Jesse - Lichess studies are exactly what you are looking for in terms of a file where a student can input their games and analysis. I've been doing this since I started playing OTB a few months ago and I've found it really valuable, I do it with my online rapid games now too.
Yep, I do this with all of my classical Lichess games. Has been very helpful! I don't use the engine anymore, thanks to Dojo's input. It really has been worthwhile in my development as a player.
Suppose you study math at school. Suppose you have a test. You do the test, you bring your work to the teacher, and teacher tells you that he will not score you work or give you any feedback. He tells you that you should do you own analysis and learn that way. Does it make any sense to you? It does not to me, because I would go back, use the same knowledge, same skills and come up with the same answers. Nothing new. If you actually see some GM-level streamers doing post-game analysis - they spell the same thoughts, same conclusions, same decisions as during game. It's just repeat, nothing new. What's new is when they turn on computer and start getting surprised. Oh, why this move is considered bad? Oh, what was another good move here?
I totally agree with u and wrote something similar in the comments above...At least one person who understands my problem. Nice example with the test btw. Now I feel a lot better :)
You're totally missing the point. During analysis, you can move the pieces, you have more time, and you have insights about how the game ended. Hence you have very different tools
@@felbas4224 You make a good point: you can still find things in your post-game analysis that you missed during the game for the reasons you cogently expressed. What Roman is saying is that there are going to be many mistakes in your reasoning and ideas you've missed that you would not be able to realistically find on your own. That's where feedback from a coach, be it computer or human, would come in.
In my view, the saddest part of improvers is they are not analyzing their games to any significant level of depth and are just spamming the new game button.
When I started playing chess last year, I could not analyze my own games. I didn't know how. The computer honestly helped me a lot with knowing how to analyze my games. Obviously it's "better" to analyze on your own, but if I can analyze with the engine literally 100 times faster, it's still doing me a lot of good to do it, even if I only get one tidbit out of every few games
@aname4141 Well-put! Me too. The engine's strength and precision is alluring, but I always feel a bit "lost" on why it thinks some moves are good or bad.
Exactly! You grow so much more analysing without an engine. I use lichess study and take 30-90 minutes analysing without it on, depending on the complexity of the game. Then after taking notes I’ll try it with engine on but lines and best move arrow turned off, finally I’ll check with the engine in full. So I go through the game in 3 stages. Wish I implemented this method years ago!
Have you noticed an improvement over another method (like using the engine immediately)? I've reached 1600 rapid using immediate engine analysis, perhaps not knowing better.
@@marcofrey2903 theres nothing wrong with that, but you should at least have a look without it first you might surprise yourself. Fair enough if time is a factor though
@@Chessdrummer83 I play on mobile is one problem. I just like it better. More tactile and I can sit cross legged on a nice chair. But I plan on importing at least one long game per week into a Lichess study and analyzing on my own. Does the “Check with Engine” work on all the sidelines you put into a study?
Another reason to analyze just using a real chess board: it helps you practice visualizing the position. For example, you play out a variation you were curious about during the game, so you start moving pieces. Do you move 25 moves like clicking thru an engine, then instantly set them back? No, you move forward a few moves until it resolves the position somewhat, maybe 3-5 moves (6-10 ply), and then you have to move the pieces back to their original positions using your memory. I've done this lots of time with human opponents OTB during a postmortem. It's very natural and it forces you to remember the "character" of a particular position, when you have to set it back up again. As for writing down everything--I'm not sure that's a great idea, because that's very time-consuming, and unless you plan on reviewing your annotated games afterward, it seems like a waste of time. More important is to just go over the game and talk to yourself, articulate in chess terms what is going on (strong pawns, weakened king position, pins, offsides knight, etc.), and try to remember what you were worried about during the game. This is why it's best to analyze your games as soon as you can afterwards, so your hopes and fears are fresh in your mind. That should help you learn what your mindset was.
Engines cannot tell you what plan you should pursue, nor really address your most important weaknesses at your level. All they can do is suggest the strongest possible move in any given position. They're not really "coaches". I totally agree with David, that "not knowing" is just part of chess in general, esp. because it's such a complex game, and so when did we all of a sudden become precisionists? Just because computers got so strong? Before computers, masters had to suss out for themselves, over an actual board, what was strongest. If no one in the Vienna cafes could prove them wrong, their variation was, de facto, the "best". They would eagerly contribute their analysis to other master games, organically thinking of plans, connecting and coordinating their pieces together in an orchestrated way, towards a goal. I used to use an engine a lot, but now I see the value of just shutting it all off and going through the game, thinking about what I was worried about, what I felt was strong or weak, and why, and coming up with stronger alternative lines. And that's enough, because I went over my loss to begin with, and I not only found stronger moves, I understood structurally WHY all those stronger moves were better, because I thought of them and tested them out, and played against myself. Putting that analysis into an engine will just be demoralizing, though, because the computer will just shred it and make you feel stupid. And if you care more about precision and accuracy in analysis to the point of downplaying or downgrading your own analysis, then you will eventually succumb to the "plug and chug" engine crowd, who will come out with a list of isolated mistakes and the "strongest move", and think if they just replace this move with that, that's learning. I used to do that, but I still made the same mistakes...UNTIL I started doing the hard work of looking at my games myself, and asking my own questions, and getting a feeling for what was playable or possible for me, at my level (~1850 USCF). An engine cannot tell you that, for example, in any given position, giving up the bishop pair is a good idea because you create pawn weaknesses that can easily be attacked in a few moves.
I'm very guilty of immediately analyzing my games with an engine. But that's only true of my blitz games! I'll quickly check the opening book, do a quick blunder review, and then confirm my feeling of the evaluation was accurate throughout. In classical games, I'm following David's advice though all the time. The more interesting the game, the more time and effort I'm willing to give my personal analysis. Only after an hour or so will I then check my in-game thoughts and my analysis thoughts with the engine, look for further improvements, etc.
It's time-efficient to see any major turning points rather than spendind centuries trying to find those spots on your own. It helps you refine your move order.
After all the work I do with my own analysis, plugging it into an engine is completely demoralizing, shredding many of my ideas. This has led me to rely more on the engine. But I just don't learn that much by looking at isolated move evaluation scores. I need to connect the moves in a coherent plan or structure, and engines just can't explain any of that.
I (1600) use the engine to see the blunders and mistakes only, drastic changes in evaluation. But before computer analysis I input my moves which I think was better than the one played.
Agreed, I don't bother trying to understand every inaccuracy. But the missed tactics can be really valuable. Some of the missed tactics in the openings I have added to a private chessable course and drilled them and subsequently won many games (at low rating levels - damiano defense, I'm looking at you).
Ok. You are right to say the engine is thrust upon you immediately on lichess, for example, and you will see the big mistakes and turning points straightaway. The point is though, even having that high level information which you ought to have sensed during the game anyway, you should then analyse the game on top of that.
what do you guys think about this usage of the engine -- I'm playing rapid games focusing on changing my thought process and doing blunder-checking and CCT on every move. a quick check with the engine immediately after the game shows me how accurate that thought process was during the game, and then I try to understand how to improve it. do you think this way of using the engine is harmful?
At some point I got used to not use the engine. It’s just not what I do for long time control games. I do my lichess study and contemplate about the games. If I fell, I don’t get it, I ask my coach or submit to the dojo. Half of the games, I never counter check them with the engine anymore.
For someone who doesn't really know the best way to objectively analyze my own games (only 1400-1500 rapid), the engine has undoubtedly improved my gameplay.
The engine does not help me when I am just outplayed. Many of the suggestions are so subtle and the suggested lines so long, it doesn't help me play better.
Im still halfway so maybe I shouldnt write this but I feel this could be simplified a lot. "Computer" is just a tool which spits out good moves. This in and of itself can never hurt you. It's just knowledge. Only you can hurt yourself. And this generally comes from not understanding nature of good moves. If you think about it this is a laziness problem. Blame the computer all you want for making people lazy in reality overcoming laziness is about willpower. And many other things like not playing long time controls, cheating is about lacking willpower.
I agree. See my comment above. They don't even mention how you can work to understand an engine move if you don't at first. You can ask it questions: What if this? What if that? Go down the lines, etc. Perhaps there's a bit of cognitive overload but if I can rest assured that I'll remember even 5% of the tactics or positional ideas, then I know I've improved.
Kostya's opinion is that it's more important to understand your openings than to know what the engine thinks. The engine does become more useful for higher rated players who already have a fundamental understanding
I want to use it less. It took me a long time to even start looking at my games at all even with a computer. I think looking with a computer is better than not looking at all so someone who has this vice is still doing better than 90% of players from an improvement perspective compared to players who just play and do no review at all. I use it after blitz games to find tactics I missed which I then try to commit to memory. I also use it to point out my own blunders that went unpunished. I agree with you all and plan to work on this vice.
The reason kids improve faster now is because we have the chess engines... Yes, it's important to think on your own but it also saves infinite hours to have a computer highlight the most critical moves that you can then look into further...
Is class E player really supposed to analyse his own game by himself? For what I see from this analysis your supposed to find opponents strongest moves (or answers) more so than your own.
Here's a non-authoritative opinion coming from someone who's reached 1600 in 2 years or so, without a coach. There has to be a middle ground. Pruess seems to think using the engine actually harms progress. If that were true, I wouldn't be at the level I'm at now. Of course, I've also watched lots of TH-cam videos, done some chessable courses, etc, but I have to believe the engine has sped up my improvement. There's a way to use the engine in a human way, by asking it questions, ignoring its suggestions sometimes, and giving yourself a pass for finding a brilliant move that can only be refuted by an obscure counter. Several times y'all say that the engine suggests moves only a GM could understand. Perhaps, but then I'll check out the third or fourth suggestion until I find one that makes more sense to me intuitively. What's more--just ask the engine questions. What if this? How does this change this tactic I was going to try? Follow it down the lines. I think this is totally valid. I've even used the engine to find novelties in sidelines only viable at my lower level, something I cannot enjoy playing once I've leveled up (chess is for enjoyment after all). What I did glean from this discussion is to take time to analyze with the quiet and curiosity of your own mind, something I should definitely start doing. This has to be key to taking things to a higher level and being more surefooted in ones own style and choice of moves, moves you can live with even if they're not the engine's best. But I can still see myself checking with the engine afterwards: It's also great feedback to check if I calculated a line accurately and found my opponents resources or not. While I truly admire the era before computers, and sometimes wish we could go back, it's just not possible when everyone else has access. It truly was remarkable that games were simply played and people went home with nothing but a hunch and further flawed analysis on their own. But also, this meant only the privileged could ever hope to reach levels anyone can reach today. Think: If you didn't have access to higher minds, you were only left to your own devices. In some ways it's a sad thought. Nevertheless, sitting and tinkering with pieces after a game like a child might, like a young Magnus did, trying things out on our own--there has to be gold in that.
Personally I'm also using the engine a lot...*For me It's very hard to see things during analysis, I already didn't see during the game.* I mean there is a reason for decision making during the game, why should u change your mind afterwards? How is one supposed to get better when he is just analyzing with his own knowledge. It's like trying to get better at some Science like Math for instance by just calculating or sticking to what you know when the real thing needed is for instance a knew concept or technique? Like u are trying to solve an equation without knowing how to transform it. If you don't have a better player nor a coach to show you new concepts and analyze with u why should u get better by thinking only on your own of your game afterwards? Maybe it's just me but I really struggle with it. I mean I really like to exchange some thoughts after an otb-game with my opponents. They tell me what they were thinking and I tell them what I was thinking. And I usually stick with the "was" I just don't get knew ideas during analysis...so why shouldn't I immediately check my thoughts with an engine after the game? :c
So, what's the harm in using computer analysis? Is it fear that if student uses computer analysis he will not use his own brain? Should we have the same fear when student use a human coach? Or is it merely that a student will not understand what's computer saying? What's the harm in not understanding what's computer saying??? Really, guys, you are so against it - it must be something really bad with computer analysis.... and I don't understand what. There is old saying that in order to understand the system - you have to get out of the system. When a student plays with his own brain and uses his own brain for analys - he stays within the same system of his brain. He will not learn much, he will just re-live the experience. Computer analysis may be pointless and hard to understand, but at least it's another point of view - a view outside the system. So I disagree with you guys.
10/10 thumbnail
Jesse - Lichess studies are exactly what you are looking for in terms of a file where a student can input their games and analysis. I've been doing this since I started playing OTB a few months ago and I've found it really valuable, I do it with my online rapid games now too.
Yep, I do this with all of my classical Lichess games. Has been very helpful! I don't use the engine anymore, thanks to Dojo's input. It really has been worthwhile in my development as a player.
Suppose you study math at school. Suppose you have a test. You do the test, you bring your work to the teacher, and teacher tells you that he will not score you work or give you any feedback. He tells you that you should do you own analysis and learn that way. Does it make any sense to you? It does not to me, because I would go back, use the same knowledge, same skills and come up with the same answers. Nothing new. If you actually see some GM-level streamers doing post-game analysis - they spell the same thoughts, same conclusions, same decisions as during game. It's just repeat, nothing new. What's new is when they turn on computer and start getting surprised. Oh, why this move is considered bad? Oh, what was another good move here?
I totally agree with u and wrote something similar in the comments above...At least one person who understands my problem. Nice example with the test btw. Now I feel a lot better :)
You're totally missing the point. During analysis, you can move the pieces, you have more time, and you have insights about how the game ended. Hence you have very different tools
@@felbas4224 You make a good point: you can still find things in your post-game analysis that you missed during the game for the reasons you cogently expressed. What Roman is saying is that there are going to be many mistakes in your reasoning and ideas you've missed that you would not be able to realistically find on your own. That's where feedback from a coach, be it computer or human, would come in.
In my view, the saddest part of improvers is they are not analyzing their games to any significant level of depth and are just spamming the new game button.
When I started playing chess last year, I could not analyze my own games. I didn't know how. The computer honestly helped me a lot with knowing how to analyze my games. Obviously it's "better" to analyze on your own, but if I can analyze with the engine literally 100 times faster, it's still doing me a lot of good to do it, even if I only get one tidbit out of every few games
@aname4141 Well-put! Me too. The engine's strength and precision is alluring, but I always feel a bit "lost" on why it thinks some moves are good or bad.
Exactly! You grow so much more analysing without an engine. I use lichess study and take 30-90 minutes analysing without it on, depending on the complexity of the game. Then after taking notes I’ll try it with engine on but lines and best move arrow turned off, finally I’ll check with the engine in full. So I go through the game in 3 stages. Wish I implemented this method years ago!
Interesting approach, thanks for sharing.
@@tobelli Yep I still do this, I think its quite thorough
Have you noticed an improvement over another method (like using the engine immediately)? I've reached 1600 rapid using immediate engine analysis, perhaps not knowing better.
@@marcofrey2903 theres nothing wrong with that, but you should at least have a look without it first you might surprise yourself. Fair enough if time is a factor though
@@Chessdrummer83 I play on mobile is one problem. I just like it better. More tactile and I can sit cross legged on a nice chair. But I plan on importing at least one long game per week into a Lichess study and analyzing on my own. Does the “Check with Engine” work on all the sidelines you put into a study?
Top notch thumbnail
The proud look on Jesse when David disses the computer.
Another reason to analyze just using a real chess board: it helps you practice visualizing the position. For example, you play out a variation you were curious about during the game, so you start moving pieces. Do you move 25 moves like clicking thru an engine, then instantly set them back? No, you move forward a few moves until it resolves the position somewhat, maybe 3-5 moves (6-10 ply), and then you have to move the pieces back to their original positions using your memory. I've done this lots of time with human opponents OTB during a postmortem. It's very natural and it forces you to remember the "character" of a particular position, when you have to set it back up again. As for writing down everything--I'm not sure that's a great idea, because that's very time-consuming, and unless you plan on reviewing your annotated games afterward, it seems like a waste of time. More important is to just go over the game and talk to yourself, articulate in chess terms what is going on (strong pawns, weakened king position, pins, offsides knight, etc.), and try to remember what you were worried about during the game. This is why it's best to analyze your games as soon as you can afterwards, so your hopes and fears are fresh in your mind. That should help you learn what your mindset was.
Engines cannot tell you what plan you should pursue, nor really address your most important weaknesses at your level. All they can do is suggest the strongest possible move in any given position. They're not really "coaches". I totally agree with David, that "not knowing" is just part of chess in general, esp. because it's such a complex game, and so when did we all of a sudden become precisionists? Just because computers got so strong? Before computers, masters had to suss out for themselves, over an actual board, what was strongest. If no one in the Vienna cafes could prove them wrong, their variation was, de facto, the "best". They would eagerly contribute their analysis to other master games, organically thinking of plans, connecting and coordinating their pieces together in an orchestrated way, towards a goal. I used to use an engine a lot, but now I see the value of just shutting it all off and going through the game, thinking about what I was worried about, what I felt was strong or weak, and why, and coming up with stronger alternative lines. And that's enough, because I went over my loss to begin with, and I not only found stronger moves, I understood structurally WHY all those stronger moves were better, because I thought of them and tested them out, and played against myself. Putting that analysis into an engine will just be demoralizing, though, because the computer will just shred it and make you feel stupid. And if you care more about precision and accuracy in analysis to the point of downplaying or downgrading your own analysis, then you will eventually succumb to the "plug and chug" engine crowd, who will come out with a list of isolated mistakes and the "strongest move", and think if they just replace this move with that, that's learning. I used to do that, but I still made the same mistakes...UNTIL I started doing the hard work of looking at my games myself, and asking my own questions, and getting a feeling for what was playable or possible for me, at my level (~1850 USCF). An engine cannot tell you that, for example, in any given position, giving up the bishop pair is a good idea because you create pawn weaknesses that can easily be attacked in a few moves.
I'm very guilty of immediately analyzing my games with an engine. But that's only true of my blitz games! I'll quickly check the opening book, do a quick blunder review, and then confirm my feeling of the evaluation was accurate throughout.
In classical games, I'm following David's advice though all the time. The more interesting the game, the more time and effort I'm willing to give my personal analysis. Only after an hour or so will I then check my in-game thoughts and my analysis thoughts with the engine, look for further improvements, etc.
It's time-efficient to see any major turning points rather than spendind centuries trying to find those spots on your own. It helps you refine your move order.
After all the work I do with my own analysis, plugging it into an engine is completely demoralizing, shredding many of my ideas. This has led me to rely more on the engine. But I just don't learn that much by looking at isolated move evaluation scores. I need to connect the moves in a coherent plan or structure, and engines just can't explain any of that.
I (1600) use the engine to see the blunders and mistakes only, drastic changes in evaluation. But before computer analysis I input my moves which I think was better than the one played.
Agreed, I don't bother trying to understand every inaccuracy. But the missed tactics can be really valuable. Some of the missed tactics in the openings I have added to a private chessable course and drilled them and subsequently won many games (at low rating levels - damiano defense, I'm looking at you).
I am around 1800. Dont have a coach. How to analyse if not with a computer??
Ok. You are right to say the engine is thrust upon you immediately on lichess, for example, and you will see the big mistakes and turning points straightaway. The point is though, even having that high level information which you ought to have sensed during the game anyway, you should then analyse the game on top of that.
what do you guys think about this usage of the engine -- I'm playing rapid games focusing on changing my thought process and doing blunder-checking and CCT on every move. a quick check with the engine immediately after the game shows me how accurate that thought process was during the game, and then I try to understand how to improve it. do you think this way of using the engine is harmful?
Kostya thinks that makes sense but this was part of the discussion we had in the video. David would probably be opposed!
At some point I got used to not use the engine. It’s just not what I do for long time control games. I do my lichess study and contemplate about the games. If I fell, I don’t get it, I ask my coach or submit to the dojo. Half of the games, I never counter check them with the engine anymore.
Funny thumbnail
For someone who doesn't really know the best way to objectively analyze my own games (only 1400-1500 rapid), the engine has undoubtedly improved my gameplay.
The engine does not help me when I am just outplayed. Many of the suggestions are so subtle and the suggested lines so long, it doesn't help me play better.
Im still halfway so maybe I shouldnt write this but I feel this could be simplified a lot. "Computer" is just a tool which spits out good moves. This in and of itself can never hurt you. It's just knowledge. Only you can hurt yourself. And this generally comes from not understanding nature of good moves. If you think about it this is a laziness problem. Blame the computer all you want for making people lazy in reality overcoming laziness is about willpower. And many other things like not playing long time controls, cheating is about lacking willpower.
I agree. See my comment above. They don't even mention how you can work to understand an engine move if you don't at first. You can ask it questions: What if this? What if that? Go down the lines, etc. Perhaps there's a bit of cognitive overload but if I can rest assured that I'll remember even 5% of the tactics or positional ideas, then I know I've improved.
how about the usage of engines for openings?
Kostya's opinion is that it's more important to understand your openings than to know what the engine thinks. The engine does become more useful for higher rated players who already have a fundamental understanding
@@ChessDojo thank you for the reply!
I want to use it less. It took me a long time to even start looking at my games at all even with a computer. I think looking with a computer is better than not looking at all so someone who has this vice is still doing better than 90% of players from an improvement perspective compared to players who just play and do no review at all.
I use it after blitz games to find tactics I missed which I then try to commit to memory. I also use it to point out my own blunders that went unpunished.
I agree with you all and plan to work on this vice.
The reason kids improve faster now is because we have the chess engines...
Yes, it's important to think on your own but it also saves infinite hours to have a computer highlight the most critical moves that you can then look into further...
Is class E player really supposed to analyse his own game by himself? For what I see from this analysis your supposed to find opponents strongest moves (or answers) more so than your own.
I agree with Jesse and David more so than Kostya here
Here's a non-authoritative opinion coming from someone who's reached 1600 in 2 years or so, without a coach. There has to be a middle ground. Pruess seems to think using the engine actually harms progress. If that were true, I wouldn't be at the level I'm at now. Of course, I've also watched lots of TH-cam videos, done some chessable courses, etc, but I have to believe the engine has sped up my improvement.
There's a way to use the engine in a human way, by asking it questions, ignoring its suggestions sometimes, and giving yourself a pass for finding a brilliant move that can only be refuted by an obscure counter. Several times y'all say that the engine suggests moves only a GM could understand. Perhaps, but then I'll check out the third or fourth suggestion until I find one that makes more sense to me intuitively. What's more--just ask the engine questions. What if this? How does this change this tactic I was going to try? Follow it down the lines. I think this is totally valid. I've even used the engine to find novelties in sidelines only viable at my lower level, something I cannot enjoy playing once I've leveled up (chess is for enjoyment after all).
What I did glean from this discussion is to take time to analyze with the quiet and curiosity of your own mind, something I should definitely start doing. This has to be key to taking things to a higher level and being more surefooted in ones own style and choice of moves, moves you can live with even if they're not the engine's best. But I can still see myself checking with the engine afterwards: It's also great feedback to check if I calculated a line accurately and found my opponents resources or not.
While I truly admire the era before computers, and sometimes wish we could go back, it's just not possible when everyone else has access. It truly was remarkable that games were simply played and people went home with nothing but a hunch and further flawed analysis on their own. But also, this meant only the privileged could ever hope to reach levels anyone can reach today. Think: If you didn't have access to higher minds, you were only left to your own devices. In some ways it's a sad thought.
Nevertheless, sitting and tinkering with pieces after a game like a child might, like a young Magnus did, trying things out on our own--there has to be gold in that.
Personally I'm also using the engine a lot...*For me It's very hard to see things during analysis, I already didn't see during the game.* I mean there is a reason for decision making during the game, why should u change your mind afterwards?
How is one supposed to get better when he is just analyzing with his own knowledge. It's like trying to get better at some Science like Math for instance by just calculating or sticking to what you know when the real thing needed is for instance a knew concept or technique? Like u are trying to solve an equation without knowing how to transform it.
If you don't have a better player nor a coach to show you new concepts and analyze with u why should u get better by thinking only on your own of your game afterwards? Maybe it's just me but I really struggle with it.
I mean I really like to exchange some thoughts after an otb-game with my opponents. They tell me what they were thinking and I tell them what I was thinking. And I usually stick with the "was" I just don't get knew ideas during analysis...so why shouldn't I immediately check my thoughts with an engine after the game? :c
Yes it's hard to see different ideas than what you played, but by do finding them (even if it's hard) you improve.
So, what's the harm in using computer analysis? Is it fear that if student uses computer analysis he will not use his own brain? Should we have the same fear when student use a human coach? Or is it merely that a student will not understand what's computer saying? What's the harm in not understanding what's computer saying??? Really, guys, you are so against it - it must be something really bad with computer analysis.... and I don't understand what.
There is old saying that in order to understand the system - you have to get out of the system. When a student plays with his own brain and uses his own brain for analys - he stays within the same system of his brain. He will not learn much, he will just re-live the experience. Computer analysis may be pointless and hard to understand, but at least it's another point of view - a view outside the system.
So I disagree with you guys.
I see. Best to turn that thing off
You first!
Takbiir
Deep