Why is this an equal result in this joseki?

[Kirby]

The stone difference 1 can be compensated by 1 imagined white influence stone elsewhere on the board (modifier -1 in White's favour for the influence stone difference). We get:

Compensated influence stone difference = 1.

Thus, (locally) the position favours Black! The thread title is wrong.

I'm skeptical of this analysis every time I see it. I don't think it takes nearly enough into account. Do other strong players accept this method of analyzing a position?

Personally, I feel analysis of influence in particular, comes with experience. This is because, even if there is a proven way for measuring influence, not everyone is able to effectively use influence for attacking and profit. Because of this, someone that's not good at utilizing influence may be at a disadvantage when they play a joseki that gives them more outward influence and less immediate profit.

As players gain experience and learn more, their way of playing may change along with preferences, and people can come to realize for themselves how useful a position is for black and white.

tl/dr: Even if there is a "book answer" or a "correct" way to analyze a position, it would seem more practical for an individual to come to appreciate a position on their own from their own experience and skill. Eg. if you think a joseki is better for black, either: (a) learn why your opinion varies from what stronger players call joseki (b) Play that joseki as black

[RobertJasiek]

is it really that simple? You don't count for
[...] stability [...] it is a factor to be listed.
Also the cuttability of the white group at b15 seems a factor.

It is possible to produce much fog in an analysis by listing also all minor aspects. Stability is not a minor aspect, but here both groups are reasonably stable, although the black group is a bit more stable. B15, however, is not such a danger, because a white stone at C15 activates white F15, if Black tries to start a cutting fight. If Black first plays on the left side and White reinforces at C15, Black also needs to reinforce, so that White can then attack Black's first play on the left side. If, however, first a fight on the left side starts and White fails to defend the aji B15 indirectly, then White might have to answer B15 at A17; now imagine Black C15, and it is White's next turn in the supposed fight on the left side; i.e., it is not straightfoward for Black to cut. This is so, especially if assuming that the players chose to create the upper left corner shape, when Black must have wanted to develop the upper side, while White wanted to develop the left side. IOW, if White was weak on the left side when creating his corner group, the corner situation would be even more favourable for Black than what my earlier analysis says.

Black B15 creates a problem for White, but White C15, F13 or D14 creates a problem for Black; also his group is not perfectly stable. The problem with the white group's stability is greater though.

I'm skeptical of this analysis every time I see it. I don't think it takes nearly enough into account.

This analysis is an extract of a more fully worked out analysis, which takes into account also other factors (to possibly identify the value classes Territory Disadvantage Combined with Other Advantage or Territory Advantage Combined with Other Disadvantage), further options for conversions from stone difference 1 or 2 to stone difference 0, and the positional context.

E.g., in a joseki, White's significant territory advantage could be compensated by a disadvantage of his group's significantly smaller degree of stability. In our corner position, such a compensation is not given; so stability considerations cannot alter the Favourable for Black assessment of the value analysis.

The analysis considers stone difference, territory count, influence stone difference and the assumption of the players' expected intention of good positional embedding of their important corner groups as the major aspects and considers all other possibly significant factors as "the Other Advantages / Disadvantages".

When the other factors (stability, efficiency, development directions etc. etc.) are not significant or obviously reasonably balanced, an analysis can be simplified by forgoing the minor factors. Of course, if you want a full analysis, you need to consider each existing tactical variation and its impact for the analysis.

EDIT:

Do other strong players accept this method of analyzing a position?

The theory is rather new, most strong players live where there is no easy access to the theory etc., so I'd say that at the moment other strong players do not accept this method, e.g., because they do not know it at all.

When seeing other strong players' stated joseki evaluation, it often ignores the stone difference and amount / value of influence. Now, tell me: do you think that ANY analysis ignoring these two ultimately important aspects can be any good...? (Occasionally, one sees other stronger players saying something like "Black has played one stone more, so the result must be favourable for him." or "Black has (more) influence.". I.e., such statements are very imprecise.)

[emerus]

When seeing other strong players' stated joseki evaluation, it often ignores the stone difference and amount / value of influence. Now, tell me: do you think that ANY analysis ignoring these two ultimately important aspects can be any good...? (Occasionally, one sees other stronger players saying something like "Black has played one stone more, so the result must be favourable for him." or "Black has (more) influence.". I.e., such statements are very imprecise.)

Yes, Positional judgement in joseki or otherwise is easily the most discussed topic with my teachers. All of them are extremely quick to point out the stone difference, the overall balance, and the value of influence. I believe they rely totally on a combination of A. Reading, B. Their own experience, C. Borrowed experience from stronger players(pros).

You are trying a new approach to add some science to positional judgement, and from my perspective, even you do not rely on it as much as any of the aforementioned points. As you said, it's an extract.

....

Compensated influence stone difference = 1.

Thus, (locally) the position favours Black! The thread title is wrong.

I have a bad feeling when reading the conclusions. This is a strong statement where you use a new, inexact science as the primary evidence. Forgive me, I get that it is your opinion just like anyone else's posts here and it works for you and perhaps others, but whether intended or not, it seems like a shortcut to the traditional(?) methods. I am just interested in an expert second opinion before I buy in.

[RobertJasiek]

you do not rely on it as much as any of the aforementioned points.

For joseki / corner variation evaluation, I rely first of all on the connection, life+death and stability statuses; stone difference, territory count, influence stone difference, very significant other (dis)advantages (the fact that they exist for a player); my analysis method (which integrates the previous four aspects); the positional environment / development directions; the inherent strategic choices.

As you said, it's an extract.

Do you know my complete method? When I have said "an extract", I have referred to other parts of my analysis method. (In this thread, you see just a particular application of it.)

where you use a new, inexact science as the primary evidence.

I have not seen anything more exact for generally applicable joseki evaluation methods...! It is not as exact as pure maths, but small inexactness is part of the theory (e.g., it allows a value range from about 1.5 to about 3.5) and non-mathematical parameters ("a significant other (dis)advantage") are used.

However, I do not just throw random numbers around, but provide justification for 7 excess points of territory (which is related to the miai value 14 of early opening moves) or for 1 excess influence stone, when converting 1 excess stone. Besides, I have applied the method successfully to 399 of 400 josekis (and for the 1 remaining joseki, there is a very sound explanation why the method does not apply).

it is your opinion just like anyone else's posts here

Anyone else does not provide a general theory that succeeds for at least 399 of 400 josekis. Do your teachers' theories (what are they...?) succeed as well?

it seems like a shortcut to the traditional(?) methods.

The (best of the) traditional methods apply to (rather small) fractions of all corner variations. There is, AFAIK, no generally applicable traditional method that always distinguishes josekis from one-sided results.

I am just interested in an expert second opinion before I buy in.

Read the theory, compare its success of application to other theories and trust your own ability to understand what you see. Go theory is no longer for experts only, but go theory should be for everybody. During a game, you cannot ask an expert, but you need theory that you can apply by yourself. Count territory, count stones and express how much excess territory there is per excess influence stone.

[RobertJasiek]

emerus, my opinion in this thread is not "just like anyone else's posts" here. I explain this below. For the core of my opinion, see my earlier analysis. Since I (and everybody) has too little time, everybody's analysis does not mention and discuss all aspects. However, the following three aspects must be assessed, but everybody else (incl. the professionals in the GoWorld) fails to do so:

Everybody else's analysis:
- forgets the stone difference (a few mention it)
- forgets the territory count
- does not assess the amount or value of influence

gostudent's analysis:
- a partial count does not assess the territory count
+ correctly assesses that the black shape is thicker

Dusk Eagle's analysis:
- studies a different sequence and result instead of the given sequence and result, without conclusively allowing a relation

daal's opinion:
- speculates instead of assessing
- says that playing first in a corner is expected to give a slight advantage, but this is not so; there are josekis with perfectly equal results
+ it is correct to observe that the white group is not enclosed in the corner, but has access to the left side
- however, it should also be mentioned that such a feature applies in a territory versus influence / side moyo joseki if the territory count favours the corner player and is balanced with the opponent's influence, or does not apply in such a joseki if the territory count favours the corner player greatly and is not balanced with the opponent's influence (the player's territory advantage is compensated by the disadvantage of his group having no access to a side); here, the territory count is not favourable for the corner player, so it is insufficient to observe the white group's access to the left side; instead, every would-like joseki must fit in one of the functional classes, which here would be 'territory versus influence / side moyo' and have one of the aforementioned features; observing only the white group's access to the left side is insufficient information to assess a result's functional type behaviour
+ assessing some stability of the white group at all is correct
- however, there is aji also in the white shape, and the stabilities of the black and white groups should be compared if stability shall be considered as a significant aspect for possible differences between the groups

Takao's opinion:
Does he provide any justification?

SmoothOper's analysis:
+ it is correct to assess White's sente
- however, this is an aspect relevant in the global context; locally, having sente to play elsewhere does not give an advantage, but what matters locally is the stone difference
- White does not have "one or two forcing moves", but he has two
+ it is correct to assess that White has these forcing moves
+ some aji exploiting move is mentioned
- other such moves are not mentioned
- speculation about Black's global strategy does not assess a corner result

Uberdude's analysis:
- studies a different sequence without conclusively allowing a relation to the given sequence
- makes the wrong assessment that a sacrifice of two stones would be bad per se (there are josekis with sacrifices!)

Joaz Banbeck's opinion:
- it is a failure to reject analysing results with a player's stone excess
- a 'sente/gote differential' is something relevant for the global context, not for assessing the local result; for the latter, one needs to stick with the stone difference
+ it is correct that an unequal number of [played] stones complicates analysis
- however, this says nothing about how to analyse the given result
+ it is possible to analyse a position also by means of a non-joseki move
- however, the suggested non-joseki move is very suboptimal and so invalid for analysis; it is suboptimal because a) it adds only 6 points of territory, b) does not increase the number of white influence stones, c) the added development direction to the center is worth less than increasing the number of white influence stones by 1; +6 points and a too small increment of influence (added direction) mean that the move is too far from what it should be: a move with the miai value 14; it would be correct to imagine play of 1 influence stone elsewhere on the board, because such a move has an (idealised) miai value of 14 points; still the relation to Black's influence stones would be missing in the analysis

Otake Hideo, Haruyama Isamu, Kobayashi Satoru in GoWorld 68, p. 10ff:
- the result itself is not analysed
+ a ladder condition for creating a variation is mentioned
o other variations are studied and give at least a hint to a few strategic choices leading to the result; in particular, a resonable variation (Dia. 4) is given:

$$W
$$ ----------------
$$ | . . . . . . . .
$$ | . . . . X . . .
$$ | . 6 O O X . X .
$$ | 8 O X X O O . .
$$ | . 4 2 1 X . . .
$$ | . 5 7 3 . 9 . .
$$ | . . . . . . . .
$$ | . . . . . . . .

Click Here To Show Diagram Code

[go]$$W
$$ ----------------
$$ | . . . . . . . .
$$ | . . . . X . . .
$$ | . 6 O O X . X .
$$ | 8 O X X O O . .
$$ | . 4 2 1 X . . .
$$ | . 5 7 3 . 9 . .
$$ | . . . . . . . .
$$ | . . . . . . . .[/go]

Bill Spight's opinion:
+ he states the correct reference miai value 14
- guessing that the result is worth around 13 points and calling this slightly better for White is not justified by reasons and does not assess influence

cyclops's opinion:
+ stability can be considered
- many other strategic concepts etc. can also be considered in principle; the point is not to assess one particular other concept, but to consider every concept with a significant impact for the result

Kirby's opinion:
- experience does not assess influence in a manner that can be discussed well
- using influence well is not part of an assessment of locally created influence

EDIT: correcting typos.

[Uberdude]

Black's descent at 1 is a bit of a greedy move compared to solid connection at a, so if white sacrifices the 2 stones black has a somewhat nice result. The downside of the descent though is it means white can more readily save the two stones and fight with the sente push at 4 and then you'll get some big messy fight.

Uberdude's analysis:
- studies a different sequence without conclusively allowing a relation to the given sequence
- makes the wrong assessment that a sacrifice of two stones would be bad per se (there are josekis with sacrifices!)

I never said sacrifices were bad per se. I said that if you compare the solid connection and sacrificing:

$$B
$$ ----------------
$$ | . . . . . . . .
$$ | . 4 . . . . . .
$$ | . . O O X 1 X .
$$ | . O X X O O . .
$$ | . . . 3 X . 5 .
$$ | . . 2 . . . . .
$$ | . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ ----------------
$$ | . . . . . . . .
$$ | . 4 . . . . . .
$$ | . . O O X 1 X .
$$ | . O X X O O . .
$$ | . . . 3 X . 5 .
$$ | . . 2 . . . . .
$$ | . . . . . . . .[/go]

with the descent and sacrificing

$$B
$$ ----------------
$$ | . . . . . . . .
$$ | . 4 . . 1 . . .
$$ | . . O O X . X .
$$ | . O X X O O . .
$$ | . b a 3 X . 5 .
$$ | . . 2 . . . . .
$$ | . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ ----------------
$$ | . . . . . . . .
$$ | . 4 . . 1 . . .
$$ | . . O O X . X .
$$ | . O X X O O . .
$$ | . b a 3 X . 5 .
$$ | . . 2 . . . . .
$$ | . . . . . . . .[/go]

It should be fairly obvious that black got more points and white fewer in the descent variation (a plus for black). There is also a not insignificant thickness difference on the outside in that white a is sente to pull out the 2 stones so it's harder for black to use the aji of the cut at b (a minus for black). But balancing those two factors I'd judge the second diagram is an improvement for black over the first.

And giving white the choice to start a fight is certainly a plus for white as more choices are good. This way of thinking about a move: what are the pros, what are the cons versus some other move that you may be more familiar with is one I find very useful when I play and I hoped it would be helpful to others too.

[RobertJasiek]

Ok, now your comparison of the two variations is clear, thank you. It does not provide an absolute judgement though.

[emerus]

Ok, now your comparison of the two variations is clear, thank you. It does not provide an absolute judgement though.

My earlier post is just about this point - your analysis alone doesn't provide an absolute judgement anymore than uberdude's. It's too strong of a statement.

[RobertJasiek]

Regardless of what one calls "absolute", stone difference, territory count and a useful value for influence are essential. Every analysis should be studied in such a context. In particular, it is insufficient to know that variation A is better than variation B as long as one does not know how good or bad variation A is, i.e., whether it creates an equal result or how far it is from an equal result WRT to the mentioned or other aspects.

[cyclops]

I hope the "Beginners" are still with us

[joellercoaster]

Yup. Still pretty interesting, if not immediately applicable by us yet

[Kirby]

whether it creates an equal result or how far it is from an equal result WRT to the mentioned or other aspects.

Bolded part of quote above is an important point. Strong players may have different methods for determining what constitutes an equal result, and these methods are not always aligned.

Sometimes, even pros disagree on what constitutes an equal result.

In the end, I think you have to decide for yourself a method you believe in for evaluating a position. If it leads to winning more games, great. If not, maybe you can consider a different method.

[RobertJasiek]

Although equality is the ideal, a joseki result need not be exactly equal, but it suffices that Black's and White's aspects approach equality.

The value of the combination of Black's territory, influence and other aspects must be similar to the value of the combination of White's territory, influence and other aspects modified by the compensation for the stone difference and viewed in the global positional context.

Therefore, it is not necessary for different players to agree on exact equality, but it suffices for one player to determine for himself an approximation of equality meaningfully. Since stone difference, territory and influence always are relevant aspects, a player must include these aspects in his consideration, if he wants to determine if an approximation of equality is given. He must consider also other aspects if their impact is significant, but he can ignore insignificant aspects, whose impact is too small to alter the distinction between "approximately equal" and "not approximately equal".

[badukJr]

but it suffices for one player to determine for himself an approximation of equality meaningfully.

I don't think that's entirely correct, because any particular player considering some sequence of moves as a near equal result doesn't make that sequence a joseki. Joseki transcend individual games. You are talking about something else, that doesn't have its own term (yet).

[RobertJasiek]

I encourage everybody to surpass the age of letting joseki determination depend on waiting for general consensus, but develop the skill to recognise josekis by himself and also during a player's own games. During a game, you cannot ask others and wait for their input, but you need to rely on your already existing knowledge and determine josekis / approximate equality on your own.