Click Here To Show Diagram Code

[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X X O . . . . . . . . . . . . . . |
$$ | . . O X O . . . . O . . . . . . . . . |
$$ | . X X O . . . . . , . . . . . , . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

My comment on this joseki has been that the black group, which is mainly on the inside, has the thicker shape so that the slightly disadvantageous territory/influence ratio can be tolerated.

I would think that normally joseki is shown with the move orders rather than just commenting on the final results.

_________________
http://tchan001.wordpress.com
A blog on Asian go books, go sightings, and interesting tidbits
Go is such a beautiful game.

Figuring out the move order is an exercise:)

I'd like to ask a general question about your method to evaluate josekis (or if it's a joseki).
As I understand it you assign values for territory, influence, etc.
When you have all the single values can you then compute a value for the whole (e.g. by adding or multiplying them) ?
With "value for the whole" I mean something that a bigger value implies a better result.

Click Here To Show Diagram Code

[go]$$B The 21st Century Dictionary of Basic Joseki
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 9 . . . . . . . . . . . . . . . |
$$ | . . 2 B 8 . . . . O . . . . . . . . . |
$$ | . . 3 4 . . . . . , . . . . . , . . . |
$$ | . . 5 W . . . . . . . . . . . . . . . |
$$ | . 7 6 0 . . . . . . . . . . . . . . . |
$$ | . . 1 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

Click Here To Show Diagram Code

[go]$$Bm11 The 21st Century Dictionary of Basic Joseki
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 7 X 6 . . . . . . . . . . . . . . |
$$ | . . O B O . . . . O . . . . . . . . . |
$$ | . 1 X O . . . . . , . . . . . , . . . |
$$ | . . X W . . . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . . X 2 . . . . . . . . . . . . . . . |
$$ | . . 3 4 . . . . . . . . . . . . . . . |
$$ | . . 5 a . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

Note: Black does not need to answer at but can extend to . But after W , Black must be able to cut at 'a'.

_________________
My "guide" to become stronger in Go

asura, as soon as the local territory count T and the influence stone difference I are known, calculate |T/I|, i.e., take the absolute of the quotient. Simply speaking, values from 1.5 to 3.5 are ok; for greater or smaller values, there must be a compensation for the player having a disadvantageous value. Note: if the stone difference is not 0, more theory is applied.

Thanks, SoDesuNe. The position did not ring a bell with me.

Click Here To Show Diagram Code

[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X X O . . . . . . . . . . . . . . |
$$ | . . O X O . . . . O . . . . . . . . . |
$$ | . X X O . . . . . , . . . . . , . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . X a . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

My first thought upon seeing this position was, That's joseki? It looks good for White to me. And it turns out that Suzuki-Kitani say that it is somewhat favorable for White.

My second thought was, Why didn't White push at "a"? The modern sequence has White making that play. Guess I'm not over the hill yet, eh?

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Not sure this is the question, but the reason for white making the 14 for 15 exchange before playing 16 is that if he plays 16 for 17 first, then black can hane at A (instead of 15) after white push at 14.

@Robert: For a quotient the range between 1.5 and 3.5 seems quite big. So the main application is to exclude bad variations quickly but not so much to find the best variation inside this range based on the value of this quotient?

There are different types of josekis etc., so there is also quite a range of values. (If one side makes territory in the corner, the ratio can be even bigger.) Yes, you might say that it is a first filter of detecting and discarding the truely mediocre and the bad variations. Afterwards, it is still an independent task to choose the best joseki / almost-joseki / variation in a given positional context.

I have just a beginner's question: For the black territory, I would count approx 15 points (depending on the endgame(?)). However I'm not sure how the number I for influence is determined. Could you give a numerical example for this position?

Another aspect may be the aji of the dead C17 stone, which seems to be at least good for three successive ko threats (e. g. B18, C19 and E19 if black replies on B17, D19 and C17, respectively). If there is a big ko fight going on elsewhere, white has the potential to gain about 15 points here (the three black stones + a few points of eliminated corner territory). Is that also considered in the numerical evaluation?

I don't know RobertJasiek's method but I like one John Fairbairn presented in Games of Honinbo Shuei vol. 3. It was proposed by Abe Yoshiteru 9-dan und looks like this: (n*(n+1))/2.
n are the number of stones from the third line, which build the wall. The idea is to also count forcing moves which would add stones to the wall.
This Joseki would therefor give White at least 15 points ((5*6/)2), with the - exchange, White has 21 points ((6*7/)2). For Black I would count 13 points without the exchange and 15 with it. So it's clear why e.g. Suzuki-Kitani say, White is favourable.

_________________
My "guide" to become stronger in Go

It might worthwhile to note that
n(n+1)/2 = 1+2+3+...+(n-1)+n = sum of first n integers

Possibly wortwhile.
Speaking just for myself, though: What the ♥? : D

_________________
My "guide" to become stronger in Go

Click Here To Show Diagram Code

[go]$$B influence stones
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X X O . . . . . . . . . . . . . . |
$$ | . . O X W . . . . W . . . . . . . . . |
$$ | . X X W . . . . . , . . . . . , . . . |
$$ | . . X W . . . . . . . . . . . . . . . |
$$ | . X O W . . . . . . . . . . . . . . . |
$$ | . . X W . . . . . . . . . . . . . . . |
$$ | . . B . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

The 1 black and 6 white influence stones with significant outside influence are marked. The influence stone difference is 1 - 6 = -5.

Black has, I'd agree, 15 points of current territory. Counting White's current territory is trickier; I count 8 points; this makes the territory count 15 - 8 = 7 points.

Then the ratio is | 7 / (-5) | = 1.4.

***

Abe's formula is only a rule of thumb, which amounts to drawing a sector line and counting the intersections inside the created imagined triangle. Instead, determining territory should rely on imagined reduction sequences.

There are several things wrong with this, all showing RJ has not read Abe's book - usually a good reason for not criticising or caricaturing it.

First, Abe counts the value of the thickness, not the territory, and so the extension stone does not come into his equation. Ultimately, the value of the thickness has to come down to territory points, but not necessarily in their own vicinity. Therefore his method is not so much a means of evaluating a local exchange (though it can do that) but of providing help in evaluating the game as a whole.

Second, his method cannot be characterised as drawing a sector line dividing a rectangle because the point is that he allows wraparound and irregular walls facing two or even three directions (and also allows for intersection of walls).

Third, although with an entirely difference emphasis he does allow imagined sequences to influence the count. You'll have to read the book to see how.

As a general point, I think we need to remember the hoary chestnut of the difference between precision and accuracy. If I understand RJ's methodology correctly, he is concerned primarily with precision, that is repeatability of his results. Even if he achieves this, they may not be accurate or too local. Abe's method probably has very low precision but he must regard it as being accurate enough to publicise it, and he is recommending it as a whole-board measure. If go is like a dart board where you are given six arrows and you need a 20 to start, using RJ's type of method seems to suggest you might be able to score triple 18 six times of six - close but not close enough, and you would never be able to start. Using an Abe type method, you may drop one dart in your foot and pitch another into someone's beer, but the rest are dotted around the top half of the board and - hey presto - one hits the 20! Game on!

If that happens consistently and pals decide to set up a pub darts team, the sometimes errant tosser is much more likely to be picked over the metronome. Since go, like life, is in practice essentially a probabilistic game where accuracy does better than precision over the long run, pros will presumably always favour the accuracy of rules of thumb (which of course they can tweak on the basis if experience).

It is also the equation for what are called triangular numbers. And, indeed, there is a triangle involved.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

I have not studied neither of the methods, but following your darts analogy, and taking correction on the fact that Go is not darts, I think the question to ask here is:

In your positional evaluation, is it generally better to be

• consistently off by, say, 2 points, or
• sometimes spot on, sometimes off by 20 points, with anything in the middle at other times?

_________________
- Bantari
______________________________________________
WARNING: This post might contain Opinions!!

Is it better to be:

[qoute]consistently off by, say, 2 points, or
sometimes spot on, sometimes off by 20 points, with anything in the middle at other times?[/quote]

I am not on familiar ground with such questions but i suspect if you load them in a different way (e.g. Consistently off by 10 points) you get a different answer. You also need to determine consistency. Thrice in a game might be the usual number of evaluations in ordinary games.

But the question does not apply in darts. If you need double-20 to finish and are consistently hitting double-1 next door, you'll never finish. You'd then be better off closing your eyes and hoping fir the best. Or sinking a few more pints until you lose control

I also have not read Abe's book, but the formula n(n+1)/2 is mathematically equivalent to drawing a 45-degree line from the wall and counting the enclosed triangular area. For this joseki, the triangle might look something like this:

Click Here To Show Diagram Code

[go]$$B
$$ +---------------------------------------+
$$ | . . . . . o o o o o . . . . . . . . . |
$$ | . . X X O o o o o . . . . . . . . . . |
$$ | . . O X O o o o . O . . . . . . . . . |
$$ | . X X O o o . . . , . . . . . , . . . |
$$ | . . X O o . . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . X a . . . . . . . . . . . . . . . |
$$ | . . b , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |[/go]

JF, would Abe claim that this joseki is favorable to W, because he gets equivalent 15-point territory from the wall, without even counting the valuable extension stone? Or is there some penalty for the cutting point aji which makes the wall less than perfect?

If I understand correctly, Abe would claim that the exchange a-b is worth 6 points for W (because the wall height goes from 5 to 6) and 2 points for B? That would explain why the diagram is joseki and the extra push makes it better for W, if B responds submissively.

RJ, after the a-b exchange, would your T/I ratio go from 7/5=1.4 to (7+2)/(5+1)=1.5, indicating the exchange is not good for W? That would be clearly wrong. Or would you add some points to your estimate of W territory, in addition to adding an extra influence stone?

Heh... sure - I understand this.
And did not mean to load the question, your darts example suggested an error of anywhere of 0-20 on one side and consistent error of 2 on the other, from what I understand.

I think that generally it is an interesting question what levels of consistent error out-weight what levels of permanent uncertainty. Is it 2 vs. 0-20? Or 5 vs. 0-20? Or 2 vs. 0-50? Or does it ever? Is there a line?

In either case we should strive to lower both the consistent error and the uncertainty, but being humans there is always some of both in us. Just saying...

_________________
- Bantari
______________________________________________
WARNING: This post might contain Opinions!!