Life In 19x19
http://prod.lifein19x19.com/

how to evaluate the reading result
http://prod.lifein19x19.com/viewtopic.php?f=10&t=861		 Page 1 of 2
Author:  entropi [ Tue Jun 15, 2010 6:15 am ]
Post subject:  how to evaluate the reading result
I have difficulty in assessing the local result in the opening and early middlegame. This is probably because I severely lack joseki knowledge and have nothing to compare the result with. How do you do it? Proverbs, intuition, experience, ...? Or is it obligatory to learn joseki just for this purpose (even if for nothing else)?
Author:  topazg [ Tue Jun 15, 2010 6:24 am ]
Post subject:  Re: how to evaluate the reading result
There's a difficult question. A lot of the value judgements I assign results are based on information I've picked up from various books written by professionals. It's partly intuition, not at all proverbs, not at all joseki, and partly a sense of direction for where the stones are supposed to go from that point onwards.
Author:  John Fairbairn [ Tue Jun 15, 2010 6:39 am ]
Post subject:  Re: how to evaluate the reading result
entropi: you might want to look at tewari. Essentially the answer is that if you have made no bad or wasted moves you can be satisfied.

In the specific case of joseki, evaluations often hinge on things like "has the opponent played on both sides", "has the inside player got more than 10 points in the corner", "has the outside wall got any defects", "is the outside wall facing the right way".

"Who has sente" is one aspect that many players overlook.

Ultimately, of course, the criterion is whether your strategic goal has been achieved, or the opponent's thwarted. Don't win the battle but lose the war.
Author:  GoCat [ Tue Jun 15, 2010 6:42 am ]
Post subject:  Re: how to evaluate the reading result
Yes, difficult question. I agree with topazg that part of the evaluation has to be based on intuition -- especially at my SDK-level play. This is one of the reasons I like to play thorough pro games -- it hopefully builds that intuitive pattern base to lead me into moves and sequences with a good result.

One of the difficulties I have is assessing the value of thickness and influence. I'm getting better at assessing territory, but in the early and mid game, just looking at territory is, of course, insufficient to evaluate the board. Again, I hope that looking at higher-level play (with commentary) is helpful. At this point when I read out a sequence, if the result involves thickness or influence, I am just barely able to see if that will be useful or not (e.g., end up with a wall that faces the wrong way). But I'm not to the point where I can assign a value to that result, and compare, say, to the territory gained by my opponent.

It's probably not obligatory to learn joseki, but I'm sure it's helpful. (I could sure use a bit more knowledge in that area!) I would like nothing better than to be able to choose a joseki for a situation, knowing the outcome I expect of it -- whether it is secure territory, outside thickness, or maybe just a stone that nullifies my opponent's influence. Learning and using joseki is hard! (Maybe time to reread Whole Board Thinking in Joseki.)
Author:  entropi [ Tue Jun 15, 2010 7:00 am ]
Post subject:  Re: how to evaluate the reading result
John, I read something about tewari but I have never tried it. I thought it requires quite some time to do and it is rather difficult to perform the analysis mentally. Is it also done (maybe by some professionals) during real games?

The questions like

Quote:
"has the opponent played on both sides", "has the inside player got more than 10 points in the corner", "has the outside wall got any defects", "is the outside wall facing the right way".


seem to be very relevant in that context.

Topazg, with proverbs, I meant simple things like "second line is the line of defeat, fourth line is victory, etc.". They are surely not enough for all situations but might eventually give a little hint.

The joseki knowledge could be helpful in the sense of giving you a reference to compare the result with. If you know your opponent has the same wall shape as in the joseki but you have more territory, it is already an indication (of course keeping in mind that more territory might also mean more bad aji).
Author:  topazg [ Tue Jun 15, 2010 7:03 am ]
Post subject:  Re: how to evaluate the reading result
entropi wrote:
Topazg, with proverbs, I meant simple things like "second line is the line of defeat, fourth line is victory, etc.". They are surely not enough for all situations but might eventually give a little hint.


I'm not saying they may be useful for some, but I've found that the stronger I have got the more often I can see where proverbs don't apply. As such, I just avoid them - I want to have a result which makes sense to me regarding the board as a whole, and not necessarily one that adheres to a proverb.
Author:  Chew Terr [ Tue Jun 15, 2010 7:17 am ]
Post subject:  Re: how to evaluate the reading result
Entropi, I was just thinking about this a day or two ago. I'm in the same boat you are. Only there might be even more leaks in my rowboat. =D

John, tewari analysis is one of those things I forget exists except for when other people bring it up. I'm going to try to look at some of my sequences the next time I do a self-review. If I'm not mistaken, tewari analysis is meant to be a tool for analysis to mine information for the future, correct? It seems like it wouldn't really be reasonable to try to analyze sequences in that way before you play them.
Author:  entropi [ Tue Jun 15, 2010 7:19 am ]
Post subject:  Re: how to evaluate the reading result
topazg wrote:
I'm not saying they may be useful for some, but I've found that the stronger I have got the more often I can see where proverbs don't apply. As such, I just avoid them - I want to have a result which makes sense to me regarding the board as a whole, and not necessarily one that adheres to a proverb.


I think that's a nice attitude in general. There are indeed so many exceptional situations that it is sometimes more difficult to judge whether a specific proverb is applicable, than just applying your own intuition. And blindly applying proverbs would be an obstacle for your improving as well, I think.
Author:  Violence [ Tue Jun 15, 2010 8:32 am ]
Post subject:  Re: how to evaluate the reading result
What I do is just feeling and intuition-based evaluation. Look at the result and try to see if there were good exchanges or bad exchanges that were made. As you get stronger, you get better at evaluating who is good and who is bad when a sequence is over.
Author:  Bill Spight [ Tue Jun 15, 2010 8:38 am ]
Post subject:  Re: how to evaluate the reading result
Chew Terr wrote:
John, tewari analysis is one of those things I forget exists except for when other people bring it up. I'm going to try to look at some of my sequences the next time I do a self-review. If I'm not mistaken, tewari analysis is meant to be a tool for analysis to mine information for the future, correct? It seems like it wouldn't really be reasonable to try to analyze sequences in that way before you play them.


Tewari is an example of evaluating the efficiency of the stones. It is straightforward enough to use during the game. :)
Author:  Toge [ Tue Jun 15, 2010 9:50 am ]
Post subject:  Re: how to evaluate the reading result
entropi wrote:
I think that's a nice attitude in general. There are indeed so many exceptional situations that it is sometimes more difficult to judge whether a specific proverb is applicable, than just applying your own intuition. And blindly applying proverbs would be an obstacle for your improving as well, I think.


- I've never thought that proverbs are obstacles to improvement for a thinking person. Understanding cuts and connections in small and large scale is food for thought: how to play as to prevent my groups being cut? Or "second line is a line of defeat" - what if you live by crawling there? It depends very much on possibilities afterwards. If living by crawling was the only possiblity, doing so is a big success.
Author:  Dusk Eagle [ Tue Jun 15, 2010 1:17 pm ]
Post subject:  Re: how to evaluate the reading result
I think the point of the proverb that says "The second line is the line of defeat" isn't to say: don't crawl if it makes you live. The point of it is to say, avoid situations where you have to crawl on the second line to live.
Author:  ethanb [ Tue Jun 15, 2010 5:56 pm ]
Post subject:  Re: how to evaluate the reading result
Helel wrote:
Don't follow proverbs blindly

Quote:
There is a distinction between a heuristic, which helps you find a solution to a problem in many cases, and a rule, which has a strict logical status.


I assume a dan player has already internalized all relevant proverbs and so no longer need the quiet reminder to think twice whenever one breaks them. Otherwise it seems slightly excessive to entirely disregard the wisdom of the centuries, well at least IMHO. ;-)


This from someone who says if he meets a dan player on the street, he will kill him?

(I understand and fully approve of the Zen nature of this comment, I just thought it was an amusing contrast :))
Author:  entropi [ Wed Jun 16, 2010 2:23 am ]
Post subject:  Re: how to evaluate the reading result
Helel wrote:
But I'm always right, even when I contradict myself. :mrgreen:


That's a nice proverb too :geek:
Author:  HermanHiddema [ Wed Jun 16, 2010 6:20 am ]
Post subject:  Re: how to evaluate the reading result
My opening evaluation consists of:

  • Balance of territory: who has more points?
  • Balance of power: who has more influence & thickness?
  • How do the above balances relate? Is it fair?
  • Do I have weak groups?
  • Does my opponent have weak groups?
  • Do my stones work together efficiently?
  • Can i get the last big point?
  • Do I have any good moves available?
Author:  HermanHiddema [ Wed Jun 16, 2010 7:07 am ]
Post subject:  Re: how to evaluate the reading result
Here's a recent opening of mine, I've tried to document my thoughts at every move.
Author:  Chew Terr [ Wed Jun 16, 2010 7:19 am ]
Post subject:  Re: how to evaluate the reading result
Thank you Herman, this is a good demonstation of the fundamental opening principles. My goal is to get good at this sort of thing so I can be ahead by move 50 in every game. Can't blame me for dreaming =).
Author:  HermanHiddema [ Thu Jun 17, 2010 3:59 am ]
Post subject:  Re: how to evaluate the reading result
Here's another game with my thoughts on every one of my opening move, this time against a 6 dan. I think this one is a little more interesting, as it didn't devolve into middle game fighting as quickly as the previous one. :)
Author:  John Fairbairn [ Thu Jun 17, 2010 4:54 am ]
Post subject:  Re: how to evaluate the reading result
Herman: A propos the lower left joseki in the second game, you may wish to look at the 1973-10-22 game between Kato and Takemiya which features it in two corners in a peculiar form of mirror go (the mirror occurs post hoc after both sides of the board are played out independently).

I think that game will suggest that White's strategy was flawed even before D4, but I also think you are not using "thick" properly. The fact that you felt compelled to play C14 rather than "approach on the wider side" suggests to me that you saw your supposed thickness (centre left) as weak, which is a contradiction.
Author:  HermanHiddema [ Thu Jun 17, 2010 5:06 am ]
Post subject:  Re: how to evaluate the reading result
John Fairbairn wrote:
Herman: A propos the lower left joseki in the second game, you may wish to look at the 1973-10-22 game between Kato and Takemiya which features it in two corners in a peculiar form of mirror go (the mirror occurs post hoc after both sides of the board are played out independently).


Thanks, I will have a look! I think I've seen the position from that game before, is it mentioned in one of the recent New in Go articles on mirror go?

John Fairbairn wrote:
I think that game will suggest that White's strategy was flawed even before D4, but I also think you are not using "thick" properly. The fact that you felt compelled to play C14 rather than "approach on the wider side" suggests to me that you saw your supposed thickness (centre left) as weak, which is a contradiction.


Yes, you're right. I was using "thickness" where "influence" would be far more appropriate. The two point jump is light and thin, so thickness was never really an appropriate term. I actually mention that it is thin in later comments, so I'm not even being internally consistent :)
Page 1 of 2 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/