Professional Game Study

[hailthorn011]

Let me offer my own insight as to why I thought this would be a good idea. Or, rather, go in depth on my original reason. It's not that I'm trying to decipher each move and figure out why it was played by the professional. Because, as has been noted, my reasoning would likely be very flawed anyway.

No, I'm trying to justify it from MY point of view. I look at it like it was a game I was playing. Why would I do that? Is there something else I would have likely have played? Would it have succeeded? Would it have failed?

At the end of the day, I can't entirely be sure. And I'll probably misunderstand the meaning of every move, from a stronger player's point of view, but I strongly believe there is something to gain from gradual pattern association, direction of play, and even life and death situations. And these are all factors that, in the end, could help me improve from 9k-10k.

However, I also do not seek to refute the points others have made. I know I may not gain too much from this exercise. But at the end of the day, it's really just for fun, right? I find a certain beauty in professional games, so even if I don't necessarily learn from them, observing them is 90% of the fun!

This is just the ideology I walked into this form of study with. Fun. Plus, I think it'd be kind of cool to be able to replay a pro game from start to finish from memory. I wasn't able to get much Go time in yesterday due to being insanely busy, but I intend to continue playing out the game. Currently, I've got the first 23 moves perfectly memorized.

[John Fairbairn]

Currently, I've got the first 23 moves perfectly memorized.

The idea of memorising games comes up over and again here, and invariably the goal is to see how many moves you can remember. That's the wrong goal. At least according to Kikuchi Yasuro, who may be among the very best go teachers on the planet, being able to teach from toddlers right up to top pros - Yamashita Keigo, the current Honinbo, is one of his charges.

Kikuchi uses memorisation in his teaching but the aim is not to see how far you can go, but rather to learn to play through the moves in your head with your eyes closed. He recommends starting with 30 moves, then you can try 50 or even 80 moves, but that's as far as it goes. You should also visualise life and death solutions in your head in the same eyes-closed way. Kikuchi says he regards this as a very important element in his teaching.

On the other aspects, broadly speaking studying skills like go can be broken down into two methods. There's the method of simply repeating elements for ten thousand hours. This is the only method that gets you to the top. It's obviously very hard work, and only people who can make that kind of effort fun are prepared to do it.

The other method is trying to understand. This tends to be fun for many more people than repetitive work. A sudden flash of insight can gives us a thrilling frisson. An extra fact tucked away in the memory gives a nice feeling of progress. But none of that leads to real strength. To take F1 as an example, there are plenty of people who, say, understand the mathematics of a car's aerodynamics, and others who understand the intricacies of the F1 rule book. These people have an important role in the sport, and can have great fun doing what they do. But they don't become racers. Their understanding and knowledge, however great, are irrelevant for that. The guys who become racers skip the maths and skip the rulebook. They just put in ten thousand hours driving cars.

Go players have to make a similar choice. Do they want to become knowledgeable or do they want to play like pros? Lots of fun and little work, or some fun if you're lucky and lots of work?

[Vesa]

I guess "the other method" is the life of a go amateur.

Cheers,
Vesa

[hailthorn011]

I guess "the other method" is the life of a go amateur.

Cheers,
Vesa

What's "the other method"?

[yawnguy]

On the other aspects, broadly speaking studying skills like go can be broken down into two methods. There's the method of simply repeating elements for ten thousand hours. This is the only method that gets you to the top. It's obviously very hard work, and only people who can make that kind of effort fun are prepared to do it.

10,000 hours? That's merely 27 hours a day for a year.

My understanding of becoming an expert in a subject includes learning thoroughly (and practising expertly in real-life situations, not in theory only) all the many, many minutiae that make up that subject. A master needs to be an expert in all aspects of the subject. So one could put in even 50 hours a day for a year ( ) and still miss if one did not also include all the (significant) angles.

Paul

[ethanb]

Kikuchi uses memorisation in his teaching but the aim is not to see how far you can go, but rather to learn to play through the moves in your head with your eyes closed. He recommends starting with 30 moves, then you can try 50 or even 80 moves, but that's as far as it goes. You should also visualise life and death solutions in your head in the same eyes-closed way. Kikuchi says he regards this as a very important element in his teaching.

The ability for strong-ish go players to do this is one reason I find the article on GoGod about the "only person who can play blind go" really weird and unlikely. Maybe it's just that nobody's really talked about it before. (The other reason is my own experience playing blind go... quite difficult, but definitely not impossible - a friend and I played out almost 120 moves one night when he had just reached 1d and I was still a very active player. The hardest part wasn't remembering the patterns - we'd played one-color go games all the way through before - but converting the decision of where to play into coordinates took longer than deciding where to play most of the time, for me anyway.)

[hyperpape]

"Just make sure you only review YOUR games when..."

(Emphasis added) Not sure what that means; everyone else here is talking about pro games.

Araban says to only review pro games if they've been commented. But if you just replace "pro games" with "your games" you can repeat the reasoning. But saying you should not review your own games is terrible advice. So I'm challenging him to tell me the difference. It's like a reductio ad absurdum, but I'm just using it to ask a question.

[Kirby]

I agree with hyperpape here. This is kind of the idea I was getting at earlier, but the example of reviewing your own games makes it clearer. I guess the only additional benefit that I mentioned earlier was that reviewing a pro game vs. your own game might give you some new and interesting ideas that you hadn't come up with yourself.

[Solomon]

"Just make sure you only review YOUR games when..."

(Emphasis added) Not sure what that means; everyone else here is talking about pro games.

Araban says to only review pro games if they've been commented. But if you just replace "pro games" with "your games" you can repeat the reasoning. But saying you should not review your own games is terrible advice. So I'm challenging him to tell me the difference. It's like a reductio ad absurdum, but I'm just using it to ask a question.

Isn't that the ideal scenario? To be able to review your own games that have been commentated by pros (or anyone stronger in this case, since we're talking about my games instead of pro games)? The problem is financial feasibility. When I used to take lessons from bigbadwolf, he would occasionally review some of my KGS games that I had already reviewed. It was mind-blowing how many more mistakes he picked out and how much more I learned thanks to his guidance. If I had the money to, I'd have this become routine and I feel I would improve much more rapidly if this could be done.

Just like how it's better to review your own game than not review it at all, I do think it's better to review pro games by yourself than not at all. However, I don't think it's efficient for reasons I explained earlier and given that financial feasibility is much, much less of an issue given that there are plenty of commentated pro games you can find online and in books, it makes little sense to review pro games without commentary.

[hyperpape]

Thanks. After reading your first post, I didn't see the nuance. It seemed like you were saying that reviewing professional games on your own is bad, instead of the actual point that it's just not worth doing because reviewing commented games is so much better.

[Kirby]

I guess I missed that nuance as well.

...
Just like how it's better to review your own game than not review it at all, I do think it's better to review pro games by yourself than not at all. ...

I agree with this, and I think that this is consistent with what most people here have been saying: It's good to study pro games that don't have commentary. It's even better if you can find pro games that have some commentary.

That sounds right to me.

[snorri]

I guess I missed that nuance as well.

...
Just like how it's better to review your own game than not review it at all, I do think it's better to review pro games by yourself than not at all. ...

I agree with this, and I think that this is consistent with what most people here have been saying: It's good to study pro games that don't have commentary. It's even better if you can find pro games that have some commentary.

That sounds right to me.

Pro commentary doesn't often focus on the questions you would have, at any given level. They only have so much space and so much time to write these reviews, after all. Even if they comment 10% of the moves, that leaves 90% for you to (potentially) misunderstand, and likely much of the other 10% if you don't understand the comments! It would seem to be a hopeless situation. But I suppose children learn language in an equally bad situation, and they manage somehow. I like pro games. I have learned many correct and incorrect things from them, but I still like them.