GM Grigoryan on the "Myth" of Solving Puzzles

So, there's a theory I've heard put forth by a number of chess educators and high-level players about the value of puzzles being mainly training for pattern recognition.  In this model, the value of puzzles is far less to practice slow, deliberate calculation (a skill set that for many people seems to not improve very rapidly with practice) but instead to practice doing easier puzzles faster.

The goal, ultimately, is to move increasingly complicated tactics out of the realm of calculation and into the realm of instantly recognizing board positions that lead to them.  The hope would be that in real games, these pattern recognition skills would replace calculation and make finding tactics easier.

I'm not at a level of experience or skill to be able to judge this idea independently, but I will say that starting to practice puzzles this way has led to seeing certain things in real games that I would not have seen before.

cool

Sure there is a rationale. In a normal position there are many different ways to think about it: you could use your memory, you could use analogy, you could use positional judgement, you could use calculation. In a puzzle, you can only use calculation. It's more specific, it's targeted, it trains the particular thing you are interested in training. Your position reminds me of the old days when it was argued that athletes should not lift weights, they should just train at their own sport. Nowadays even golfers lift weights as a routine part of their training.

nowdays i can see, and play against, many youngsters who are really great in tactics, they see motives, they execute combinations like top class masters, when it comes to positional, when game is calm down and demands proper, deeper, strategy they fall down in just few moves. particularly noticed this at regular club otb blitz and rapid tournaments.

RE: the value of puzzles being mainly training for pattern recognition

I wouldn't say mainly. That's one use of puzzles, but they can be used to train many different skills in chess, provided they are well selected for the skill being trained, e.g. pattern, depth (look-ahead), breadth (short-term memory), imagination, etc. In fact pattern and imagination are at odds with one another, pattern is something you have seen before, and imagination is thinking outside the box.

Yes, it's common. What is also common is some of these youngsters go on to become GMs, whereas we do not. So we should take whatever points we can get now, while the taking is good.

3600 puzzles pb, 30/40/72 rush pbs.

my game stats are all 1600+ average & climbing:  i believe the 11,000 puzzles i've attempted have certainly helped these scores to improve.

training with attempting to find a tricky solution in a position that you KNOW is there increases pattern-recognition, planning, determination, creative-thinking, calculation & evaluation skills, while reinforcing the excellent habit of patiently searching for the best move every time it's your turn to play.

i'm down to 8 achievements left to get so when only 'puzzle addict' (solve 10,000 puzzles correctly) is left i'll probably drop off the radar for another year.

the problem i see with a lot of these conversations is a 'level' issue... you can't properly understand the value of solving puzzles until you really start trying to pass about 60+ survival or 3500+ rated ones.

it stands to reason:  if you can find the correct sequence of moves in a position that 99% of other people can't, wouldn't that translate well to real tactical chess skill in actual games?

just some observations from my experience.

To be fair, one person did say (and says the same thing every chance they get) that studying openings was useless, which is a form of training. I have been hard on puzzles, but I never said they are useless, just that the time invested vs. the expected gain in number of wins is small. I still think analyzing games, or even studying openings is more useful. By studying openings, I don't mean memorizing lines, I mean studying openings: "why is this a good move? what does it accomplish? what plan does it further? what are possible responses? why are those responses good? why are they bad?" You can learn a lot from studying long lines and reading the annotations.

Should you do some puzzles? Sure, but don't expect them to turn you into a tactical wizard or radically improve your rating. Most people are good at puzzles because they are good at tactics, they didn't magically become good at tactics because they did a bunch of puzzles.

That one person is me and yes, studying openings is 100% useless before a certain level(which i think is the master level). So far in slow games i know exactly 0 openings and have reached 1800 so for atleast 1800 studying openings is completely useless. Will it be useless even for 2000 rating? i don't know, if i am able to reach there then i can answer that better

There are people who get to almost 2 000 chess.com rating just by playing games. Does that mean that everything but playing games is useless below 2 000 rating?

eh what? this comparison doesn't make any sense. I am saying is that if i can reach 1800 by not knowing literally ANYTHING about openings then so can anybody and it just shows how useless openings really are. I would have been higher if only my tactics weren't so horrible. Every game i lose is because of a missed tactic, never once have i felt that "oh man i wish i knew this opening line"

Chess isn't 100% tactics. "Tactics" refers to specific types of move combinations (forks, skewers, pins etc) which win material or checkmate by force. Calling everything in chess "tactics" just makes the word meaningless. 

As for openings, of course it's valuable to study how to play the first third of the game at any rating level, although some people focus too much on it. Every type of study whether it's openings or tactics puzzles is useful as long as you don't overdo it and ignore other elements of the game. I think a good rule of thumb is to make sure you're studying at least 3 different aspects of chess (e.g openings, endgames and tactics puzzles) so you don't get too focused on one.

As I've said, then by that logic if someone gets to 2 000 just by playing games, he could say that everything but playing games is useless because anybody can get to 2 000 just by playing games. This is just to show you the flaw in logic.

But to be a bit less extreme. I've never touched master games before getting to 1 600. Does it mean that checking them out is totally useless?

I don't think so. I am sure there are many people out there that got to certain rating (say 1 500) without game analysis. It still doesn't mean that game analysis is useless before 1 500 rating.

There are probably even people who got to some respectable rating without doing anything related to tactics. It doesn't make tactical vision useless.

So the fact that you didn't do something up to certain rating doesn't make it useless, it just means that you personally didn't have to do it before getting to 1 800 rating, nothing more.

It's just false that studying openings is useless. In another thread I explained in great detail how my best win ever was because of the opening I chose and my prep. I've won so many games because I know way too much theory, knew without having to think that an opponent made a mistake, and punished it immediately. Had I had to calculate it I may have come up with it, but maybe not. I still can't believe that people are saying that it's useless. It's just straight-up not borne out by the facts. Besides, I'll ride with Daniel Naroditsky, who is in the top five leaderboard on chesscom and is a beloved educator over some guy who just can't understand the logic of if you know something and your opponent doesn't, you're at an advantage.

There is not one person on the planet who got to a respectable rating by not knowing tactics at all lol

There many chess teachers who claim the exact opposite, that openings are 100% useless. People like ben finegold who have been teaching for decades and even started their own chess clubs are among those. IM lawrence trent also said that to reach 2000 u just do tactics. That's just how it works.

So far i have  not seen this supposed "strength" as far as openings are concerned, if and when i see it, i will agree

And again in case you missed it, because you apparently did, studying openings yields the same kinds of benefits as analyzing games. I'm not talking about memorizing line, I'm talking about *studying* an opening, which means reading annotation and understanding plans and why some moves are good and some are bad. Unbelievable that people say it's useless.

I've repeated the story far too many times in too many threads to explain it again. If you really wish, you can message me and I'll cut and paste you the story.

"Plans" are useless too if one keeps on missing stuff.  That's why studying openings has no value for weaker players... they are not good enough to take advantage of some inaccuracy and will probably hang a piece as soon as they are out of their prep

Oh but I didn't say they do not know tactics, I just said that they might get to a certain level without studying tactics.

Just like for instance you might not have studied openings, but you have some knowledge of the openings simply by experience. 

Here: https://www.chess.com/analysis/game/live/34667369383

you get some 9 move theoretical line.

Or here: https://www.chess.com/analysis/game/live/35688383351

Pretty good  Maroczy setup, and a main line position. So perhaps you did not really study openings, but you can't say that you know nothing about openings.