This desirable move needs to be prepared better.

7…a6 sometimes prepares queenside expansion with …b7-b5 to come. This is a reasonable idea, but is a bit slower than our line. Shaked chose a more active course of play in the center.

The solid 7…c6 is a flexible move that often prepares to advance on the queenside with …b7-b5 a little later. While this is probably playable, Shaked has chosen a more active course of play in the center.

7…Nc6 is an aggressive move that tries to gain control of the e5-square and pressure White’s d4-pawn soon. As we will soon see, Shaked had already foreseen White’s next move.

8.Qd2 attacks White’s h6-pawn and seems to puts Black on the defense. Instead 8.d5 Na5 9.Ng3 c5 or 8…Ne5 9.Ng3 Nfd7 10.Qd2 h5 would lead to unclear positions.
———————————————————
There already is early tension in this line.

8…e5! strikes back in the center and doesn’t bother to defend Black’s h6-pawn. This is Black’s best move by far. If Black instead took care of the attacked h6-pawn by playing a move such as 8…Kh7?! or 8…h5?!, White could next improve his position with 9.Rd1 and stop Black’s idea of advancing with …e7-e5 for some time.

9.Bxh6 captures Black’s h-pawn. This will lead to an unclear position and is better than taking on e5 first. After 9.dxe5 dxe5, Black benefits from opening the position since he is better developed. If White then tried to win a pawn with 10.Bxh6??, Black would reach a winning position with 10…Bxh6 11.Qxh6 Nb4! Black then threatens both 12…Nc2 mate and 12…Nd3+.
———————————————————
Artur Jusupov was a candidate for the World Championship in the late 1980s. In this game, however, his opening choice is not good enough to give him any hope for an edge.

9…Bxh6 forces White’s queen away from the center of action. Usually it is quite dangerous to allow the opponent’s queen to get so close to your king. But in this case there is no danger since White’s queen is all by herself. If Black instead captures on d4 with the e5-pawn or the c6-knight, White could next exchange bishops on g7 and then keep his extra pawn by taking on d4.

10.Qxh6 is forced.
———————————————————
Again there aren’t many good choices.

Taking with the e5-pawn on d4 is a reasonable move. After 10…exd4, White should choose the sharper 11.Nd5 Nxe4 12.Nf4 over 11.Nb5 Nxe4 12.Nbxd4? Nxd4 13.Nxd4 Qf6. Once a White knight appears on f4, Black has to watch out for a knight sacrifice on g6. We prefer a different means of regaining the pawn.

10…Nxd4 regains the pawn and threatens the knight fork on c2. Black’s only reasonable alternative is 10…exd4. White should then choose the sharper 11.Nd5 Nxe4 12.Nf4 over 11.Nb5 Nxe4 12.Nbxd4? Nxd4 13.Nxd4 Qf6.

11.Qd2 retreats the queen to a square where it covers c2 and also threatens to win a pawn on d4 next.
———————————————————
One should always try to make things as difficult as possible for one’s opponent.

11…c5 reinforces the knight on d4. If White now takes this powerful horse, Black would get a protected passed pawn on d4. This is clearly better than 11…Nxe2? 12.Bxe2, which would help White catch up in development.

12.Ng3 improves the position of White’s kingside pieces. It is safe to say that White’s opening isn’t a success. Black is better developed and has all the time in the world to improve his position on the following moves.
———————————————————
After a more or less forced sequence of moves, both sides will now look to complete development and find a good plan for the coming middle game.

12…a6 prepares active play on the queenside. This is a fine plan, especially considering that White may be bold enough to castle on that side of the board. Black had a few other reasonable moves to look at. 12…Kg7 allows Black to play …Rh8 later, while 12…Nh7 gives Black the option to play moves such as …f7-f5 and …Qh4 next. 
Instead 12…Qa5 threatens to win material with 13…Nb3 next. This motif will occur in our line as well, but in a different form. White should probably try to simplify things with 13.Nd5. Then 13…Nc2+ 14.Kd1 Qxd2+ 15.Kxd2 is unclear. We don’t want to exchange queens in this line since White’s king will have a difficult time finding a safe haven.

13.Bd3 gets the light-squared bishop into play. Note that the presence of the center pawns on c4 and e4 greatly limit the scope of this piece. Black can be very happy to have traded off his inferior dark-squared bishop for White’s superior dark-squared bishop.
———————————————————
Unlike many other King’s Indians, this position can quickly open up more and lead to sharp tactical play.

13…b5 continues Black’s plan of pressuring White’s queenside and opening some lines on that side of the board.

14.h4 tries to create pressure against Black’s king by opening the h-file. By now it is obvious that White does not want to castle on the kingside in this game. If White castled kingside now with 14.0-0, Black could quickly get the initiative on the kingside with moves such as …Kg7, …Rh8 and …Ng4. For example, 14…Kg7 15.Qg5 Rh8 16.Nf5+ Bxf5 17.exf5 Rh5 18.Qg3 e4! wins a pawn for Black. 
Note that White was ill advised to take on b5. Following 14.cxb5? axb5, White cannot capture on b5 next without allowing the knight fork …Nb3. Since White’s a2-pawn is pinned along the open a-file in that line, Black would then win the exchange.
———————————————————
Is it time for Black to start defending now?

14…Ng4? does not place this knight on a good square. White could always push you back with f2-f3.

14…Bg4?! tries to make it harder for White to advance with h4-h5 next, stops White’s idea of castling queenside next and hopes for the blunder 14.f3?? Bxf3. But after 15.Qg5, White renews the threat of h4-h5 and eyes the loose g4-bishop. We prefer a different post for this piece.

14…Be6 increases the pressure against White’s c4-pawn. Clearly taking with 14…bxc4? 15.Bxc4 would have only helped to activate White’s light-squared bishop. Also 14…b4? 15.Nd5 is undesirable for Black since he wants to open up lines on the queenside, not keep them closed. 
Black could have also addressed White’s idea of playing h4-h5 now. 14…Kg7 is a fine idea, since 15.h5 could then be met with 15…Rh8. Black should avoid playing 14…Bg4?!, since after 15.Qg5, the g4-bishop is loose and White renews the threat of h4-h5. In summary, Black does not have to begin defending against h4-h5 until after it has been played.

15.0-0-0 castles queenside. This will not give White’s king much safety, but connects White’s rooks. Perhaps moving the king to f1 and g1 later on would have been preferable. If White instead continues the kingside attack with 15.h5, Black can defend with 15…Qe7, e.g., 16.Qh6 Ng4 or 16.hxg6 fxg6 17.Qh6 Qg7. 
Clearly White doesn’t want to open lines on the queenside with 15.cxb5? axb5. White’s a2-pawn is then pinned along the open a-file in that line, allowing Black to win the exchange with …Nb3 if White takes the b5-pawn. But 15.Nd5 instead looks like a reasonable alternative to avoid the bloodbath of this game.
———————————————————
The stage is set for a wild game.

15…bxc4 wins a pawn with gain of time and allows Black to open the b-file leading to White’s king. This is Black’s best move. The other capture 15…Bxc4 threatens 16…Bxa2 next. But after 16.Kb1, it would be harder for Black to open a file leading to White’s king. Black now had to resist the temptation to play the active looking 15…Qa5? The vis-?-vis with White’s queen allows a common tactical motif. After 16.Nd5, both Black’s queen and the f6-knight are attacked. Trading queens on d2 and retreating the lady to d8 both don’t help Black make progress in his attempts to attack White’s monarch. The piece sacrifice 16…Qxa2? instead doesn’t seem to work after 17.Nxf6+ Kg7 18.Qg5!, when White threatens to mate Black’s king with Ngh5+ next. Black, on the other hand, doesn’t seem to have a perpetual since White’s king can run to e1. 
After 15…bxc4, the game Yusupov-Shaked (1997) continued with 16.Bb1. Black then took a move to defend on the kingside with 16…Nh7. This allows Black to respond to h4-h5 with either …Qg5 or …g6-g5. In either case, White’s chances on the kingside have been stopped. Following 17.Nf1 a5 18.Bc2 a4, White stopped Black’s a-pawn with 19.a3. Then 19…Qb6 and 20…Rfb8 gave Black strong pressure against White’s king. Black later chased White’s monarch all the way to the kingside. Clearly castling queenside in this line was too optimistic since White’s king soon became the center of all attention while Black’s monarch was safe on g8. 
In this case trading Black’s h-pawn for White’s d-pawn in the opening worked out very well for Black. This is because White was unable to get other pieces involved in his efforts against Black’s king. White should probably look to improve on this game early on. Both 6.Ng3 and 7.Bh4 are candidates for that. We don’t like White’s position after 8…e5, but it appears that 15.Nd5 could have minimized the damage. In this line Black was able to play an early …e7-e5 because White’s knight on e2 was placed awkwardly, blocking both the king and the f1-bishop. Black has many interesting alternative on moves 5 and 6, but the chosen path appears to be most accurate.