Lockdown Correspondence Chess

AM playing on board three for England in semi-final group four of the Carlos Flores Gutiérrez Team Tournament at the International Correspondence Chess Federation.
My first four games to finish were drawn. The fifth to finish can be seen at https://beauchess.blogspot.com/2020/10/morphing-french-xvi.html. This is the sixth game to finish.
Spanton (England - 2260) - Serge Maurer (Canada - IM2221)
Maróczy Bind
1.e4 c5 2.Nf3 e6 3.c4!?
The sixth-most popular move in ChessBase's 2020 Mega database, but it has been played by Magnus Carlsen.
3...Nc6 4.d4 cxd4 5.Nxd4 Nf6 6.Nc3 d6!?
By far the most common choice in Mega20 is 6...Bb4.
7.Be2 Be7 8.0-0 0-0 9.Be3 Bd7 10.Rc1 Nxd4 11.Qxd4 Bc6 12.Rfd1 Qb8 13.b4
An alternative plan, successfully played in Petr Velička (2449) - Jan Votava (2511), Czech Championship (Lázně Bohdaneč) 1999, starts with 13.f4. That game continued 13...a6 14.a4 a5!? 15.Bf3 Nd7 16.Qd3 Rd8 17.b3 b6 18.Nb5, when the analysis engines Stockfish12 and Komodo11.01  reckon Black could have equalised with the natural-looking 18...Nc5. Instead the game saw 18...e5?! 19.f5, after which Black had no easy way of breaking the bind (1-0, 42 moves).
13...b6 14.Qd3!?
Komodo11.01 prefers the known move 14.f4, while Stockfish12 likes 14.f3 which, as with the text, does not appear in Mega20. The idea of 14.Qd3!? is to vacate d4 for the white dark-square bishop, which can put pressure on the black kingside as well as, indirectly, on the d6 pawn.
14...Rd8 15.Bd4 a6 16.a4 Nd7 17.Qg3 Bf6 18.Bxf6 Nxf6 19.Qe3
Black has swopped his inactive bishop for its active white counterpart. That gives him the better surviving bishop, but he is sensitive on the dark squares.
19...Nd7 20.Qd4 Nf6
I am not sure what SM hoped to achieve by his knight-dance, but note that Black was not ready for the thematic pawn-break ...d5, eg 19...d5? 20.cxd5 exd5 21.e5 Ne8 22.b5 axb5 23.axb5 Bb7 24.Qxb6.
21.f4 h6 22.Qe3 Nd7 23.Qd4 Nf6 24.h4
I was not interested in a draw.
24...e5!?
Trying to give himself some space before White opens lines on the kingside with g4 etc.
25.fxe5 dxe5 26.Qxd8+ Qxd8 27.Rxd8+ Rxd8 28.Nd5
There is no longer any question of a mating attack on the black king, but now White has the much more mobile pawn-majority.
28...Bxa4
Not 28...Nxe4?? 29.Ne7+ and 30.Nxc6.
29.Nxb6 Bc6 30.Nd5 Kf8 31.Bd3 Rb8 32.Ra1 Bb7?!
White gets two connected passed pawns after 32...Bxd5 33.exd5, but this might be the better option.
33.Ra4
White's bishop makes a poor impression, but it is doing a useful job on d3 defending e4 and c4. Meanwhile the white knight is very strong, and Black does not want to capture it because White gets connected passed pawns after the recapture exd5.
33...Ke8 34.Kf2
Both kings enter play, but White's has a more important role to play.
34...Kd8
34...Kd7 is not really an improvement as Black effectively has to wait on White as there is no counterplay in the position.
35.Ke3 Kd7 36.Kd2 Rd8 37.Kc3 Re8 38.Ra1 Rg8 39.Rf1 Ke6 40.Kb3 Bc6 41.Ra1 Bb7
The bishop is forced back as 41...Ra8?? would be a gross blunder thanks to 42.Nc7+. But now the white king activates further.
42.Ka4 Kd7 43.Ka5 Rd8 44.Bc2 Rb8 45.Nxf6+?!
Probably not the best. I somehow thought we had a position without the move 44.Bc2, in which case Nxf6+ probably is the best move. Fortunately for me, White's advantage is so large - three white pawns on the kingside are basically holding up four black ones - that the outcome should not be affected.
45...gxf6 46.Rf1 Ke7 47.c5 Rg8 48.Kb6 Bc8 49.Bd3

Position after 49.Bd3 - who stands better and by how much?

*****

*****

*****

*****

*****
I am sure this is a position a lot of players would look at and be unsure who has the advantage, or at least be unsure as to how big that advantage is. Engines, however, have no such difficulty - they reckon White is winning, and winning easily. Stockfish12, for example, has White the equivalent of more than 6.5 pawns ahead, while Komodo11.01 has White more than 4.5 pawns ahead. Give the engines longer, and the evaluations rise rather than fall.
The game finished:
49...f5!? 50.exf5 Rg4 51.f6+ Ke8 52.Kc7 Rd4 53.Kxc8 Rxd3 54.c6 1-0

I have been unable to pinpoint exactly where Black lost this game. It seems his game gradually went downhill without any glaring mistakes.

One thing I have noticed with the Maróczy Bind is that, at my level at least, it is easier for White to keep control of the position in correspondence chess rather than over the board.

I suspect this may be because the bind takes time to set up and maintain, which gives Black active piece-play that humans find harder to handle than engines do.

Conversely, Black needs a plan against the bind rather than just relying on tactics, and failure to come up with a suitable plan will be punished more efficiently in correspondence chess.