Why analyzing losses is mainstream?

Last time we spoke about analyzing our wins. Today let's go back to basics: why is analyzing losses so mainstream? When I asked myself that question, one idea came to mind: when we win, we might have played a perfect game, and then there is nothing to analyze. For example, we may simply deliver Légal's mate. We already know everything about it, so what is there to analyze? But if we lose, we certainly made a mistake somewhere!

Let's dive into the examples!

A well-known position, and I believe 99% of readers know the best move here! White wins a pawn or mates the opponent. Because we have all studied and memorized this, we just reproduce it on the board perfectly.
Now the game that I lost. Recently I played a rapid tournament in Osečina, and in round 5, after scoring well in the first four rounds, I faced Grandmaster Kostić from Serbia with the Black pieces.
White opened with 1.Nf3, and after some moves we reached the following position:

I opted for 8...Ne4, simplifying the position, while in the next position

I opted for 12...Qd5, inviting complications. This is inconsistent: first I chose the more solid line, and now I am complicating things. I'm not sure I knew what I was expecting from this game! More about the moves in the Lichess study below. Here I want you to also pay attention to the non-chess factors, which have a great impact on the game. I believe that after we analyze our losses well, we will have some idea of what went wrong, and later, of course, we will try to improve in that area!

Here are the games as a Lichess study, where you can play through the moves.

First, the famous Légal's mate:

The game with GM Kostić, which we started to analyze: