How should I do puzzles? • page 2/2 • General Chess Discussion • lichess.org

Don't do random puzzles.
Do endgame puzzles, tactics, and themed midgame from your preferred openings.
25% puzzles 25% study your games and gmaster games 40% training nonrated games and 10% of your time on "rated" games. Find 2 to 4 people to play nonrated games with to improve.

Play your rated games at end of week after learning. Once you're feeling stronger play more rated games.

Don't do random puzzles. Do endgame puzzles, tactics, and themed midgame from your preferred openings. 25% puzzles 25% study your games and gmaster games 40% training nonrated games and 10% of your time on "rated" games. Find 2 to 4 people to play nonrated games with to improve. Play your rated games at end of week after learning. Once you're feeling stronger play more rated games.

Overcooker

#12

Thanks everyone for the advice. My plan for now is:

"Complete" the practice section, at least to familiarize with the concepts
Themed puzzles as long as necessary then random or a variation, and not spending 20 minutes on each puzzle
Practice opening principles, controlling the center etc. and later learn a couple openings
Play at least two real time games a week, more if I have more time, at least 15+ minutes games
Probably correspondence too

I don't know if I'm missing something important. About studying games, I don't really know how to do that on my own right now. I know about the studies section, but because it doesn't really have a structure, I don't know what to look for.

@Monist said in #10:

Actually, it is clear what you need to do in random mix puzzles.
[...]
But if it remains too baffling do some practice puzzles with clear goals and motifs first and then come back to these "open ended" puzzles.

I like the challenge of random puzzles and I agree that you should always be prepared for everything, but when I get puzzles that I can't understand I feel like I'm just doing guesswork and wasting time without learning much. Sometimes I've had to trade pieces but I still don't know why I had to. Like this puzzle, where I apparently even guessed the right solution:

https://lichess.org/training/SQlYD

All I know is that the computer says that's the best move, but I don't know why. I don't know whether I missed something or I just can't calculate too far ahead.

Thanks everyone for the advice. My plan for now is: - "Complete" the practice section, at least to familiarize with the concepts - Themed puzzles as long as necessary then random or a variation, and not spending 20 minutes on each puzzle - Practice opening principles, controlling the center etc. and later learn a couple openings - Play at least two real time games a week, more if I have more time, at least 15+ minutes games - Probably correspondence too I don't know if I'm missing something important. About studying games, I don't really know how to do that on my own right now. I know about the studies section, but because it doesn't really have a structure, I don't know what to look for. @Monist said in #10: > Actually, it is clear what you need to do in random mix puzzles. > [...] > But if it remains too baffling do some practice puzzles with clear goals and motifs first and then come back to these "open ended" puzzles. I like the challenge of random puzzles and I agree that you should always be prepared for everything, but when I get puzzles that I can't understand I feel like I'm just doing guesswork and wasting time without learning much. Sometimes I've had to trade pieces but I still don't know why I had to. Like this puzzle, where I apparently even guessed the right solution: https://lichess.org/training/SQlYD All I know is that the computer says that's the best move, but I don't know why. I don't know whether I missed something or I just can't calculate too far ahead.

mcgoves

#13

@Overcooker said in #6:

I'd like to add another question though: is it really better to do few puzzles correctly or would it be better to do a lot but failing more?

I find the best way to do puzzles is to treat it like it's in a newspaper or a book. Really calculate the whole thing, and be confident of the full solution before you enter your first move. Otherwise, it's really just a guess, and that trains nothing.

I often spend several minutes on a puzzle. If it gets much past five minutes, I may give up and move on. If several of those happen in a row, I'll stop for a while, or reduce difficulty. When I'm working at the proper difficulty, it's fairly rare to solve one in under 30-40 seconds.

If I miss it, I'll analyze. If I can't figure out the analysis (even with the engine helping), then the puzzle is too hard for me.

If I don't have the patience for all that and just start guessing quickly, it's past time to stop. Or not ... sometimes that's fun for its own reasons.

@Overcooker said in #6: > I'd like to add another question though: is it really better to do few puzzles correctly or would it be better to do a lot but failing more? I find the best way to do puzzles is to treat it like it's in a newspaper or a book. Really calculate the whole thing, and be confident of the full solution before you enter your first move. Otherwise, it's really just a guess, and that trains nothing. I often spend several minutes on a puzzle. If it gets much past five minutes, I may give up and move on. If several of those happen in a row, I'll stop for a while, or reduce difficulty. When I'm working at the proper difficulty, it's fairly rare to solve one in under 30-40 seconds. If I miss it, I'll analyze. If I can't figure out the analysis (even with the engine helping), then the puzzle is too hard for me. If I don't have the patience for all that and just start guessing quickly, it's past time to stop. Or not ... sometimes that's fun for its own reasons.