I used to play the King's Indian Defense as Black only when opponents committed to an early Nf3, and I would switch to the Janowski Indian, Tango or Budapest if they didn't. This way I could avoid the dreaded Four Pawns Attack and Sämisch Variation. But I recently decided to commit to the King's Indian completely, which means learning a defense against everything White can throw at you.
The Sämisch was a big worry of mine until I came across the system developed by the late American GM Robert Byrne (1928-2013) that goes 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 c6 6.Be3 a6 with the plan of ...b5 (E80 in ECO). As GM Efstratios Grivas has analyzed in some key articles and GM Eugene Perelshteyn helpfully discusses in a series of videos, this is quite a viable line -- especially if Black correctly delays castling in order not to provide White an easy target on the kingside. This is exactly how Byrne himself played it, so it is puzzling to find so many sources giving 5....O-O(?) as the standard move in the Byrne (E81 in ECO). Castling early not only creates a target, but it delays Black's queenside counterplay. The ...b5 advance indirectly strikes at the center, and one idea that Black has is to exchange off White's c-pawn by bxc4 and then break with d5. Black can also attack the center with ...e5 or ...c5 in some lines. It is a very flexible system.
Those adopting the Byrne against the Saemisch might also want to use it against the Hungarian 5.Nge2 with 5...c6 6.Ng3 a6 7.Be2 b5 (as recommended by Joe Gallagher in Beating the Anti-King's Indians, for instance), since White otherwise might transpose to the Sämisch later with 6.f3 leaving you flat-footed.
The system with ...a6, ...c6, and ...b5 was the subject of renewed interest following Byrne's death last year, so I suspect it is due for revival. I have personally never been a fan of the widely recommended anti-Sämisch gambit with 5.f3 O-O 6.Be3 c5, especially after I analyzed the game Elsness - Gallagher, Gotheberg 2005, where I found it hard to prove compensation for the pawn after 7.dxc5 dxc5 8.Qxd8 Rxd8 9.Bxc5.
I have only been able to consult publications on the Byrne from the last decade, but I am sure it was analyzed in print much earlier. I therefore welcome reader additions and suggestions -- especially regarding its early history.
The King's Indian Defense: Beating the Sämisch! - Part 3 by Eugene Perelshteyn, Chess.com (January 2014). GM Perelshteyn's introduction to the Byrne variation looks at his own games with the line -- here Fishbein - Perelshteyn, Burlington Open 2012.
The King's Indian Defense: Beating the Sämisch! - Part 2 by Eugene Perelshteyn, Chess.com (January 2014)
The King's Indian - Beating the Sämisch! by Eurgene Perelshteyn, Chess.com (December 2013) Discusses the game Novikov - Perelshteyn, San Diego 2004.
Robert Byrne and My Modern Defence by Nigel Davies, The Chess Improver (April 2013). Examines Larsen - Byrne, Leningrad 1973.
"Instructive Pawn Play in the Saemish King's Indian" by Eugene Perelshteyn, ChessLecture.com (January 22 2013). Runs 14:41 minutes. Analyzes Matthew Fishbein - Eugene Perelshteyn, Burlington Open 2012 which began 1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. f3 a6 6. Be3 c6 7. Qd2 b5 8. a3 bxc4 9. Bxc4 d5. By subscription only.
"Several Ways to an Advantage: Saemisch System, Byrne Variation" by Boris Schipkov ChessBase Magazine #140 (February 2011). Looking at the Byrne from the White perspective, Schipkov agrees with Grivas that if Black castles White gets good attacking chances by normal means (Qd2, g4, Bh6, h4 etc.), but he also tries to show chances for a White edge after 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 c6 6.Be3 a6 by one of three methods: (1) 7.c5 b5 8.cxb6 with play on the dark squares or by attacking the c-pawn; (2) 7.Bd3 b5 8.e5 Nfd7 9.f4 bxc4 10.Bxc4 Nb6 11.Be2 or 11.Bb3; and (3) 7.g4 b5 8.h4 h5 9.g5 with some advantage in space for White. Games include Campos Moreno - Can, Khanty Mansiysk 2010; Narcisco Dublan - Sielicki, Andorra op 2008; Botsari - Managadze, Nikaia op 2009; Del Rio Angelis - Jaksland, Calvia 2005; Volodin - Seeman, Tartu 2010; Graf - Hug, Mitropa Cup 2002; Tomashevsky - Soto Paez, Khanty Mansiysk 2010; Chekhover - Shamkovich, Leningrad 1953; Piket - Van Wely, Hoogovens 1999; and Sakharov - Nesterov, Udmurtia 2008. Schipkov chooses games that favor White and do not feature best play by Black. The lines he recommends are challenging, but they should not discourage amateur players who will rarely if ever encounter the most theoretically correct lines of play from opponents.
"Initiative at all Costs!" by Eugene Perelshteyn, ChessLecture.com (January 13, 2011). Runs 19:58. Analyzes Chanda Sandipan - Eugene Perelshteyn, King of Prussia 2010, which opened 1. d4 Nf6 2. c4 g6 3. f3 Bg7 4. e4 d6 5. Nc3 a6 6. Bg5 c6 7. Qd2 b5 8. O-O-O Bb7!? (unusual) 9. e5 dxe5 10. dxe5 Qxd2+ 11. Bxd2 Nfd7. By subscription only.
"Tricky Pawn Moves in the Opening" by Eugene Perelshteyn, ChessLecture.com (October 22, 2010). Runs 23:10. Analyzes the game Darwin Yang - Eugene Perelshteyn, Arlington 2010, which began 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 a6 6.Be3 c6 7.c5 O-O 8.Nge2 b6 9.cxd6 exd6 10.Qd2 b5 11.g4 b4. By subscription only.
"Recent Developments in the Byrne System" by Efstratios Grivas, New in Chess Yearbook 92 (2009): 187-195. Grivas has written a number of articles on the Byrne, and here he does his usual detailed and thorough analysis. Main games include Przedmojski - Witek, Warsaw 2008; Moiseenko - Grivas, Kemer 2009; Andrianov - Grivas, Athens 1993; Schandorff - Erdogdu, Dresden 2008; and Papaioannou - Grivas, Iraklion 1995.
"379. 22 April 2008: Stunning computer move in a pre-computer game" by Tim Krabbe, Open Chess Diary (2008). Examines Timman - Greben, Amsterdam 1967.
King’s Indian Saemisch System (CD) by Boris Shipkov, ChessBase (2007)
Reviewed by Carsten Hansen.
The Sämisch King's Indian Uncovered by Alexander Cherniaev and Eduard Prokuronov, Everyman Chess (2007): 73-88. The authors begin already with a grave error by only considering lines that arrive at the Byrne via the move order 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O(?) 6.Be3 c6, when Black is prematurely castled and exposed to a straightforward attack on the kingside. The relatively slim volume offers scant coverage of the best lines in the Byrne Variation, though the games offered show relatively balanced play. Sample games include Wang Yue - Zvjaginev, Petrosian Memorial 2004; Riazantsev - Novikov, Moscow 2007; Haba - Golubev, Bundesliga 2001; and Postny - Socko, Moscow 2002. I did not find this book useful, even as an overview of the Saemisch.
King's Indian Saemisch, Byrne Variation (E80), by Efstratios Grivas, ChessBase Magazine #115 (December 2006). Also available in the ChessBase's Encyclopedia of Chess Openings. Grivas analyzes over 30 of his games with the Byrne Variation. Those looking to analyze the line closely would do well to begin with Grivas's PGN files here and with his article in the 2009 NIC Yearbook (see above). Games analyzed include Mishra - Grivas, Sharjah 1985; Marinelli - Grivas, Vinkovci 1989; Ionescu - Grivas, Kavala 1990; Mitrandzas - Grivas, Athens 1991; Kalesis - Grivas, Corfu 1991; Kalesis - Grivas, Athens 1991; Davies - Grivas, Tel Aviv 1991; Sarno - Grivas 1992; Raicevic - Grivas, Athens 1992; Knaak - Grivas, Athens 1992; Kalesis - Grivas, Athens 1993; Botsari - Grivas, Corfu 1993; Andrianov - Grivas, Athens 1993; Papaioannou - Grivas, Iraklion 1995; Konstandinou - Grivas, Aegina 1995; Orfanos - Grivas, Aegina 1996; Elsness - Grivas, Yerevan 1996; Atalik - Grivas, Karditsa 1996; Kanellakis - Grivas, Athens 1997; Efthimakis - Grivas, Athens 1997; Mouroutis - Grivas, Athens 1999; Nikolaou - Grivas, Athens 2000; Kaminellis - Grivas, Athens 2000; Meister - Grivas, Hungen 2002; and Vafiadis - Grivas, Aghia Pelaghia 2004.
The Controversial Sämisch King's Indian by Chris Ward, Batsford (2004): 118-126, 188-192. Ward offers a useful overview of the Saemisch, but his game selection seems a little idiosyncratic and White-focused. The treatment of the Byrne Variation is also rather limited, and games include Levitt - Friedgood, Birmingham 1998; Campos Moreno - Candela Perez, Orense op 1997; and Motwani - Hanley, British Ch 2004.
King's Indian Battle Plans by Andrew Martin, Thinker's Press (2004): 75-77, 86-88. There are many original ideas in Martin's book, so he naturally does cover the Byrne Variation in two games: Khenkin - Kozul, Belgrade 1999 and Ward - Hebden, British Ch Scarborough 2001.
The King's Indian Defence (2nd edition) by Leonard Barden, William Hartston, & Raymond Keene, Batsford (1973).
I am sure that there were articles before the turn of the Century on this line, and I welcome reader additions.