Annotated chess game

14...e6 or 14...Nf6 look reasonable.

Agreed. 14...e6 exploits the half-pin that has suddenly appeared in the d-file, allowing Black to initiate pawns exchanges in the centre and free up some squares for his underemployed pieces. 14...nf6 brings in some supports for Black's projected K-side pawn advance. Make no mistake: ...g5 is part of Black's plan, it just seems, as the annotator indicates, a bit premature here.

This is not a good move IMHO. White should play on the queenside or in the center. 15. Rfd1 looks solid. There is no real danger on the kingside yet, so no need to touch the castle pawns. This can only speed up Black counterplay (if it develops).

This is a seriously weakening move that offers Black counterchances, however slender. I might have been inclined to play 15.f3 here (or, more likely, a move or two later). Strenngthening the light squares in this way has the consequence of weakening the dark, but these won't be easy for Black to come at. 15.Rd1 and 16.Rac1 are aggressive moves that reinforce White's strengths.

15...f3!? looks interesting. 16. Bc4, h5!? and it's showtime, especially if White is in time trouble. This is why 15.g4?! was not a good move.

I liked 15...f3 as well. It would have been a bone in White's throat for some time, and the follow-up ...h5 introduces some interesting possibilities. 15...f3 16.Bc4 h5!? 17.gxh5 Bxh3 18.Re1 g4 (18.Bxg5 Bxf1 19.Rxf1 or 19.Kxf1 doesn't look good enough...).

I like 16. Kg2 here, consolidating f3 and allowing Rh1 in the future if necessary. 16. Bf3!? is also interesting, Bishop here influences 2 diagonals, preventing h5 and holding d5.

My method was an asterisk after the move - again to signal what I considered a critical point in the game. (I used to write down times, too. Is that legit?). It's actually quite hard for Black to find moves here. However, the half-pin earlier mentioned having vanished with 17.Rfd1, White was threatening 18.dxc6 Rxd6 19.Bxf4 ... i.e. exploiting the half-pin on the b-file, and the masked battery on the d-. As Black, I would have liked to play ...Rh6. getting the Q off the critical file will permit the rook move later.

Well he is right   especially considering that players are not allowed to do that (scoresheets should contain only moves, times on the clock and draw offers). I usually mark "critical positions" during the game by circling the move number. This is not against the rules and is useful for my later analysis. Besides all that, you are wrong and 17...Qc7 is not a bad move at all   Edit: yes writing time after individual moves is legit and I do it all the time.

18...Rh6 (preventing h4 and clearing f6) or 18...h5 with 19...Rh6 look OK. The queenside is stable at the moment so 18...b5 is not necessary. 18...b6 is maybe also better, keeping the option of later b5 but retaining option of bc5.

This is very bad. Now b pawn is actually pinned to the "developed" Bishop. Bishop was OK on c8, at least it was not in the way of other pieces   since White is threatening dc6 with Bb5, Black can move 19...b4; 20.Na4, Rh6 (clearing f6 for other pieces); 21. Kg2, Bd7 with some improvement of the position (White slightly better as usual).

Obviously for the last few moves Black had this idea in mind all along. I am not convinced it was worth it. 20...Rh6 is again better. Rook still defends the sixth rank but doesn't clog up the f6 square.

Agreed. Black has induced some weaknes on White's K-side, however difficult to exploit, and ought to be looking to press home whatever advantage he has obtained there. Placed on b8 and c8, the Q-side pieces were doing useful work, and the B was still capable of being transferred quickly to the K-flank when the opportunity arose. Black hasn't improved his Q-side and central position very much (if he has at all!), and White still has reserves to draw on.

22...Rh6  

I don't think even 22...Rh6 will help any more, not after 23.Kg2. Black looks pretty thoroughly busted it seems to me.

Ordinarily in this kind of game, White isn't interested in locking down the Q-side in this way, but it makes a lot of sense in this particular position. The threat of busting through the centre remains of serious concern to Black, and on the K-side, it is White that has the serious breakthrough chances with a timely h3-h4 lurking in the background. Finally, White has retained a key to reopening the Q-side should the need arise: a3-a4. Black has no play whatever.

Now it is just horrible, 23...Nc7 is better, or 23...Rh6 (but this move is obviously beyond Black's vision).

Maybe Black is hoping to open the h-file with ...h7-h5.

Black lost a few moves in a difficult position with questionable manoeuvring, and now he is losing the game.