A subtle detail to understand about this topic is that although it is called "spaced repetition" what makes it effective is NOT the repetition. What solidifies the neural patterns is the retrieval process. This is why testing is a great learning tool. The research shows that the more you struggle to retrieve/recall the information the stronger the neural patterns become and retention will last longer. Time intervals of repetition do matter, and this is where the woodpecker method might lose some effectiveness. The woodpecker method increases volume of exercises and the intervals for recall become shorter. Spaced repetition unlike Woodpecker method is the opposite approach. The intervals for recall increase as your recall becomes stronger. Recalling the patters by increasing the time in-between recall is much better because it INTERRUPTS the forgetting curve, which is what helps store information to long term memory. There is a great book on this topic called "Make it Stick". There are other topics like "interleaving" that I still have not figure out to integrate into my chess, but I'm working on it.
Came here to say exactly this. Woodpecker / Seven Circles is the opposite of spaced repetition. The intervals shorten, and as Axel Smith told Ben in an interview, he didn't retain his tactical gains after a few years of not woodpeckering. Spaced repetition, on the other hand, is defined by expanding intervals with the idea that there's a "forgetting curve" such that the longer you wait on reviewing, the longer term the memory is that you encode the information into--as long as you don't wait so long that you're beyond the threshold of being able to recall the information.
One idea I had for implementing spaced repetition in Chess practice was making interactive Lichess studies and adding a link to them in an Anki flashcard. Probably a good way to review GM games/openings/endgames. I do something similar with LeetCode questions.
@@AFistfulofDynamite There used to be a great chess software that had spaced repetition implemented very well. It was called Chess Position Trainer. I still have it and when I use it I always see results. I do not know if it is still available. But I recommended greatly. Also the software will save only one position at a time, so as you create your opening rep, it will show you transpositions between openings. When you practice with it it will also tell you your recall percentage as well.
It sounds like a lot of people don't quite get the purpose of spaced repetition for tactics. It's not intended to have you memorize solutions to specific tactical problems. It's purpose is to get various tactical patterns/motifs ingrained in your long-term/subconcsious memory so you're more likely to recognize those patterns during a game. I use Chessable to drill the "The Checkmates Patterns Manual" and "1001 Chess Exercises for Beginners" problems. That's a total of 2051 basic tactical positions that are getting ingrained in my brain. This has greatly helped me in spotting tactical possibilities in my games and when working on "Best Move" or calculation problems.
@@chesscomdpruess Have you had any students who were able to develop master-level tactical vision as an adult? My hypothesis is that the people who are able to deeply acquire thousands of tactical patterns without spaced repetition had to do at least some of that when they were young.
I agree with this and has been my experience that memorizing calculation puzzles is a waste of time. Everyone knows what a smothered mate position looks like and how to set it up, but it was only after doing the checkmate manual that I learned many new patterns, like the Greco mate, which was on my radar in my last OTB game, that would've been really hard to think of and set up further away from the ideal position. I see immense value in drilling the courses you mentioned for drilling, and would be near the top of the list for me to drill for tactics.
@@chesscomdpruess Jesse's idea that repeating them makes them come more quickly during play seems intuitive though, because generally speaking speed comes with repetition. Your point about about the tradeoff between repeating things and doing something new is also a good one. We need science to sort this kind of stuff out. Can Kabadayi (FM, cognitive scientist) said a while ago he's considering starting a project to answer questions about chess learning. Let's hope it happens.
22:48 that's only an issue when you have the course set to "key moves" instead of "all moves" - that's a setting you can set for yourself, so you always play from the very first move through the whole variation
I have so many chess books that I haven’t actually read. Spaced repetition with Chessable has not only helped me stick with material, but also drilled it. It’s really helped my digestion of the content
@@oldman-badchess one thing I really like is the flexibility of Chessable courses. Maybe you just want to start with the “short and sweet” version for a taste. Then you buy the course and just start with quickstarter. Then go on to the “priority” lines, then finally you can choose to do the whole course. The level to which you dive is up to user and their time
Yep, same here. I tried several times to get through "100 Endgames You Must Know" but could never get very far before giving up. Chessable made it totally accessible to me. I completed the whole Chessable course in about a week and have been drilling it over a year now.
The default repetition schedule on Chessable is 4hrs, 10 hrs, 1 day, 2.5 days, 1 wk, 2.5 wks, 1.5 months, 4 months and then continues with 4-month spacing. The idea is that as long as you correctly repeat the moves in the given position, you advance to the next spaced repetition time, but if you fail, you fall back to an earlier spaced repetition time. You have the option of changing this schedule if you wish. For my opening and endgame drills, I changed the max repetition time to 2 weeks. For my tactical drills I left the default schedule in place.
Farming points is fun, to see your name at the top of the leaderboard for a given course. I just played against the French for the first time OTB yesterday, after spending a few days drilling lines and was able to produce the 12 moves in my course no problem. I set the interval to every hour, and just drilled it several times a day, and by the end of the week, I was able to play the 90 lines I drilled to 100% or near 100% accuracy consistently. Furthermore, I was also able to continue to play the best moves a bit further thanks to the themes in the position ingrained.
all respects but evey 60 mins i think its mentale health challenging. Also no duty in life, no physiological needs? I know we supposedly should replicate alphazero but....
100% agree with Jesse being dropped into move 15 with out seeing the lead up moves is less helpful. Chessable needs an option or a button to see the lead up moves super fast.
well you can choose to learn and review "All moves" instead of "Key moves". The problem is that you have to manually play the same moves countless times that's very time consuming. Try doing that in an opening like the Marshall is super-boring
I think part of it _might_ depend on the "knack" or "talent" of the student. I remember one of my old classmates who simply attended lectures and did the assigned homework, and he scored at the top of the class. I shared that rank with him, but I spent countless hours in the library every week working additional problems from practically every other textbook I could find on the subject. I exhausted myself to get the results he got with minimal work because he was more "talented", or whatever, than I. Famously, Magnus Carlsen has said he never did much in the way of tactical exercises, yet from what I've heard he's still a very strong chess player. Likely he has a "knack" for the game.
There isn't any scientific research on spaced repetition for chess tactics afaik, but some people who have had success with it recommend adjusting the spaced repetition schedule in Chessable to have MUCH longer intervals between reviews, starting with 7 or 30 days then scaling 2-2.5x after that. The idea behind this schedule is to discourage rote memorization of the puzzle solutions like you would get with Chessable's default schedule, and encourage learning the general patterns instead. It ends up being more like The Woodpecker Method, but it avoids a problem in that method where you can spend too much time on stuff you already know.
Spaced repetition/woodpecker style in general and Chessable in particular: I think it might not work/falls sort for some forms of study: openings and fastest mates (sometimes) if you have the text hidden (like i do), visualization exercises, etc. Also people probably should override book defaults to their preferences (i don't bother). It has flaws but the spacing algorithm is basically the same as language learning apps so it's pretty proven for any straight memorization task. The visualization issue gets at the fundamental issue/con about spaced repetition is what you plan on getting out of it. With openings (I gather) and most endgames (say Silmans book at around 1700+ chapters) the usual goal is either memorization or pattern recognition: that's why spaced repetition seems to be less disputed. In something like tactics: if it is pattern recognition, then it's probably more effective to space out a "single book." But, many do tactics to improve their visualization, calculation and/or evaluation (esp when you get to what Kanneman and Jesse like to call "system 2 thinking") where when "3 books one time" would be better since you are pushed out of being able to lean as heavily on memorization/leverage your lazy and efficient system 1 brain. "Tactics" is in that gray area where it's a bit of all of that, and people with different weaknesses and different goals. Depending on your specific context, you can make a strong argument againt spaced repetition, esp if someone gets caught up on the point system (in Chessable) and doesn't really figure out WHY that move is the right one (which probably happens a lot in opening courses there?).
Woodpecker spaced repetitions start with promlems that help your calculation. The repition though is desgned to help build pattern recognition. Pattern recognition is someting you get less of by doing 3 different books. Lower rated players often need a lot more help building an internal database of patterns.
I agree. Space repetition is probably best for endgames but also checkmating patterns. The only thing that will help improve calculation is more calculation so I think memorizing puzzles is not effective. As for openings, I'm surprised that databases weren't mentioned. I dreaded studying openings in the past. Since I learn to study openings using a database I'm hooked. My only rating jumped 100 points in a two-month period where I spend lots of time building an opening repertoire.
Good point by Kraii on being presented with move #18 and being drilled on how to proceed, without knowing how you got there... That's why I am now playing the french, the classical dutch and just playing 1. e3 for now and trying to have a clue what the hell I'm doing...
This is why I drill with pause at end of variation/line, so that I can go back to the start and see how I got to the line. Just because there is an option to automatically move to the next line doesn't mean you should.
Dang it, I remembered this too late... but I'll still post it because it's so pithy and apropos. Bruce Lee famously said "I do not fear the man who has practiced 10,000 kicks once. but I fear the man who has practiced one kick 10,000 times."
Seems to me that rote learning lines is close to pointless, because in 99.9% of games your opponents will deviate at some stage and even if they don’t, after you’ve sacked 2 pieces, a rook and half a dozen pawns, “with compensation for the sacrificed material” is hardly sufficient information on how to proceed.
One thought I have on woodpeckering calculation sets is you're basically converting a calculation exercise to a memory recall exercise, which might not be ideal. If you're drilling tactical motifs, it seems like it's a bit better (but it doesn't seem to be any better than more unique examples of a given concept).
Yes, the Woodpecker method/spaced repetition method works well to get tactical patterns/motifs ingrained into your long-term memory so that you're more likely to recognize them when playing a game, but it doesn't help with improving your calculation. Tactical pattern recognition and calculation go hand-in-hand but are different skills that need to be trained in different ways.
@@davidfranklin5426go to your course, open it and where the repetition options are, Learn: Everything, Study: All Moves, Review: Whole Variation, Depth: Full Depth. So start your course with those options. If it’s a course you’ve already through. You will have to make the changes and go back and “learn” every line over again and you will “learn” all the beginning moves again.
Yes, I agree. For opening drills I much prefer to start from move 1. You can change this setting by going to the "Chapters" page for the course and in the "Course Settings" along the right-hand side, change the "Review" selection to "Whole Variation".
I agree with Kostya in that I'd prefer if I could drill my opening lines using a physical board and pieces. I've submitted to Chessable a suggestion to implement an interface for e-boards so I could use a physical board when drilling opening lines but who knows if they'll ever do that.
You can do that with at least some of the Chessnut Eboards (Chessnut Air is one of them). Not sure but there should be some other eboards with the same functionality. . Still not the great stuff because the board recognizes "the original flow" but at the moment you start reviewing your mistakes it does not recognize the position because it goes from lets say move 12 to move 5 again and all the pieces are elsewhere, so you would need to arrange the pieces manually to that position for every reviewed mistake
I have a chessable account and every now and then I try to use it. But sorry… I just can’t stand it. The repetition method is so dull and infantilizing. I know it works like a charm for others but it’s just not for me.
Chessable is only half the battle. You used the spaced repetition to get to a position then THAT position is your Tabiya to spar from and fight from. Memorizing and learning their little blurbs isn't enough. You need to train in those final positions for it to be useful, otherwise, the opening doesn't matter bc you don't even have experience in what you're learning.
Spaced repetition isn't an opening specific technique, it's a learning tool, even outside of chess. Ignoring openings: tactics, strategy and endgames are still very significant areas of study that benefit from spaced repetition and were talked about during the video you've just commented on... Maybe rewatching it (or watching it the first time?) may benefit your understanding of spaced repetition in an ironic way :)
A subtle detail to understand about this topic is that although it is called "spaced repetition" what makes it effective is NOT the repetition. What solidifies the neural patterns is the retrieval process. This is why testing is a great learning tool.
The research shows that the more you struggle to retrieve/recall the information the stronger the neural patterns become and retention will last longer.
Time intervals of repetition do matter, and this is where the woodpecker method might lose some effectiveness. The woodpecker method increases volume of exercises and the intervals for recall become shorter. Spaced repetition unlike Woodpecker method is the opposite approach. The intervals for recall increase as your recall becomes stronger. Recalling the patters by increasing the time in-between recall is much better because it INTERRUPTS the forgetting curve, which is what helps store information to long term memory.
There is a great book on this topic called "Make it Stick". There are other topics like "interleaving" that I still have not figure out to integrate into my chess, but I'm working on it.
Thanks! If only one of us had realized to say this on the actual episode!
Came here to say exactly this. Woodpecker / Seven Circles is the opposite of spaced repetition. The intervals shorten, and as Axel Smith told Ben in an interview, he didn't retain his tactical gains after a few years of not woodpeckering. Spaced repetition, on the other hand, is defined by expanding intervals with the idea that there's a "forgetting curve" such that the longer you wait on reviewing, the longer term the memory is that you encode the information into--as long as you don't wait so long that you're beyond the threshold of being able to recall the information.
One idea I had for implementing spaced repetition in Chess practice was making interactive Lichess studies and adding a link to them in an Anki flashcard. Probably a good way to review GM games/openings/endgames. I do something similar with LeetCode questions.
I saw Ben Johnson's podcast on Make it Stick. For me it's a must read.
@@AFistfulofDynamite There used to be a great chess software that had spaced repetition implemented very well. It was called Chess Position Trainer.
I still have it and when I use it I always see results. I do not know if it is still available. But I recommended greatly. Also the software will save only one position at a time, so as you create your opening rep, it will show you transpositions between openings. When you practice with it it will also tell you your recall percentage as well.
I watch this while reviewing my lines on Chessable.
It sounds like a lot of people don't quite get the purpose of spaced repetition for tactics. It's not intended to have you memorize solutions to specific tactical problems. It's purpose is to get various tactical patterns/motifs ingrained in your long-term/subconcsious memory so you're more likely to recognize those patterns during a game. I use Chessable to drill the "The Checkmates Patterns Manual" and "1001 Chess Exercises for Beginners" problems. That's a total of 2051 basic tactical positions that are getting ingrained in my brain. This has greatly helped me in spotting tactical possibilities in my games and when working on "Best Move" or calculation problems.
I understood that; it’s just that in my own personal experience, I don’t forget tactical patterns, so it’s unnecessary.
@@chesscomdpruess Have you had any students who were able to develop master-level tactical vision as an adult? My hypothesis is that the people who are able to deeply acquire thousands of tactical patterns without spaced repetition had to do at least some of that when they were young.
If you mean who had never played chess as a kid, I don’t think I have.@@dsrguru
I agree with this and has been my experience that memorizing calculation puzzles is a waste of time. Everyone knows what a smothered mate position looks like and how to set it up, but it was only after doing the checkmate manual that I learned many new patterns, like the Greco mate, which was on my radar in my last OTB game, that would've been really hard to think of and set up further away from the ideal position. I see immense value in drilling the courses you mentioned for drilling, and would be near the top of the list for me to drill for tactics.
@@chesscomdpruess Jesse's idea that repeating them makes them come more quickly during play seems intuitive though, because generally speaking speed comes with repetition.
Your point about about the tradeoff between repeating things and doing something new is also a good one. We need science to sort this kind of stuff out. Can Kabadayi (FM, cognitive scientist) said a while ago he's considering starting a project to answer questions about chess learning. Let's hope it happens.
22:48 that's only an issue when you have the course set to "key moves" instead of "all moves" - that's a setting you can set for yourself, so you always play from the very first move through the whole variation
Would love to see a video on how Chessable Mistakes and How to get the most out of Chessable!
I have so many chess books that I haven’t actually read. Spaced repetition with Chessable has not only helped me stick with material, but also drilled it. It’s really helped my digestion of the content
Same. I have a ton of books. Finally caving and getting Chessable has made me want to study and keep the streak alive!
@@oldman-badchess one thing I really like is the flexibility of Chessable courses. Maybe you just want to start with the “short and sweet” version for a taste. Then you buy the course and just start with quickstarter. Then go on to the “priority” lines, then finally you can choose to do the whole course. The level to which you dive is up to user and their time
Yep, same here. I tried several times to get through "100 Endgames You Must Know" but could never get very far before giving up. Chessable made it totally accessible to me. I completed the whole Chessable course in about a week and have been drilling it over a year now.
The default repetition schedule on Chessable is 4hrs, 10 hrs, 1 day, 2.5 days, 1 wk, 2.5 wks, 1.5 months, 4 months and then continues with 4-month spacing. The idea is that as long as you correctly repeat the moves in the given position, you advance to the next spaced repetition time, but if you fail, you fall back to an earlier spaced repetition time. You have the option of changing this schedule if you wish. For my opening and endgame drills, I changed the max repetition time to 2 weeks. For my tactical drills I left the default schedule in place.
Farming points is fun, to see your name at the top of the leaderboard for a given course. I just played against the French for the first time OTB yesterday, after spending a few days drilling lines and was able to produce the 12 moves in my course no problem. I set the interval to every hour, and just drilled it several times a day, and by the end of the week, I was able to play the 90 lines I drilled to 100% or near 100% accuracy consistently. Furthermore, I was also able to continue to play the best moves a bit further thanks to the themes in the position ingrained.
all respects but evey 60 mins i think its mentale health challenging. Also no duty in life, no physiological needs? I know we supposedly should replicate alphazero but....
@@tizianoricci7509 Just because it says you have stuff to review doesn't mean you have to. I said I reviewed it a few times a day, not every hour.
100% agree with Jesse being dropped into move 15 with out seeing the lead up moves is less helpful. Chessable needs an option or a button to see the lead up moves super fast.
well you can choose to learn and review "All moves" instead of "Key moves". The problem is that you have to manually play the same moves countless times that's very time consuming. Try doing that in an opening like the Marshall is super-boring
@@NgoTheVinh90 i agree that is a fail as an alternative
I think part of it _might_ depend on the "knack" or "talent" of the student. I remember one of my old classmates who simply attended lectures and did the assigned homework, and he scored at the top of the class. I shared that rank with him, but I spent countless hours in the library every week working additional problems from practically every other textbook I could find on the subject. I exhausted myself to get the results he got with minimal work because he was more "talented", or whatever, than I.
Famously, Magnus Carlsen has said he never did much in the way of tactical exercises, yet from what I've heard he's still a very strong chess player. Likely he has a "knack" for the game.
There isn't any scientific research on spaced repetition for chess tactics afaik, but some people who have had success with it recommend adjusting the spaced repetition schedule in Chessable to have MUCH longer intervals between reviews, starting with 7 or 30 days then scaling 2-2.5x after that.
The idea behind this schedule is to discourage rote memorization of the puzzle solutions like you would get with Chessable's default schedule, and encourage learning the general patterns instead. It ends up being more like The Woodpecker Method, but it avoids a problem in that method where you can spend too much time on stuff you already know.
Well Pruess just had a photographic memory when he was younger..
Spaced repetition/woodpecker style in general and Chessable in particular: I think it might not work/falls sort for some forms of study: openings and fastest mates (sometimes) if you have the text hidden (like i do), visualization exercises, etc. Also people probably should override book defaults to their preferences (i don't bother). It has flaws but the spacing algorithm is basically the same as language learning apps so it's pretty proven for any straight memorization task.
The visualization issue gets at the fundamental issue/con about spaced repetition is what you plan on getting out of it. With openings (I gather) and most endgames (say Silmans book at around 1700+ chapters) the usual goal is either memorization or pattern recognition: that's why spaced repetition seems to be less disputed.
In something like tactics: if it is pattern recognition, then it's probably more effective to space out a "single book." But, many do tactics to improve their visualization, calculation and/or evaluation (esp when you get to what Kanneman and Jesse like to call "system 2 thinking") where when "3 books one time" would be better since you are pushed out of being able to lean as heavily on memorization/leverage your lazy and efficient system 1 brain. "Tactics" is in that gray area where it's a bit of all of that, and people with different weaknesses and different goals. Depending on your specific context, you can make a strong argument againt spaced repetition, esp if someone gets caught up on the point system (in Chessable) and doesn't really figure out WHY that move is the right one (which probably happens a lot in opening courses there?).
Woodpecker spaced repetitions start with promlems that help your calculation. The repition though is desgned to help build pattern recognition. Pattern recognition is someting you get less of by doing 3 different books. Lower rated players often need a lot more help building an internal database of patterns.
I agree. Space repetition is probably best for endgames but also checkmating patterns. The only thing that will help improve calculation is more calculation so I think memorizing puzzles is not effective. As for openings, I'm surprised that databases weren't mentioned. I dreaded studying openings in the past. Since I learn to study openings using a database I'm hooked. My only rating jumped 100 points in a two-month period where I spend lots of time building an opening repertoire.
I used to think "oh wack, here comes david's opinon" now I love his way of thinking.
Good point by Kraii on being presented with move #18 and being drilled on how to proceed, without knowing how you got there... That's why I am now playing the french, the classical dutch and just playing 1. e3 for now and trying to have a clue what the hell I'm doing...
This is why I drill with pause at end of variation/line, so that I can go back to the start and see how I got to the line. Just because there is an option to automatically move to the next line doesn't mean you should.
Dang it, I remembered this too late... but I'll still post it because it's so pithy and apropos. Bruce Lee famously said "I do not fear the man who has practiced 10,000 kicks once. but I fear the man who has practiced one kick 10,000 times."
Thank you
Seems to me that rote learning lines is close to pointless, because in 99.9% of games your opponents will deviate at some stage and even if they don’t, after you’ve sacked 2 pieces, a rook and half a dozen pawns, “with compensation for the sacrificed material” is hardly sufficient information on how to proceed.
This was part 2 of the series. But I dont see part 1 ?
ruclips.net/video/vTlgfV48LbE/видео.html will add to the description!
I like the repetition …it works in education
One thought I have on woodpeckering calculation sets is you're basically converting a calculation exercise to a memory recall exercise, which might not be ideal. If you're drilling tactical motifs, it seems like it's a bit better (but it doesn't seem to be any better than more unique examples of a given concept).
Yes, the Woodpecker method/spaced repetition method works well to get tactical patterns/motifs ingrained into your long-term memory so that you're more likely to recognize them when playing a game, but it doesn't help with improving your calculation. Tactical pattern recognition and calculation go hand-in-hand but are different skills that need to be trained in different ways.
ould you define: system two thinking?
23:24 you can change that setting. I found I didn’t like that either. Much better to play from move one.
How do you change it?
@@davidfranklin5426go to your course, open it and where the repetition options are, Learn: Everything, Study: All Moves, Review: Whole Variation, Depth: Full Depth. So start your course with those options. If it’s a course you’ve already through. You will have to make the changes and go back and “learn” every line over again and you will “learn” all the beginning moves again.
Yes, I agree. For opening drills I much prefer to start from move 1. You can change this setting by going to the "Chapters" page for the course and in the "Course Settings" along the right-hand side, change the "Review" selection to "Whole Variation".
I agree with Kostya in that I'd prefer if I could drill my opening lines using a physical board and pieces. I've submitted to Chessable a suggestion to implement an interface for e-boards so I could use a physical board when drilling opening lines but who knows if they'll ever do that.
You can do that with at least some of the Chessnut Eboards (Chessnut Air is one of them). Not sure but there should be some other eboards with the same functionality. . Still not the great stuff because the board recognizes "the original flow" but at the moment you start reviewing your mistakes it does not recognize the position because it goes from lets say move 12 to move 5 again and all the pieces are elsewhere, so you would need to arrange the pieces manually to that position for every reviewed mistake
I have a chessable account and every now and then I try to use it. But sorry… I just can’t stand it. The repetition method is so dull and infantilizing. I know it works like a charm for others but it’s just not for me.
I highlight, underline, use marginal notes in my books. The reread is much faster.
Can you guys please review the book logical chess move by move by Irving chernev
Chessable is only half the battle. You used the spaced repetition to get to a position then THAT position is your Tabiya to spar from and fight from. Memorizing and learning their little blurbs isn't enough. You need to train in those final positions for it to be useful, otherwise, the opening doesn't matter bc you don't even have experience in what you're learning.
Let's gooo
Listening to David argue against proven science, yeah, I'm out.
Play chess960 and stop this insanity ty
Spaced repetition isn't an opening specific technique, it's a learning tool, even outside of chess. Ignoring openings: tactics, strategy and endgames are still very significant areas of study that benefit from spaced repetition and were talked about during the video you've just commented on...
Maybe rewatching it (or watching it the first time?) may benefit your understanding of spaced repetition in an ironic way :)