Learning vs improving

A 2005 study found that "serious study", the task of reviewing games of stronger players and trying to guess each move in advance, was the strongest predictor of chess skill. It found that grandmasters spent an average of 5000 hours of serious study in their first 10 years of serious play, more than five times more than intermediate players.

http://onlinelibrary.wiley.com/doi/10.1002/acp.1106/abstract

This is very different from what most people do: read chess books and learn new knowledge, but no "serious study". Are there two parts (learning knowledge, serious study), or is only "serious study" important?

1.e4 best by test

As with any skill in the world, knowledge alone is NEVER enough to excel. You have to see how it is applied. So I would say both are equally important and each requires the other to excel at any skill whether it is chess, painting or playing football.

Probably what makes it "serious study", is the fact that it is tedious and requires application of knowledge. So most people don't bother.

There is no contradiction here. You can't predict moves in GM game without knowledge - you will be unable to understand what is going on. And when reading chess book you can do it as "serious study" (guessing moves, playing example against an engine etc.) - this can help to convert your "knowledge" into "skill".

Serious study falls under the action of learning.

thx for posting. It's an very interesting article.

So 5000 hours of using knowledge is only for one component of improving, and many more hours of acquiring knowledge are required to reach a high level.

great so in the next 5 years i need to spend about 2.7hours a day to reach that 5000 hours of serious study...and add another, what, 2hours a day learning knowledge and playing some? so 4.7 hours of chess a day (lets round to 5) in five years. Guess I need to look for a sponsor for the next few years to fund my chess study. :P

@rtr : are you sure by 'serious study' the authors were referring to reviewing master games and guessing the moves, rather than 'active learning tasks' in general (ie. involving solving puzzles and/or analyzing various positions) ?

...and if serious study is what I've seen referred to as 'active learning', then it's definitely the most important part of study, as you're going to pick a lot of new knowledge this way.

I'm afraid it is true. All evidence I have seen supports it. A number of strong players recommend something like 4-6 hours per day for 3 years, others gave estimates that were in the 5k hours ballpark. This seems to be a minimum, and I'm sure learning and improvement must continue on from there. Even the best players and "child prodigies" seem to require the same effort before they are top players.

This seems to hold true for fields that are very competitive, like chess, music, day trading, and so on. However, if you want to learn some other skill like IT server and networking infrastructure, it is low hanging fruit. There are certainly people who work hard to be knowledgable about their work, but after a point they know enough to do their job and they stop learning. At that point, you are competing mostly with people who are spending exactly ZERO hours improving their skill, so you can be a top performer and make loads of money after years of hard work. Years of hard work in chess usually won't lead to being a top performer or loads of money.

In this and other articles/books, I have seen the term "deliberate practice" used, which I think is in line with what you call active learning. Maybe it's the same thing. Deliberate practice has a number of properties, such as (1) being an activity that pushes you just beyond your current abilities, (2) provides rapid feedback, (3) allows for many repetitions, and others. The "analyze-master-games-via-guessing" method is just one that I have seen written about in several places over the years, but good puzzles and key positions should fit in that category at least as well as random positions from master games. Maybe the full game analysis is relatively the most effective because it ties all of the key ideas/positions together, showing how they fit into the big picture of the game and getting to see how those ideas were executed in practice.

D_for_DJ wrote:

1.e4 best by test

I don't think so

This is true, application cannot happen without learning the required knowledge first.