- Blind mode tutorial
lichess.org
Donate

Calculation Problems

#1

It strikes me that one of the weakest areas of my game is calculation, not necessarily seeing tactics, just calculating quickly and accurately. A lot of the time I play on autopilot and some days I have games were I feel like I see the board very clearly, and I play well, and other times my autopilot is... well pretty bad.

My tentative goal is to try to get to 2700 in puzzles (I've bounced between about 2400-2500 for ~400 puzzles) but it seems to me that that might not be the best way to solve the problem.

If anyone has any suggestions they would be appreciated!

It strikes me that one of the weakest areas of my game is calculation, not necessarily seeing tactics, just calculating quickly and accurately. A lot of the time I play on autopilot and some days I have games were I feel like I see the board very clearly, and I play well, and other times my autopilot is... well pretty bad. My tentative goal is to try to get to 2700 in puzzles (I've bounced between about 2400-2500 for ~400 puzzles) but it seems to me that that might not be the best way to solve the problem. If anyone has any suggestions they would be appreciated!
#2

If somebody posts “play long games” one more time I’m going to scream. I’d say you just need to work on skyrocketing your puzzles. I’m 2600 here and on chess.com. I’d help you work on that.

If somebody posts “play long games” one more time I’m going to scream. I’d say you just need to work on skyrocketing your puzzles. I’m 2600 here and on chess.com. I’d help you work on that.
#3

Hello dear @Politics_Matter!

Your post sounds like you are seeking for an advice how to improve your puzzle rating and not how to improve your practical chess. To get better at puzzles = do more puzzles. As in every life/chess area, practice makes it perfect. :-)

I think you cannot compare calculation while solving puzzles to playing a game yourself.
In puzzles you know that there is something decisive.
In real game you dont have this hint and there is a big difference between training calculation and training tactics.
Solving tactics depends on recodnizing patterns, training calculation depends on the ability to think far ahead without confusing yourself or getting lost in long random variations, which are most likely not even going to happen.

In general, calculation starts with knowing when to calculate.
Trying to calculate everything, after each move, tires your brain and you end up making moves that you havent calculated properly in any case!
In most positions you dont even need to calculate a lot of variations, it is rather "boring" positional thinking of improving the placement of pieces...
There is a recommendation to see for checks, threats, captures (forced moves) first.
Better not deep but correct - and the most important thing is to calculate as far as you are able to visualize and evaluate the position you are aiming to get after your intended line.
Happy chessing and all best :-)

PS: Megadoggah - wanna hear you scream - so yeah, go play long games! :D

Hello dear @Politics_Matter! Your post sounds like you are seeking for an advice how to improve your puzzle rating and not how to improve your practical chess. To get better at puzzles = do more puzzles. As in every life/chess area, practice makes it perfect. :-) I think you cannot compare calculation while solving puzzles to playing a game yourself. In puzzles you know that there is something decisive. In real game you dont have this hint and there is a big difference between training calculation and training tactics. Solving tactics depends on recodnizing patterns, training calculation depends on the ability to think far ahead without confusing yourself or getting lost in long random variations, which are most likely not even going to happen. In general, calculation starts with knowing when to calculate. Trying to calculate everything, after each move, tires your brain and you end up making moves that you havent calculated properly in any case! In most positions you dont even need to calculate a lot of variations, it is rather "boring" positional thinking of improving the placement of pieces... There is a recommendation to see for checks, threats, captures (forced moves) first. Better not deep but correct - and the most important thing is to calculate as far as you are able to visualize and evaluate the position you are aiming to get after your intended line. Happy chessing and all best :-) PS: Megadoggah - wanna hear you scream - so yeah, go play long games! :D
#4

#1
"some days I have games were I feel like I see the board very clearly, and I play well, and other times my autopilot is... well pretty bad"
We all have good days and bad days.
Indeed in a real game nobody tells us if there is a tactic for us or even more importantly for our opponent. So in a real game you have to assume there is a tactic and only be reassured if you did not find any tactic. So in a real game you have to play a move after more thinking time than you use to solve a puzzle, where you know that there is a tactic and for which side.
A good book is "Think like a grandmaster" by Kotov, where he explains how to calculate.
A good training method is to take a grandmaster game, a chess set, a sheet of paper, a chess clock, and a notepad. Set the chess clock at your preferred time control. Cover the game with the sheet of paper. Say you play white. Uncover the first few moves and play them. Now start your chess clock and think about the next white move. Write down what variantions you think of. Now uncover one full move and play the actual white and black grandmaster moves. Continue. Later compare your notes with grandmaster annotations or engine analysis.

#1 "some days I have games were I feel like I see the board very clearly, and I play well, and other times my autopilot is... well pretty bad" We all have good days and bad days. Indeed in a real game nobody tells us if there is a tactic for us or even more importantly for our opponent. So in a real game you have to assume there is a tactic and only be reassured if you did not find any tactic. So in a real game you have to play a move after more thinking time than you use to solve a puzzle, where you know that there is a tactic and for which side. A good book is "Think like a grandmaster" by Kotov, where he explains how to calculate. A good training method is to take a grandmaster game, a chess set, a sheet of paper, a chess clock, and a notepad. Set the chess clock at your preferred time control. Cover the game with the sheet of paper. Say you play white. Uncover the first few moves and play them. Now start your chess clock and think about the next white move. Write down what variantions you think of. Now uncover one full move and play the actual white and black grandmaster moves. Continue. Later compare your notes with grandmaster annotations or engine analysis.
#5

I pretty much agree with what @tpr writes. Solving puzzles help you to get a better sense for any tactics that are hidden in a position. But in order to not cheat on yourself you have to do them within a certain time. If we apply Kotov's method to puzzles we should at least solve them with a chess clock running as well, giving us ourselves, say, 30 minutes for the complete puzzle sequence as 30 minutes is a typical time frame that one would spend for a critical position under tournament conditions.

I pretty much agree with what @tpr writes. Solving puzzles help you to get a better sense for any tactics that are hidden in a position. But in order to not cheat on yourself you have to do them within a certain time. If we apply Kotov's method to puzzles we should at least solve them with a chess clock running as well, giving us ourselves, say, 30 minutes for the complete puzzle sequence as 30 minutes is a typical time frame that one would spend for a critical position under tournament conditions.
#8

I used to train with a book of "tactics". Pearls of Azerbaijan:
https://chess.co.uk/products/pearls-of-azerbaijan-djakahngir-agaragimov

It is not about really tactics it is about pure calculation and something called intuition indeed, because without it instead calculation would be very hard , because we are not aware about what to calculate, or when, and I think that one of the best ways of improving calculation is "setting positions", as Jacob Aagard said in his books. Excelling at chess Calculation and "Grandmaster Preparation: Calculation", those books are a magic formula to improve. Actually what I think that improves your calculation (without counting your mood) is constant and daily training, not matter how you train, if is more or less correct. There are thousand of books, like The Woodpecker Method that heps a lot to improve your tactics, your pattern recognition as your calculation abilities. Actually calculation is not always related to tactics, calculations is required only when you must to solve an issue a lot of moves ahead: Like checkmate an exposed king or simply improve your position because you're a pawn down or whatever.
I would say that pattern recognition helps your calculation the 80% of the times (to those experts I would say almost 100% of the times), because your brain know how to do and solve the positions from previous knowledge that you learned studying or training with books and analyzed games.

From someone that feels that can calculate very well for his level, its all about train your skills and forget about puzzle rating or even blitz-rapid rating, because it becomes really frustrating when you are able to play good games or solve hards puzzles but it is not reflected in your rating, results comes with time, so, best regardings and good luck.

P.S: Sorry for bad my english or my bad grammar.

I used to train with a book of "tactics". Pearls of Azerbaijan: https://chess.co.uk/products/pearls-of-azerbaijan-djakahngir-agaragimov It is not about really tactics it is about pure calculation and something called intuition indeed, because without it instead calculation would be very hard , because we are not aware about what to calculate, or when, and I think that one of the best ways of improving calculation is "setting positions", as Jacob Aagard said in his books. Excelling at chess Calculation and "Grandmaster Preparation: Calculation", those books are a magic formula to improve. Actually what I think that improves your calculation (without counting your mood) is constant and daily training, not matter how you train, if is more or less correct. There are thousand of books, like The Woodpecker Method that heps a lot to improve your tactics, your pattern recognition as your calculation abilities. Actually calculation is not always related to tactics, calculations is required only when you must to solve an issue a lot of moves ahead: Like checkmate an exposed king or simply improve your position because you're a pawn down or whatever. I would say that pattern recognition helps your calculation the 80% of the times (to those experts I would say almost 100% of the times), because your brain know how to do and solve the positions from previous knowledge that you learned studying or training with books and analyzed games. From someone that feels that can calculate very well for his level, its all about train your skills and forget about puzzle rating or even blitz-rapid rating, because it becomes really frustrating when you are able to play good games or solve hards puzzles but it is not reflected in your rating, results comes with time, so, best regardings and good luck. P.S: Sorry for bad my english or my bad grammar.
#10

Since Kotov is brought a second time in a few days, there is a detailed review of 'think like a grandmaster' on chess.com, by GM Bryan Smith. He gives his opinion on the pros and cons of that book and the main controversial aspect of it (The tree)
I am unsure we can post links like this here, but you find it easily with google, search : kotov think like a grand master review

Since Kotov is brought a second time in a few days, there is a detailed review of 'think like a grandmaster' on chess.com, by GM Bryan Smith. He gives his opinion on the pros and cons of that book and the main controversial aspect of it (The tree) I am unsure we can post links like this here, but you find it easily with google, search : kotov think like a grand master review

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