Black to move

Lots  going on!  How do you react to complex positions?  Get confused, or get a grip?

Organised thinking

The landmark description of a thinking process is Kotov's: select a list of candidate moves, work your way through them in turn.  Works for Kotov, maybe, but not everyone else has found it useful (or even possible) to imitate it. 

Forcing moves first

Look at the most forcing moves first: you can have more confidence in your opponent replying as you expect.  If at the end of a forcing sequence you think the position is at least as good for you as it is now, you can go for it.

Solving

Sometimes you can cut through a complex position by arranging a tactical accident for your opponent.  The right move in this next position would never be selected as a candidate move, but you can work out it's the right one.

[Event "Moscow"]
[Site "?"]
[Date "1973.??.??"]
[Round "14"]
[White "Kuzmin, Gennadi"]
[Black "Sveshnikov, Evgeny"]
[Result "1-0"]
[ECO "B49"]
[PlyCount "43"]
[EventDate "1973.??.??"]

1. e4 c5 2. Nf3 e6 3. d4 cxd4 4. Nxd4 Nc6 5. Nc3 Qc7 6. Be2 a6 7. O-O Nf6 8.
Be3 Bb4 9. Nxc6 bxc6 10. Na4 O-O 11. c4 Bd6 12. f4 Nxe4 13. c5 Be7 14. Bd3 Nf6
15. Bd4 Nd5 16. Nb6 Nxb6 17. Bxh7+ Kxh7 18. Qh5+ Kg8 19. Bxg7 Kxg7 20. Qg4+ Kh7
21. Rf3 Bxc5+ 22. Kh1 1-0

Kuzmin,G - Sveshnikov,E [B49] Moscow (14), 1973
1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nc6 5.Nc3 Qc7 6.Be2 a6 7.0-0 Nf6 8.Be3 Bb4 9.Nxc6 bxc6 10.Na4 0-0 11.c4 Bd6 12.f4 Nxe4 13.c5 Be7 14.Bd3 Nf6 15.Bd4 Nd5
[drumroll]

16.Nb6 !
[16.Bxh7+ Kxh7 17.Qh5+ Kg8 18.Bxg7 Kxg7 19.Qg4+ Kh8 20.Rf3 Nxf4 21.Rxf4 f5]
16...Nxb6 17.Bxh7+ Kxh7 18.Qh5+ Kg8 19.Bxg7 Kxg7 20.Qg4+ Kh7 21.Rf3 Bxc5+ 22.Kh1 1-0

Just plunge in

Analysis to Corden-Nunn, 1978

[Event "analysis: just plunge in"]
[Site "analysis: just plunge in"]
[Date "1975.??.??"]
[Round "?"]
[White "corden"]
[Black "Nunn, birmingham"]
[Result "0-1"]
[ECO "C55"]
[PlyCount "24"]

1. e4 e5 2. Nf3 Nc6 3. d4 exd4 4. Bc4 Nf6 5. e5 d5 (5... Ne4 {also =}) 6. Bb5
Ne4 7. Nxd4 Bc5 {[#]  has more attacking potential} (7... Bd7 {is the book move
}) 8. Nxc6 {?} Bxf2+ 9. Kf1 Qh4 10. Qxd5 (10. Nd4+ c6 11. Nf3 (11. Nxc6 O-O)
11... Ng3+ 12. Kxf2 Ne4+ 13. Ke2 Qf2+ 14. Kd3 Bf5 {[#]  "with decisive attack".
Just plunge in:} 15. Ba4 (15. Nc3 Nc5#) (15. Bd2 Nxd2+ 16. Kc3 Qe3+ 17. Bd3
Ne4+ 18. Kb3 Qb6+ 19. Bb5 Qxb5+ 20. Ka3 Qa5+ 21. Kb3 Nc5#) (15. Be3 Nd6+ 16.
Kc3 (16. Kd4 Nxb5+ 17. Kc5 Qxe3+) 16... Qxe3+) (15. Bf4 cxb5) (15. Qe1 Nc5+ 16.
Kc3 Qxc2+ 17. Kd4) 15... Nd2+ 16. Kc3 Qe3+ 17. Kb4 a5# {you can at least
regain one piece whenever you wish, still keeping the K exposed. NUNN &
GRIFFITHS}) 10... Be6 11. Qd3 (11. Nd4+ Ke7 12. Nxe6 fxe6 13. Qb3 Bc5 {-+})
11... O-O 12. Nb4 (12. Na5 Bb6) (12. Nd4 Bxd4 13. Qxd4 {??} Ng3+ 14. hxg3 Qxd4)
12... Bc5 0-1

Try 15. Ba4 Nd2+ 16.Kc3 Qe3+ 17.Kb4 a5#
So, you can at least regain one piece whenever you wish, still keeping the K exposed.

Don't Analyse Unnecessary Tactics [DAUT] (Nunn & Nimzowitsch)

Nimzowitsch-Tartakower, 1929

[Event "analysis: play a positional mo"]
[Site "analysis: play a positional m"]
[Date "1929.??.??"]
[Round "?"]
[White "Nimzowitsch"]
[Black "tartakower (karlsbad)"]
[Result "1-0"]
[ECO "E87"]
[PlyCount "49"]

1. d4 Nf6 2. c4 g6 3. f3 Bg7 4. e4 d6 5. Nc3 O-O 6. Be3 Nbd7 7. Nh3 e5 8. d5 a5
9. Nf2 b6 10. Qd2 Nc5 11. Bg5 Bd7 12. g4 Qc8 13. h4 Kh8 14. h5 gxh5 {[#]} 15.
Bxf6 {"all of this is extremely complicated and therefore I played after no
more than five minutes' thought"...} (15. Bxf6 Bxf6 {(Nimzo analysed briefly:)}
16. Qh6 Bg7 17. Qxh5 h6 18. g5 f5 {[#]  "now it is necessary to analyse
several variations which are roughly equally good:"} 19. gxf6 (19. gxh6) 19...
Rxf6 20. Bh3) 15... Bxf6 16. Rxh5 Bg7 17. Nh1 {[#]} f6 18. Qh2 h6 19. Ng3 Kh7
20. Be2 Rg8 21. Kf2 Rh8 22. Rh4 Qe8 23. Rg1 Bf8 24. Kg2 Nb7 25. Nh5 {[#] 
"with a strong solid game and chances of attack" - NIMZO} 1-0

 Nimzo analysed briefly:
15.Bxf6 Bxf6 16.Qh6 Bg7 17.Qxh5 h6 18.g5 f5...  
"...Now it is necessary to analyse several variations which are roughly equally good:"
19.gxh6...
19.gxf6 Rxf6 20.Bh3...
"All of this is extremely complicated and therefore I played after no more than five minutes' thought:"
15.Bxf6 Bxf6 16.Rxh5 Bg7 17.Nh1 f6 18.Qh2 h6 19.Ng3 +=

Strategic planning

Similarly, in the midst of complications, pick a move you want to make from a strategic point of view, and if you can't refute it, play it.

I seem to recall Nimzo (again) advising this pragmatic approach, and playing a centralising move like 19.Nd4 to improve his position in the event of any tactics, but I can't find it... anyone?

Complex analysis: growing a tree

Analysing a tree of candidate moves is best done by progressive deepening, rather than analysing them in turn, because you might find one that works, and you might learn things from one line that make another line work.

[Event "Stockholm"]
[Site "?"]
[Date "1962.??.??"]
[Round "?"]
[White "Kortchnoi, Victor"]
[Black "Olafsson, Fridrik"]
[Result "1-0"]
[ECO "D76"]
[PlyCount "73"]
[EventDate "1962.??.??"]

1. g3 g6 2. c4 Bg7 3. d4 Nf6 4. Bg2 d5 5. cxd5 Nxd5 6. Nf3 Nb6 7. Nc3 Nc6 8. e3
O-O 9. O-O a5 10. b3 Bf5 11. Bb2 Qd7 {#} 12. Ng5 Rfd8 13. Nge4 e5 14. Nc5 Qc8
15. d5 e4 16. Qe2 Nxd5 17. Nxd5 Rxd5 18. Nxe4 Bxb2 19. Qxb2 Bxe4 20. Bxe4 Rd6
21. Rad1 Qe6 22. Rxd6 Qxd6 23. Rc1 Ne5 24. Qd4 Nc6 25. Bxc6 Qxd4 26. exd4 bxc6
27. Rxc6 Ra7 28. Kf1 Kf8 29. Ke2 Ke7 30. Kd3 Kd7 31. d5 Rb7 32. Ra6 Rb5 33. Kc4
Rb4+ 34. Kc5 Re4 35. Rf6 Ke7 36. Rf3 Kd7 37. Rxf7+ 1-0

"IC[149]... seemed to have the best-organised thought processes (...) It is not clear, however, whether this logical approach is necessarily all that effective, for it could be argued that if IC had adopted the RW[208] approach of looking at moves twice he might have spotted 1. e4 Bg4 2. d5 Nd4! which seems good for Black"
 --WEBB

Safety net perpetuals

A familiar short-cut: if you know you have a perpetual in hand, you can commit without fear, and analyse again when things may be a bit clearer.

Further reading:

NIMZOWITSCH, My System
KOTOV, Think Like A Grandmaster
NUNN, Secrets of Practical Chess
JACOBS, Analyse to Win