Annotated chess game

Black has the bishop pair and the central pawn majority. White's development lead is marginal and likely soon to disappear. Black has a decided edge here in my view.

As I have said before in many games - roll those pawns forward! -

Or even 19.c4 at once, maybe? Also to be considered was 19.Be3, forcing off a pair of bishops. After exchanges on b6 he might have been able to contrive something of interest with his Q-side pawns. However, Black's game would still be preferable.

Since Black would be reluctant to play ...Bc7, perhaps, this has the effect of forcing exchanges. But maybe the move had to be considered, however inconvenient for the moment. Black wants. preferably sooner rather than later, to play ...d5, and ...Bc7 (along with ...Be6 and ...f6) helps prepare this central advance.

There was no reason for white to resolve the tension along the g1-a7 diagonal in black's favor.

ethansiegel makes a very pertinent point, especially as by 20...Qc7, Black has deprived his dark-square bishop of options. There doesn't seem to be much percentage for Black to exchange on e3, but meanwhile, White can use the tension to improve his Q-side position.

Very awkward square for the rook. Perhaps a counter attack against the d file wouold have helped. ie Red1!

alternatively, Bd7 followed by a5 could cause a huge disruption in white's queenside pawn structure.

True, but I prefer the text move. Black has recognised that c4 is as important as c3 here, but one ought not to overlook d5. The move prepares ...f6 followed by ...d5, possibly with ...Bc4 included. White would probably have to see what he could achieve on the K-side to counterbalance the pressure. Ethansiegel's plan, by the way, could be implemented with great effect after this central bind has been established.

A good square for the knight, and it does disrupt Black's plans to some extent. But it commits the knight: it will not be easy to redeploy it anywhere else. With ...e4 or ...f5 Black can pretty much nail the knight down where it stands.

?! This is a huge mistake on your opponent's part. He get's insufficient compensation for the knight (the passed pawn on the c-file is not enough), and if black can exchange off enough major pieces he will have a won endgame.

Probably desperation! Black has built up enormous pressure on White's game, and there was plenty more to come. Rather than be reduced to complete passivity, like the victim of the sacrifice, White looks for counterplay. Sure, White ought not have enough here to save the game, but he has changed its nature into something in which he thinks he might have a chance.

This must be a mistake. It would be sound perhaps if it weren't for the Bc4 which gets in the way. Then the PAWNS WOULD ROOOLLLLLL!

I would've played either g6 or h6 here to avoid the back-rank threats. There's no need for the bishop to move.

... But see Black's next!

I like Rd4! Grabbing the open file and the pieces are tied to the back rank. However, h6 simply rids of this problem. Perhaps worth consideration at the very least...

I can't give an exclam to this. All of white's pieces have been locked down for a long time to the defense of c3. If black plays h7 or g7 and then Rd8, he starts disrupting the white queen's position.

White really should sacrifice this pawn and go for complications. He cannot just sit there watching his position fall apart.

The annotator's approach makes a lot of sense here, but there is also a very small, almost insignificant, consideration that will at some point call for attention. Suppose White were able to excange off all the heavy pieces, and, for the cost of all 6 of his own pawns, get rid of Black's 3 K-side pawns? That leaves Black with B and RP vs lone K. If White's K can reach a1 betimes, it's a draw! It is much too early to worry overmuch about this, but it will suggest things for Black to avoid. At this point, I'd probably start looking for ways to disrupt White's defences. My first thoughts are to place the bishop on f5 with a guard pawn on h5. The bishop reaches behind White's defensive lines. Another idea is to play the ...Qc7-c6-a4, possibly followed by ...a5. One thing is for sure, Black is at liberty to try this or that approach without much fear that White can achieve much. Clearly White's sacrifice hasn't worked to imrove his prospects on the board, but its psychological effect seems to have been considerable.

White may not realize he is trying to hold a position where he's down -- Rd3 would've been much better.