I much dislike the common social media phrase: “Shared without comment”, because that is a comment, and usually implied agreement. IM Kavutskiy thought GM Gelfand “roasted half of #chesstwitter” in his Technical Decision Making in Chess. Rather than share without comment, I agree: Yeah, this is a jeer:

Black will win in the long run. But again, look at the position after the time control. This got far too close!

Let me underline why 42… Nd4? is the mistake. Even if we are able to prove that Black is still winning and the move made in the game does not change the objective evaluation of the game, we should notice the length of the variations and just how many options there are along the way for Black to go wrong. | am not like one of those players that run to the computer after the game, look what evaluation it gives and then go on Twitter and say I made a mistake on move 25 and if I had made this or that move… A real mistake is determined when we look at how the game did and could have proceeded and compare the complexity. We must remember that an advantage is rarely blundered away, but slowly squandered by a series of imprecise moves. Something the engine, which has no way of telling how difficult a move is to find, can tell us little about.

To illustrate how difficult it is to assess the nuances once I played 42… Nd4?, I can tell you that Jacob and I have a recording where we both argue, after analyzing the game together, that 43…b5 would have been a mistake and 43…f5 was actually a good move. Now that our opinion has turned 180 degrees we should not pretend that this is done with absolute certainty.