What do you think I need to improve to be 2200 in lichess (classical & rapid)

@Usmagic said in #3:

As your puzzle rating is not astronomical.
Focus more on them. You did 173 today... quite a lot. and got around 50% of them wrong.

I see this criticism a lot. If you play enough games you will have a 50% win rate. If you do enough puzzles you will have a 50% success rate. That is just how the system works.

@Jemwil41 said in #9:

how do i know my strengths and weakneses

For the weaknesses, you can analyze your games (mostly the losses) to find in which point you failed or felt you were losing. It is easier to do it backwards, from the end of the game. Discard obvious tactical blunders that don't provide any new knowledge to you.

Usually (if you are careful and lucky), you'll find a few patterns, some common traits in said positions (like "poor pawn structure", "little active pieces", "premature attacks",...) and could trace back a way to avoid them.

@sammgus said in #11:

If you do enough puzzles you will have a 50% success rate. That is just how the system works.

If I understand correctly, in the default settings the system will assign to you ~50% puzzles over your rating, ~50% below, isn't it?

So having consistently a 51% success rate would mean improving a lot, at least until you approach the puzzles upper limit.

@sammgus said in #11:

I see this criticism a lot. If you play enough games you will have a 50% win rate. If you do enough puzzles you will have a 50% success rate. That is just how the system works.

That's the default setting, but I don't think it's a good one.

You learn a lot more if you calculate long enough until you see the solution. When you do that with default settings you need to calculate longer and longer until you are unable to solve them at all. They just get to hard to do that.

So my suggestion was to set the difficulty on "easier" and look until you find the correct solution. The occasional miss will reduce your rating significantly and again you can start solving the puzzles with good confidence.

@aescht said in #14:

That's the default setting, but I don't think it's a good one.

You learn a lot more if you calculate long enough until you see the solution.

That's not a setting, that is just the result if success makes the puzzles (or the opponents) harder, and failure makes them weaker. Your success rate will tend toward 50%.

So if your success rate is at 50% and you are at some rating X, then you know that you have made an improvement when that rating increases. Making the puzzles easier to get a higher success rate doesn't make much sense, because it is the harder puzzles that we need to figure out to improve. Just like playing weaker players is not the way to get better.

@sammgus said in #15:

That's not a setting, that is just the result if success makes the puzzles (or the opponents) harder, and failure makes them weaker. Your success rate will tend toward 50%.

That's due to the default setting of "normal" difficulty.

You don't want 50% success rate in puzzles if you want to improve efficiently. What are you doing? Guessing the solution? I mean how can you fail 50%? That just makes no sense. Puzzles are generally not there to be guessed, but solved, unless you just play around and have fun.

@aescht said in #16:

You don't want 50% success rate in puzzles if you want to improve efficiently. What are you doing? Guessing the solution? I mean how can you fail 50%? That just makes no sense. Puzzles are generally not there to be guessed, but solved, unless you just play around and have fun.

If you could solve them all (or even 80%+) then obviously you are doing puzzles that are too easy. The system makes the puzzles harder to challenge your problem solving. Eventually you will make mistakes in calculations. That is not a bad thing, and it doesn't mean you're guessing at the solution. If you think it is better to do simple puzzles and get them all right, than do challenging puzzles and fail half of them, then we just have different learning philosophies.

@sammgus said in #17:

If you could solve them all (or even 80%+) then obviously you are doing puzzles that are too easy. The system makes the puzzles harder to challenge your problem solving. Eventually you will make mistakes in calculations. That is not a bad thing, and it doesn't mean you're guessing at the solution. If you think it is better to do simple puzzles and get them all right, than do challenging puzzles and fail half of them, then we just have different learning philosophies.

Yes, we are getting to understand each other!

If I do puzzles to improve (which I haven't done in quite some time as I don't feel the urge to improve any more) they are very hard for me, don't get misled by the human categorization "easier". But I just take 10 or 20 minutes to get them right. Occasionally I fail them because I oversee something. But getting 50% (!!!!!!!!!!!!!!!!) right seems very much like you are guessing solutions.

If I keep the puzzles on "normal" with that philosophy I will end up spending 3+ hours to solve a puzzle or I just capitulate because I cannot figure a winning sequence with certainty at all, which is unpractical and doesn't seem an efficient use of my time.

I looked in your chess insights... @Jemwil41
Any opening above 50% is a pass. Anything below is not.
In the Italian game, you have two phases that don't pass. The other openings names below only have a phase that you were weak in. I would start with the following openings and links.

Yusupov-Rubinstein System 26.70% 27 Endgame https://lichess.org/training/Yusupov-Rubinstein_System
Bishop's Opening 31.00% 51 Middlegame https://lichess.org/training/Bishops_Opening
Italian Game 38.70% 76 Endgame https://lichess.org/training/Italian_Game
Scandinavian Defense 39.20% 29 Endgame
Zukertort Opening 44.10% 31 Endgame
Caro-Kann Defense 45.10% 200 Middlegame
Four Knights Game 47.50% 186 Middlegame
Italian Game 49.40% 189 Middlegame
Sicilian Defense 49.90% 304 Middlegame

@aescht said in #18:

But getting 50% (!!!!!!!!!!!!!!!!) right seems very much like you are guessing solutions.

If there are an average of 3 reasonable moves in each position and an average of 3 moves per puzzle, then you have roughly a 4% chance of guessing the solution. If you are getting difficult puzzles right 50% of the time then you are calculating quite well!

Occasionally I fail them because I oversee something.

And once you have done enough puzzles then you will oversee something ~50% of the time... Unless you hit the difficultly limit and are still solving strongly.