Question about improving to players rated 1800+ here in blitz or rapid


Tactics, slower games, positional study. The first two are self-explanatory, the positional study is analysis of your own games as well as some master games - don't use an engine for your initial analysis. I don't have a strict repertoire. I play the Scotch, b3, the English, Colle-Zukertort, etc as white. As black it's more limited: QGD, Budapest, Sicilian Kan. Sometimes use a Queen's Indian typ set up. Tactics is really the most important followed closely by game analysis. The answers are always the same to this question but people keep asking.

Tactics, slower games, positional study. The first two are self-explanatory, the positional study is analysis of your own games as well as some master games - don't use an engine for your initial analysis. I don't have a strict repertoire. I play the Scotch, b3, the English, Colle-Zukertort, etc as white. As black it's more limited: QGD, Budapest, Sicilian Kan. Sometimes use a Queen's Indian typ set up. Tactics is really the most important followed closely by game analysis. The answers are always the same to this question but people keep asking.
You are right, people keep asking! Although there is something that I rarely see - when people realised that they have to do this or that to make progress and what was this exactly. I guess that some people never had a plateau and keep impressed gradually. But about people who stood at 1500 for too long and suddenly did a jump in their rating?

Playing at a club and/or playing at least 1 "long" game each week was helpful (at least G/60). If you have no humans for this then consider playing an engine.
Studying my weakest area was a big help.
Choosing to stop screwing around with silly openings and try an all classical repertoire helped.
1. Did you usually go through your games to check if you played the opening according to your repertoire? (if you follow a repertoire at all!). Did you do this by comparing with a physical book, database or did you just use the opening explorer?
Yes. I often compared the opening moves to a database, even if it was a blitz or bullet game.
2. Did you try to understand everything when you used the engine in the analysis? Or did you pay only attention to mistakes and blunders, ignoring inaccuracies?
Any position where I was confused during the game, or unsure, or thought the choice was interesting, and any position where I played a clearly bad move according to the engine.
3. Any general advice you can give to us? I know that most authors suggest to play rapid games or longer formats. However, will you improve automatically your blitz by doing so? (Not that this is my first goal, but I am just curious about your experience).
Get a well regarded book on strategy or the endgame and play ever every line of analysis in the book and take notes in a notebook. For example Pachman's Modern Chess Strategy, Maizelis Soviet Chess Primer, Soltis Pawn Structure Chess, Dvoretsky's Endgame Manual, Silman's Endgame Course. Annotated game collections can also be useful for exampe Zurich 1953 or Reti's Masters of the Chessboard.
When you get better at chess in general, your blitz automatically improves, yes.
With the small caveat that if you take a long break from blitz then you'll need some days to get back into it.
Thanks. Yes, I think that playing against an engine is not that bad. I have old chess computers from the 1980s and 1990s, and perhaps I can give a try. Weekly one game, this is feasible.

I don't know, that stuff gets kind of vague and may only be useful to the individual since it's happening only in their head.
It was useful to me when I started clearly defining what my opponent's threat or idea was. When I was 1300 or so I used to just randomly calculate until I found something that looked good, but that wasted a lot of time and energy.
It also helped when I learned to calculate slowly. Very slowly. After each move I'd pause, try my best to visualize the new position, and then carefully choose a move for my opponent. This was useful for me as a 1500-ish player because you know, when you assume a capture, recapture, or check will be played you miss strong in between moves or other ideas. After I broke some of my bad habits like assuming captures I let myself calculate faster again, but for a while I was solving tactics puzzles and calculating very slowly to try and increase my accuracy.
This is insightful! I realised already that it is essential to see in between moves. I recently won a game against a guy rated 700 higher than me in blitz because such a move (chess 960, and there he is underrated). Next game I lost because ...a in between move! It was actually a simple check, and is queen and a rook were hanging, so I was just blind . I also noted that weaker players usually don’t calculate such moves when they assume that I will take material back.
Now some more specific questions.
1. Did you usually go through your games to check if you played the opening according to your repertoire? (if you follow a repertoire at all!). Did you do this by comparing with a physical book, database or did you just use the opening explorer?
2. Did you try to understand everything when you used the engine in the analysis? Or did you pay only attention to mistakes and blunders, ignoring inaccuracies?
3. Any general advice you can give to us? I know that most authors suggest to play rapid games or longer formats. However, will you improve automatically your blitz by doing so? (Not that this is my first goal, but I am just curious about your experience).
Thanks!