Is analysis of lost games really useful?

No matter how many times I analyze, I can't seem to reduce my mistakes.

At the beginner level analyzing games can help with openings so regardless of wether you win or loose just have a look at what else you could have played in the first ten moves. Secondly, you can try identifying patterns of where you're going wrong. I looked through some of your games and the main thing holding you back seems to be calculation. For example, you're not directly hanging pieces but end up doing so sometimes after a series of captures. 

@1

"No matter how many times I analyze, I can't seem to reduce my mistakes"
++ Analyse your lost games thoroughly, not superficially.
Preferably do that right after you lost, so the imprint is still fresh.
Identify your decisive mistake and then investigate what thinking led you to play the mistake.

Yes it really helps me as I avoid those mistakes. What you should be looking for is how you lost and how you can use the analyse to make better moves. You could be looking for the wrong things.

Why would you analysis won games? I only analysis if my opponent plays an interesting opening or destroys me in the opening. Or if I think there were some interesting tactics

After analyzing try to know why the computer says a move was a mistake.

For me, exchanging my knight and bishop for rook and one pawns is something I will confidently do. However I noticed that the computer doesn't like it. And in my endgames where the opponent has knight and bishop against my rook, it is really difficult to play.

When you analyze your games try to know why some of the thing you do naturally are mistakes.

For the part of making mistakes in your games, just don't beat yourself too much, mistakes is part of the game, but those mistakes you do frequently is something you should try to cut down.

Thank you for your comments.
This is one of the games I lost.

The analysis engine says "8.Bxf2" is the best.
I don't understand why.

If I play it, instead of gaining my opponent's pawn, I lose my bishop.
This will result in a piece loss.

By forcing my opponent's king to capture my bishop, I can prevent my opponent from castling, so the engine decided it was best?

Others say that what the engine says is not always true.
is that true?

You don't loose a piece if king takes, knight g4 check opens the queen to take the knight so you win a pawn + less pressure on f7

@8

Your decisive mistake was 24...Rb8?? With 24...Nc5 you could have defended and won with your extra piece. Did you consider 24...Nc5? Why did you select 24...Rb8?? How much time did you spend on 24...Rb8?? How much time did you have available then?

21...Bxb5? just helps white. Use your extra piece to defend with 21...Bc5.

19...Nxa6?? blunders a piece. Just use your extra pieces to defend and win 19...Bc6

10...Qg7? is a missed win: 10...Ng4 is a double attack on Ng5 and f2.

8...Qe7? is a missed win: 8...Bxf2+ 9 Kxf2 Ng4+ 10 Kg1 Qxg5 wins a pawn and prevents white from castling, leaving you with an attack on his king.

7...Nc6? is a missed win: 7...d5 8 cxd4 dxc4 9 Qxc4 Qxd4 10 Qxd4 Bxd4 leaves you win a pawn. 5...Nd4 and 7...Nc6? cannot be right.

Indeed, the engine is right: 6...d5 was much stronger than 6...Bc5. The point is 7 exd5 Bf5 and 8...Nxc2+. However, 6...Bc5 is not yet bad. Your later mistakes are much more consequential.

Thanks for the great advice.
After the match, it's very clear why I played such a stupid move.
But it doesn't seem to work when playing.
My brain doesn't work well...

Won games doesn’t mean perfect play. There can be mistakes and areas where a person could’ve done better.

Some GM´s or other people think thats the best way to improve.

GM Jan Gustafson dont think so, he says he never analysed a game. Also not in his childhood.

Everyone has to figure out for themselves how they want to improve. He thinks it's important to engage intensively with chess and it should be an occupation that also gives you pleasure.
The brain then already connects the things as it is necessary.
I think the most important thing is that we "feed" our brain with chess positions and patterns. If analysis is not fun, why do it as a hobby player?
If you prefer to study old master games, then you should do it. Or tasks on middlegame positions, or just analyze any positions from games. The most important thing is that you do something and that you do it with such pleasure and passion that you can really memorize it.

Translated with www.DeepL.com/Translator (free version)

If the analysis of an oven leads you to learn that putting your hand on leads to pain and injury when it is hot, yet you still do it again and again regardless, whose fault is it? The analysis, or the analyser?

Analyses help you discover where and why you were wrong. Next, you have to actively correct your course and stop doing it. You will slip up still but, at least, you'll be on a right track.

Post-game analysis isn't just about looking at what the engine says.

It's also about (and perhaps more importantly about): looking critically at your ideas and asking yourself if there were other options that you could have tried.

I heard a Grandmaster once say that most players analyze their games incorrectly. He said that analyzing a single game should take a very long time. Half an hour or more. And he also recommended one take notes, with pencil and paper, about the game, about each move, about their thoughts and emotions during each move, and about the reasons they made each decision.

This kind of analysis is far beyond what most players care to invest. But it's the kind of deep analysis and introspection that can lead to the most insights ...

Hi