- Blind mode tutorial
lichess.org
Donate

How should I do puzzles?

#1

When I do random puzzles I fail lots of them because I don't know what's the expected solution. This is a bigger issue when I'm expected to trace pieces and I don't know if I should trade or not. Even when I do trade pieces and that's the accepted solution I still don't know why that is.

I'm thinking about doing more themed puzzles but would it be "cheating" if I already know what I'm supposed to do? And with cheating I mean cheating myself, not the rating. When I do "mate in one" puzzles I do much better than my average but because I only look for mate in one move.

With doing more themed puzzles I mean doing a variety of them, not just sticking to one theme.

When I do random puzzles I fail lots of them because I don't know what's the expected solution. This is a bigger issue when I'm expected to trace pieces and I don't know if I should trade or not. Even when I do trade pieces and that's the accepted solution I still don't know why that is. I'm thinking about doing more themed puzzles but would it be "cheating" if I already know what I'm supposed to do? And with cheating I mean cheating myself, not the rating. When I do "mate in one" puzzles I do much better than my average but because I only look for mate in one move. With doing more themed puzzles I mean doing a variety of them, not just sticking to one theme.
#2

Familiarize yourself with lichess.org/practice first

Familiarize yourself with lichess.org/practice first
#3

Thanks, I'll try that. I already spent some time in the practice section but got stuck in the fork challenges. I've no idea how to force the opponent into a certain position for a fork, at least not in 3-4 moves. But I guess I don't have to do the challenges right away.

Thanks, I'll try that. I already spent some time in the practice section but got stuck in the fork challenges. I've no idea how to force the opponent into a certain position for a fork, at least not in 3-4 moves. But I guess I don't have to do the challenges right away.
#4

Most chess coaches would actually tell you to start by "cheating" In order to learn how something works you need to study that pattern multiple times. So doing 25 puzzle all involving pins will teach you pins and train your brain to look for pins. Mixing them up comes later. Think back to math class. In algebra you would have 10-20 homework problems on a single theme before moving on the the next theme. Then later on the teacher would mix up the themes, but when learning it was one theme at a time. Could you image learning math by randomly selecting problems from the book? Cause that is basically what you are doing using the Lichess puzzles.

Most chess coaches would actually tell you to start by "cheating" In order to learn how something works you need to study that pattern multiple times. So doing 25 puzzle all involving pins will teach you pins and train your brain to look for pins. Mixing them up comes later. Think back to math class. In algebra you would have 10-20 homework problems on a single theme before moving on the the next theme. Then later on the teacher would mix up the themes, but when learning it was one theme at a time. Could you image learning math by randomly selecting problems from the book? Cause that is basically what you are doing using the Lichess puzzles.
#5

I would try something with a set task (ie, mates in 2). That gives you a better idea of what you should be shooting for.

I would try something with a set task (ie, mates in 2). That gives you a better idea of what you should be shooting for.
#6

Yes, that makes sense. I've only low effort things (meaning one click away) like puzzles so far but I think I need a better study plan because I feel like I'm skipping the fundamentals.

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 remember this story when I was studying art, where a ceramics teacher divided the students in two groups: one group had to focus on quantity and make fifty pots, while the other group had to focus on quality make only one very good pot. In the end the students who focused on quantity made the best pots. I do believe this is true because "practice makes perfect," but wouldn't this apply to chess puzzles too? And I'm only mentioning this because I've heard over and over people recommending quality over quantity regarding puzzles. So far I've tried to focus on quality, but I'm wondering if that is slowing me down.

Yes, that makes sense. I've only low effort things (meaning one click away) like puzzles so far but I think I need a better study plan because I feel like I'm skipping the fundamentals. 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 remember this story when I was studying art, where a ceramics teacher divided the students in two groups: one group had to focus on quantity and make fifty pots, while the other group had to focus on quality make only one very good pot. In the end the students who focused on quantity made the best pots. I do believe this is true because "practice makes perfect," but wouldn't this apply to chess puzzles too? And I'm only mentioning this because I've heard over and over people recommending quality over quantity regarding puzzles. So far I've tried to focus on quality, but I'm wondering if that is slowing me down.
#7

Go slow enough for near perfection. And as you get better at finding the correct solution you will get quicker. It is better to get 24/25 mates in one in 15 minutes than get 17/25 in 30 seconds. Once you get good enough to get 24/25 in under a minute, start doing mates in 2. Once again go slow enough for near perfection, if it takes you 2 hours to get 24/25 that's fine keep doing them. Once you can get 24/25 in under 5 minutes start doing mates in 3 .....rince and ....repeat

Getting the wrong answer fast won't teach you anything.

Go slow enough for near perfection. And as you get better at finding the correct solution you will get quicker. It is better to get 24/25 mates in one in 15 minutes than get 17/25 in 30 seconds. Once you get good enough to get 24/25 in under a minute, start doing mates in 2. Once again go slow enough for near perfection, if it takes you 2 hours to get 24/25 that's fine keep doing them. Once you can get 24/25 in under 5 minutes start doing mates in 3 .....rince and ....repeat Getting the wrong answer fast won't teach you anything.
#8

My idea was more like you can see the solution after you fail, so you're still learning. I'm still going to analyze the puzzle solution to the best of my abilities. In theory I should still be able to memorize that pattern.

I don't know if any research has been done in this regard, quickly googling I found this reddit post from ten years ago:

https://www.reddit.com/r/chess/comments/1tj6fk/a_year_of_solving_tactical_puzzels_over_7000_and/

The post's OP went from about 779 to 1347 puzzle rating in one year doing 7142 puzzles in 43 hours, passing 49% of them. That is 21 seconds per puzzle on average.

Although that might be too fast and it's only one case, it still shows improvement. But I don't know if puzzle rating is a good metric, I got a big jump doing only mate in one puzzles.

My idea was more like you can see the solution after you fail, so you're still learning. I'm still going to analyze the puzzle solution to the best of my abilities. In theory I should still be able to memorize that pattern. I don't know if any research has been done in this regard, quickly googling I found this reddit post from ten years ago: https://www.reddit.com/r/chess/comments/1tj6fk/a_year_of_solving_tactical_puzzels_over_7000_and/ The post's OP went from about 779 to 1347 puzzle rating in one year doing 7142 puzzles in 43 hours, passing 49% of them. That is 21 seconds per puzzle on average. Although that might be too fast and it's only one case, it still shows improvement. But I don't know if puzzle rating is a good metric, I got a big jump doing only mate in one puzzles.
#9

Theres plenty to be learned from failing a puzzle. Just be sure to see the solution before moving on.

Theres plenty to be learned from failing a puzzle. Just be sure to see the solution before moving on.
#10

@Overcooker said in #1:

When I do random puzzles I fail lots of them because I don't know what's the expected solution.

Actually, it is clear what you need to do in random mix puzzles. You need to find the winning move sequence. "Winning" here means technically won thereafter. Lichess measures "winning" by its scoring system.

I would suggest these steps.

  1. Try Puzzles Streak. The puzzles basically start easy and get harder: although some early ones can be tricky sometimes.
  2. Approach these puzzles the way you approach moves in a position in a game.
  3. Use the back arrow and then forward arrow to get an idea of the move sequence that led to the position.
  4. Count pawns and pieces to get an idea of any material imbalance. Then you know you need either a checkmate or sufficient material gain to solve the puzzle successfully.

The above steps will situate your mind in the position. Take your time. Puzzle Streak is not a timed race. When it says "white to move" or "black to move" it really means "to checkmate or achieve a decisive advantage".

Follow a full stepwise mental approach. First, check opponent threats. If the opponent has threats then, given the way these puzzles work, there will be a counter move which nullifies the enemy threat and delivers one of your own. If the opponent has no clear threats that you can see then start looking for forcing moves methodically by using 4 steps.

  1. Checks and "Queen Checks"
  2. Captures
  3. Sacrifices leading to checkmate or material gain
  4. Threats (look for all possible threats you can deliver especially double-threats).

A forcing move is a move that the opponent can only reply to in one way or in a very limited number ways, if he doesn't want to face a losing position immediately. It is all about calculated forcing move sequences leading to decisive advantage. 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.

@Overcooker said in #1: > When I do random puzzles I fail lots of them because I don't know what's the expected solution. Actually, it is clear what you need to do in random mix puzzles. You need to find the winning move sequence. "Winning" here means technically won thereafter. Lichess measures "winning" by its scoring system. I would suggest these steps. 1. Try Puzzles Streak. The puzzles basically start easy and get harder: although some early ones can be tricky sometimes. 2. Approach these puzzles the way you approach moves in a position in a game. 3. Use the back arrow and then forward arrow to get an idea of the move sequence that led to the position. 4. Count pawns and pieces to get an idea of any material imbalance. Then you know you need either a checkmate or sufficient material gain to solve the puzzle successfully. The above steps will situate your mind in the position. Take your time. Puzzle Streak is not a timed race. When it says "white to move" or "black to move" it really means "to checkmate or achieve a decisive advantage". Follow a full stepwise mental approach. First, check opponent threats. If the opponent has threats then, given the way these puzzles work, there will be a counter move which nullifies the enemy threat and delivers one of your own. If the opponent has no clear threats that you can see then start looking for forcing moves methodically by using 4 steps. 1. Checks and "Queen Checks" 2. Captures 3. Sacrifices leading to checkmate or material gain 4. Threats (look for all possible threats you can deliver especially double-threats). A forcing move is a move that the opponent can only reply to in one way or in a very limited number ways, if he doesn't want to face a losing position immediately. It is all about calculated forcing move sequences leading to decisive advantage. 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.

This topic has been archived and can no longer be replied to.