Three years, thousands of games, no improvement?

I've been on this site for 7 years and my blitz rating and rapid has not improved. I did improve in bullet and puzzles though  

One problem I see with just analyzing your games with an engine is that the engine gives (what it thinks to be) the best move, but doesn't explain why it's the best move and, more importantly, how should I have spotted it.

Sometimes it's very obvious why the move (or moves, if you turn on multi-pv) it suggests is the best, but it's still not always obvious how I should have seen or noticed it. Sometimes the strength of the move only becomes apparent after the game progresses a few moves ahead, so it's not necessarily so obvious to see at that point why it's the best move.

Othertimes it's not clear at all why a move is the "best" move, especially if the evaluation barely changes. In these cases engines just don't explain why. Other than, perhaps, with an implicit "all the other possible moves are worse".

Conversely, an engine may show a move of mine to have been a blunder (if the evaluation drops significantly) but, once again, it doesn't necessarily explain why it is, and it might not be immediately apparent why. It might not even become apparent even after a half dozen moves. The engine just calculated with its sheer brute-force power that 20 moves ahead it leads to an inferior position, but that's hardly useful for me.

More useful would be to know general principles and tactics that improve the position, ie. a higher level explanation of what should be done in a particular position and why (ie. not a "20 moves from now your position becomes better by a complicated evaluation function", but a "it strengthens your position by defending these pieces" or "it weakens your opponent's positions by putting pressure on this piece" or similar.)

1. Puzzles are the best way to train with engine moves as they ask you to replay exactly " the same moves as engines".

2. About 90% of the times, the reason the engines choose those moves are within 5 moves. If the solution is beyond those moves, forget about them. Alternatively,  if you cant visualize the suggestion,  pause the move, and see the end of principal variations.

3. Actually reviewing your game doesn't mean to follow engine moves, you can review with your own and find your own mistakes . And ask engine for assistance only.

Yet another victim of all the hype about how to improve.

Puzzles are exactly the wrong way to improve.

For beginning players ... all this hype about solving puzzles simply destroys learning methodology.

You will not encounter puzzle positions in your games.  Even when they present themselves... thinking a best move solution offers itself for every move - is a sad mistake. “Tactics” present themselves at every turn = hogwash 

I will challenge drmrboss on his claim . Any day ... any hour. On his claim that solving puzzles is a primary learning tool for beginning players.

In fact - an emphasis on solving puzzles (unless highly rated) is detrimental for improvement.

You dont need to challenge me, you challenge your opponent " within your rating" and " with your prefer slower time control". Higher rated people prefer faster time control ( >90% of 2000+ play 1-3 min time control) and you wont be able to play at our usual pace.

I got from 1250 to 1770 USCF rating in slow tournament play primarily by doing lots of tactics puzzles, playing a lot, and reviewing some of my games. The key is not to do random puzzles on a web site with no explanation. As I mentioned upthread, choose books with puzzles specifically chosen for their instructional value. Do them over and over until you can spot the solutions instantly. Then move on to a harder book.

You're correct that tactics don't present themselves at every turn. But if you can't spot them when they do show up, you'll lose a lot of games. Studying tactics isn't about finding moves to win. It's about spotting the moves that would let your opponent win, so you can defend against them. Eventually, you'll reach a level of needing strategy, also, but that's around 1800 OTB. You can get that far almost entirely on tactical study.

But don't take my word for it. Ask any professional chess coach with a good reputation and lots of students who have reached high level. They'll all tell you that tactics puzzles are a key part of any study program. 

So I could progress beyond "grasshopper" with this?

https://www.old-games.com/download/1878/budokan

Lazy coaches ... blah blah blah 

Do this... do that.

I’m telling it like it is ...

from many a year of experience-

For the new players - the chess experience is one to be taught from a different view if it is to last. 

drmrboss can not step up. Thinks his rating tells all. This is exactly the issue regarding instruction for the new players. They become deceived- about the nature of our hobby.

The OP asks ... 

why no improvement after 3years?

I know why. Most here are clueless.

drmrboss ? Explain your knowledge  of hands on practical experiences? 

Besides this repetitive cliches of solving puzzles and your high rating = knowledge?

Tactic puzzles are excellent for improving one's ability to find combinations. But what do you play if you're in a position where you can't find a winning combination?

In those cases, it helps to know the thematic ideas intrinsic to certain kinds of positions.

If you're playing the Black side of a Sicilian Scheveningen, for example, where White is castled long, and has a king-side pawn storm underway, do you know a specific square, on the queenside, that you should be aiming to get a knight to? And do you know what the appropriate time to develop your king bishop is? Should your h pawn be on h6, with the intent of slowing down White's pawns? Or should you avoid playing ... h6, as it gives White's pawns a way to pry open Black's kingside?

If you aren't studying this kind of stuff, you'll be forced to figure it out over the board. Probably, you won't make all the right choices. But if you are studying this kind of stuff, then you'll have already figured out some of these answers beforehand. And you'll know immediately when your opponent has gone wrong.

And this can extend to every facet of the game. Openings. Middles. Endings.

Grandmasters play excellent chess not because they can see all the best moves at a glance, but because they've studied so many key positions, and have learned important thematic ideas.

This kind of knowledge is out there, in various books and other places, but the majority of players don't (for whatever reason) choose to study it ...

Yes, that type of thing is important, TestPatzer. Once you stop losing tactics every game. But all the strategic knowledge in the world won't help if you overlook a knight fork. Or at slightly higher levels, a combination to set up a knight fork. That's why tactics must be studied before strategy. 

There are some people who advocate nothing but tactics puzzles until you hit 1800 in slow OTB tournaments. That's probably a bit extreme, but tactics should be at least half of your study (non-playing) time. 

I think it is usually best to defer to strength. It is a little weird for someone who is 600pts lower rated in blitz to be telling the higher rated player that their approach is wrong. Why is the lower-rated player's rating so low if they have the right approach? And to appeal to the masters, many say that at the beginner and intermediate level, tactics are 90% of the game. I tend to agree since many of my blitz games against intermediate players are determined based on tactical errors

Ha ha, I'm sure everyone has noticed that so far BBJ3 hasn't been all that helpful to the OP. All he's done is give his opinion on what doesn't work - He hasn't yet given his opinion on what *does* work. Maybe he's waiting to get paid for his advice, or maybe he's pausing for dramatic effect. Maybe he wants us to beg him for his advice, idk. Regardless, it should be entertaining, if he ever gets around to it. Maybe it'll be something we've all heard before, like the Soviet training method... Whatever...I think he just likes having an audience and tearing things down...

+1

Maybe go back to Djing.

About using an engine to analyze one's game, consider this position from my most recent game:

Stockfish suggests Qa4 as the best move. It's not at all obvious to me why, not even after seeing the continuation.

In my mind it looks to me like putting the queen on a relatively precarious position, at the edge of the board where it doesn't really do much. It attacks the a6 bishop but so what? It just moves to b7 and that's it. Ok, perhaps it's a tad bit less active there, but I don't see much benefit. (The next move Stockfish suggests after this is Nb3, kicking the c5 bishop to b6, but once again I don't really see the point. Almost looks like a better position for that bishop than undefended on c5.)

This just feels like one of those engine moves that are strong only if you are really strong and know how to follow it up. It's not a move that would even cross my mind, nor do I see its reasoning even in retrospect, and even after seeing the engine's suggested followup.

The second (or third, depending on how long I allow the engine to calculate) strongest move, albeit quite a bit behind Qa4, that it suggests is Rb1. This actually makes a whole lot more sense to me. It puts the rook on a semi-open file, and it supports a future pawn to b4 advance, which looks really strong. This is one that in retrospect even I can see is a strong prophylactic move. Yet, it only considers this the second/third strongest move in this position.

(Btw, the other move that switches position back and forth with this one for the second-best is Nf3. I see benefits and drawbacks from this. On a plus side it puts the knight on a more active position and opens up the way for the c1 bishop to move. On the minus side it weakens both the c4 and e4 pawns. If I were to be asked if this is a good or a bad move, I wouldn't be sure.)

There will always be such occurrences that you just don't understand some engine move. If you made a huge mistake, and it is not easy to figure out why, go through line, as you did. If the first line is unclear as you need to find 5 good moves afterwards which are not really forced on intermediate level (it will happen even for bigger mistakes) go to the next line. Perhaps you will understand that line as it might be more logical to a human, So if you don't understand the line then just forget about it (if every move better than yours is unclear, forget about them altogether).

In most cases it will not happen for blunders or mistakes, but in this case, white has +0.4 to +0.8 advantage anyway, so it is normal that in many instances it is unclear to us why +0.8 move is better than +0.5 or +0.6 move. If you don't understand it, just forget about it. It is a marginal difference anyway.

As you study , a little by little, more things will become clear. 

As for Qa4, after his Bb7, his bishop is a bit worse there, you made that move with tempo, so after he moves the bishop to a worse square, you have the move again. So probably the fact that Qa4 is practically a free move that makes his bishop less active and you get to move again, makes that move pretty good.

As for the precarious position of the queen, how can he attack it? But of course, just in case don't play b6 afterwards and close down the escape path. 

All of that being said, I am not completely sure if I would play that move, for the same reason you've mentioned, but after the fact I kind of understand why the move is good.