GM Grigoryan on the "Myth" of Solving Puzzles
Sort:
Feb 1, 2022
0
#222
Exactly as Jetlag says. I believe it was Fischer who said tactics flow from superior positions. If you get a good position out of the opening and your pieces are on good squares, tactics become more available. You can say that x% of games are won on tactics, but are they really? That’s like saying that the assassination of the archduke started WWI. Lots of things had to happen in order to allow that event to set off the powder keg.
NikkiLikeChikki wrote:
Kartikeya - Let me show you a game that demonstrates just how useless theory is. The opening, in case you didn't know, is called the Double Muzio Gambit. It's something no sane person would play without knowing the theory because it looks absolutely idiotic. The thing is, there's a method to the madness, and in the Lichess database, white wins a huge percentage of games.
Now I castle on move 5 and my oppenent thinks I'm an idiot. If my opponent doesn't know the theory, they will think I am either a moron or accidentally premoved castling. I mean just look! I'm giving up a free piece! Idiocy. And yet, if black takes that knight, they will lose 57% of the time.
Let's continue. On move 7, I push a pawn giving up more material... still theory. It's to deflect the queen, and unless you're psychic and can calculate what is going to happen to you if you take it, you take the free pawn. So while I'm at it, let's sacrifice another piece. I mean I'm already an idiot and down a bunch of material, why not give up a bishop? The king takes the free piece and puts itself in line with the queen. White wins 60% of games from here.
I push the pawn to d4. How dumb is that? Not only am I giving up another pawn, but I'm giving up another pawn *with check*! Pure idiocy. If the queen takes that pawn, it loses 65% of games. This is still theory. White is down 8 points of material, and wins 65% of the time. Yep, people have found all of those over the board without the aid of theory.
Move 10 I move the bishop to attack the queen. I can do this because the pawn is pinned, so the black queen retreats to the most natural square defending the king. Move 11 I move out the knight. But you're hanging the bishop, you say! If black takes that bishop, that free bishop, they lose 81% of the time. My opponent takes the free bishop. Black is up two bishops, one knight, one pawn, and is dead lost. Still theory and Stockfish reads +8.2. I've done everything wrong and my opponent is dead because I know the theory of the Muzio to move 14. The rest is just a series of windmills and black is toast.
Do I play the Muzio every time? No. That would be boring. But it's a hell of a lot of fun to pull off OTB and look at your opponent's face as they think you're some kind of moron giving up all of your pieces. If you're black, you need to know the theory to survive this, and you need to not make the most seemingly obvious moves.
Ok so i tried my best to figure out how i would play against it and i did see that d4 move but understood that the idea was to activate the bishop so probably would not have taken the pawn.
Anyways, i think for blitz / bullet it can be very useful but in longer time controls this opening would likely be disastrous for white
DrJetlag wrote:
kartikeya_tiwari wrote:
DrJetlag wrote:
kartikeya_tiwari wrote:
DrJetlag wrote:
kartikeya_tiwari wrote:
I think that just goes to show how useless studying openings really is since u can basically figure out everything on your own if you just see threats and have good lookahead(which i don't but still played the "mainline").
Yeah, because reinventing the wheel is more efficient than building on collective experience. You probably also came up with 2. c4 in the Queen's Gambit (yes, it has a name and others played it before!). But the reason couldn't have been that it challenges d5 and allows white control of the center, that would be positional thinking and we're not masters yet.
Well yes... that is indeed my point... u can play c4 just based on logic without needing to learn the theory of it and memorizing it's lines.
"Collective experience" is 100% useless since at lower ratings you are not going to find people who play 15 lines deep in mainline ruy lopez(the only opening i know the name of lol. This is exactly how lower ratings are. I don't know any opening but one doesn't "need" to know the opening theory since he won't face any opening theory.
Games at sub master levels are won or lost based on tactics. They are completely based off of oversights, mistakes, calculation errors etc.
Let me give u an analogy. Let's say that u are teaching some guy calculus... he doesn't know how to add numbers but u are advising him to learn the equations of integration... would you do that? no right? you work off of "BASICS" first and till one has mastered the basics he won't need any higher level stuff.
Just funny to me that people who hang knights every move are being told to memorize opening theory.... doesn't make sense
Stick to one argument, for f's sake. One moment you talk about "below master", and the next you talk about "not being able to add numbers". There is a whole continuum between that. Then you keep equating learning theory with memorizing 15 move main lines, which is not what people have in mind when they talk about theory. I also don't believe that you came up with c4 and never heard the name "queen's gambit" before. In any case, this is a purely positional move. With a less dogmatic attitute you could maybe have made some progress in the last few years.
Bro anyone below masters is basically like "cant add numbers"... 1000 rated players can't add 2+2 and 1800 players can't add 2.3e + 3.2e .. that's the only difference.
Point being, just because people who are very low rated make huge blunders and people who are slightly above at 1600+ make less obvious blunders, doesn't mean we they don't make stupid blunders.
The games of 1600 players are also lost by calculation mistakes just like the games of 1000 rated players, it doesn't matter if they miss a 3 move combination or just hang their knight. The "CAUSE" of defeat is the same.
A youtube channel "Chess Vibes" who is a titled player did this exercise. He looked at 100 games of players all the way from 800 to 2000.... he found out that the overwhelming majority of games had "tactical oversights" as their main cause of defeat. I believe blunders + tactical oversights made up more than 65% of the defeats of the games of 1800 players with bad endgames were the next in line with 20% of the games decided based off of that, next in line was time. Openings made up less than 5% of the reason for the loss.
So yes, this just proves that the main issue lies with calculation and tactics. From this study it became clear than 1 out of every 20 games a 1800 rated player loses is due to bad opening and 13 out of 20 are lost due to tactical mistakes / blunders.
Guess how many games were lost in openings by 1200 rated players? 0
But the fact that most games are lost by tactical mistakes is as obvious as the fact that most goals in football are made by defensive mistakes. Even at the highest level. You are deliberately missing the point: you can lower the likelihood of defensive mistakes by having good positioning.
I win a good percentege of my games as white using the Greek gift sacrifice. The direct reason is that my opponents didn't see this and therefore didn't defend against it, which can happen in shorter time controls. But the more fundamental underlying reason is the fact that they usually misplay the queen's gambit and let these positions arise in the first place. I win 65% of my games as white after 1. d4 d5 2. c4 Nf6, which is a positional opening mistake (and this happens quite a lot). So while technically, and on an individual basis, the majority of these are decided by a tactical mistake, it is a positional mistake that loses these games statistically.
I won't call d4 d5 c4 Nf6 an "opening" mistake since it's pretty obvious why it's not so good. cxd5 Nxd5 and white gets to play e5 with tempo... this is not really an opening mistake, it's more of a calculation mistake.
That aside, having a good position does not really matter much since tactical mistakes can happen out of nowhere. Just missing a single move is enough to lose the game. At lower ratings you can constantly be having a good position but u will just "miss" something and the opponent will win. It's more about who made their last blunder. Strategy, positoning, opening nothing really matters at sub master level, it's just wild west of blunders, weak moves and horrible tactics.
NikkiLikeChikki wrote:
Exactly as Jetlag says. I believe it was Fischer who said tactics flow from superior positions. If you get a good position out of the opening and your pieces are on good squares, tactics become more available. You can say that x% of games are won on tactics, but are they really? That’s like saying that the assassination of the archduke started WWI. Lots of things had to happen in order to allow that event to set off the powder keg.
I mean the master who looked those games was constantly looking for if one player got a big advantage out of the opening and if that was the case then he put those games under the "opening mistake" category. It just doesn't happen often in longer time controls. We are talking about weak players here, do you think any of us could adequately punish 1.h4? i don't think so
Feb 1, 2022
0
#226
Don’t matter much because they can come out of nowhere… again, this is dumb. It’s like saying wearing a seat belt doesn’t matter because you can die anyway. It’s like saying getting vaccinated doesn’t matter because you can still get sick.
Probability vs. possibility… look into it.
Feb 1, 2022
0
#227
Wait. What? You just totally contradicted yourself! On the one hand you say that opening theory doesn't matter below Master level, and then you turn around and say it's a mistake that no master would make? Hello?!?!?! Duh! Someone who is lower rated is more likely to fall for a mistake that an opening makes possible. You JUST admitted that openings matter. Yay! I win!
Feb 1, 2022
0
#228
By the way, you're just wrong about your facts. Nakamura beat Andreikin with the Muzio Gambit in 2010. Both players were over 2600. Sure it was a blitz game, but if Nakamura had 1 minute and you had an hour, you'd still get checkmated in the Muzio.
Feb 1, 2022
0
#229
The way I see it: the more your positional play improves, the more tactics tend to work for you, and the less they tend to work against you.
And positional play doesn't have to be obscure, grandmaster-level ideas. In every main opening, there are common ideas to keep in mind, general pawn structures to remember, and thematic piece-placements to aim for.
These can start, at a basic level, as things to watch for in the opening. Then it can extend (at a more advanced level) to things to aim for in the middle-game.
A player who is aware of these general ideas is likely going to do better than the player who isn't.
And if you work on improving your knowledge of these positional ideas, you'll find that sometimes (not always, but sometimes) winning tactics will fall into your lap, because of it.
This doesn't mean that tactics are no longer important. Your tactical vision is still a main "weapon" of yours. Improving your positional understanding is simply a way to improve your knowledge of where and how to aim that weapon.
It's like a captain saying, "That hill is crucial to our stronghold! Make sure nobody gets to it!" The gunner has a plan to follow.
As opposed to another captain saying, "Just fire your weapon at anything that moves!" Then the gunner starts shooting at birds in the sky. 
NikkiLikeChikki wrote:
By the way, you're just wrong about your facts. Nakamura beat Andreikin with the Muzio Gambit in 2010. Both players were over 2600. Sure it was a blitz game, but if Nakamura had 1 minute and you had an hour, you'd still get checkmated in the Muzio.
It was blitz and i have already said that opening memorization only matters in blitz / bullet.
Also, no... nakamura can't checkmate me if he had 1 minute vs mine 60 minutes in that gambit line
Feb 1, 2022
0
#231
Ok. I went over to the Lichess database and looked ONLY at classical games, because you can do that there. The Muzio Gambit has been played in 2219 classical games and white has won 51% of those. In one of those classical games, a certain artem-us (rated 2311) defeated a certain TOMNS (rated 2099). In another, a certain Stop_rob (2315) defeated Vikas043 (2203).
2219 classical games and white wins 51%. Sorry, I've got numbers that directly refute your claim and you have wind.
And I still think you get checkmated by Nakamura....
Edit: even funnier: I looked at daily games! Even with all of that time to think, black only wins 51% of games. Your exact words were "it would be disastrous for white." Black winning slightly over half the time isn't great for white, but it's hardly disastrous. In the same time control, if white plays the Accelerated London and black responds with 1...Nf3 2...c5, black wins 55% of the time, so it's better than the London System. Disastrous would have to mean way more losses than wins. In the Muzio, a certain Yaskoetemovic (2173) defeated a certain Aleksi98 (2013) by checkmate in 15 moves... in a daily game.
NikkiLikeChikki wrote:
Ok. I went over to the Lichess database and looked ONLY at classical games, because you can do that there. The Muzio Gambit has been played in 2219 classical games and white has won 51% of those. In one of those classical games, a certain artem-us (rated 2311) defeated a certain TOMNS (rated 2099). In another, a certain Stop_rob (2315) defeated Vikas043 (2203).
2219 classical games and white wins 51%. Sorry, I've got numbers that directly refute your claim and you have wind.
And I still think you get checkmated by Nakamura....
Edit: even funnier: I looked at daily games! Even with all of that time to think, black only wins 51% of games. Your exact words were "it would be disastrous for white." Black winning slightly over half the time isn't great for white, but it's hardly disastrous. In the same time control, if white plays the Accelerated London and black responds with 1...Nf3 2...c5, black wins 55% of the time, so it's better than the London System. Disastrous would have to mean way more losses than wins. In the Muzio, a certain Yaskoetemovic (2173) defeated a certain Aleksi98 (2013) by checkmate in 15 moves... in a daily game.
Lichess ratings are very inflated so i won't look too much into that. 2100 in lichess is like 1800 in chess.com and i have said several times that anything below masters is fair game, nothing can be said about those games since people make blunders every single move.
I have never ever seen muzio but if u want we can play an unrated 45+45 game with u playing the muzio. U know the theory while i don't but i think it's still a bad line for white since white player is relying heavily on his tactical abilities.
One another thing is that in no titled classical games u will find muzio being played, not in a serious tournament... there is a reason for that.
Feb 1, 2022
0
#233
kartikeya
Here is a game I played against a lower rated player in 25+5. Of course, white lost the game due to a tactical error, namely 28.Nd3?? but his position was lost when he played a4.
However, my opponent let me equalize very, very easily early on in the game, and black had a very comfortable position. I used four minutes of my time, while my opponent was in time pressure (under 1 min) by the time he resigned.
These kinds of mistakes make your position unnecessarily difficult, which increases the chances of mistakes occurring. In my game, white continuously traded pieces down, which is exactly what black wanted to happen. In the end, there were too many infiltration points into white's position due to his large space advantage.
Stil1 wrote:
Grandmaster Grigoryan has an interesting article about what he calls the "myth" of solving puzzles for chess improvement.
His main argument, simplified: many players recommend that others "do puzzles!" to improve at chess. But while solving puzzles can be fun, and while they will help improve your tactical vision, they won't directly help you learn how to properly play chess.
Furthermore: spending too much time on tactics, in lieu of other chess elements (positional learning, for example, which Grigoryan refers to as "strategy"), can be more harmful than helpful.
Here's the full article, for those interested:
https://chessmood.com/blog/the-myth-about-chess-tactics-and-solving-chess-puzzles
What do you think? Do you believe that chess puzzles are the end-all, be-all for chess improvement? Or do you agree more with GM Grigoryan's perspective?
I agree more with GM Grigoryan's perspective
Dec 29, 2024
0
#235
I find puzzles useful, but my study is not exclusively focused on puzzles. Currently I’m reading Thomas Engqvist, 300 Most Important Chess Positions with the intention to complete the book in two months. Engqvist recommends a full year and already I have ample evidence that my rapid process courts trouble.
His 300 positions cannot be solved tactically. They require (and develop) positional understanding.
Dec 30, 2024
0
#236
The #1 problem with using puzzles for improvement is that with a puzzle you already KNOW that a winning tactic exists. You just need to spot it.
In real life, there is no such assurance. You aren't warned that a winning line exists, and you might either miss it entirely or (just as bad) waste clock time looking for a tactic that isn't even THERE.
blueemu wrote:
The #1 problem with using puzzles for improvement is that with a puzzle you already KNOW that a winning tactic exists. You just need to spot it.
In real life, there is no such assurance. You aren't warned that a winning line exists, and you might either miss it entirely or (just as bad) waste clock time looking for a tactic that isn't even THERE.
A player who solves puzzles regularly has a much better chance of spotting a tactical shot in a game. Much better. Sure, there aren't tactics in every position, but there's still a lot in any game.
Optimissed wrote:
Very often, with the rather artificial puzzles that are presented here, alternative winning lines exist which are easier to spot and therefore easier to execute, but which take more moves to complete.
Almost all of chess.com's puzzles are from real games. A very small minority (way less than 1%) are from chess compositions.
In a similar fashion, there are practically no puzzles where there's an alternative winning line. Surely no more than 1 puzzle in 10 thousand.
If you play three 90 min time controls each day. I think there is no need to solve tactical puzzles.
At the end of the day like @nklristic said - alot of it comes down to talent and style of play.
I have a friend who's 1700 and literally never learned opening lines or solved puzzles. Never. He's just talented. He was a casual 1200 before covid... His tactical vision and opening play is just sublime, he's super solid and never misses anything when we play, he could easily be 2000+ if he worked at it...
Very often he gets slightly worse positions out of the opening but he always fights back.
Me on the other hand, I was probably 800 for quite a while, and I needed to study tactics and opening theory to get to my modest 1550... (I started at age 28 and he is playing pretty much his whole life but very casually)
What's possible, and time will tell, is maybe I reach 2000 in a few years due to my work ethic and he stays at 1700 because he could never bother to do anything but play... Who knows?
So yeah, studying tactics, opening theory, strategy, endgames and master games are ways to maximize your potential, and everyone has different potential, due to age, talent, motivation, life status etc... Like a previous comment said, someone with the potential to be 2500, can very likely reach 1800 without solving puzzles or knowing theory, but he may very well have no motivation to even approach maximizing his potential, cuz what's the point, actually?...
I for one just felt I did want to push myself, because my of an inner feeling of dissonance that my understanding of the game was similar to a 1500 and yet my play was sub 1000 for a while, so I was like "hey, I wanna be 1500"... and guess I'm happy... I see no point in naturally talented 1800s trolling people who had to work for it claiming "it's a waste of time and you can reach this rating just by playing and anlaysing"... it's childish and annoying.