In a recent game on the club, we played the avalanche variation in the attached sgf. The colours were reversed and there were more stones on the board, but it's about the same. I played the upper left variation in case of any ladders (it was like that in the game). Anyway, after black 25, the following sequence in the sgf seems good for black. I'd like to hear some other people's opinions about this variation, and maybe other options for white.



edit: When we just played it, we thought it was good for black, but seeing it again, it doesn't look like that, since black still has a cut. Let me know what you guys think!

[admin]SGF tags changed so that it displays properly. -JB [/admin]

You're much stronger than me, so I won't try to offer positional judgement, but in a slightly off-topic note, it seems like pros these days tend to prefer the atari at s4 (from what I've seen). Here's an example from Iyama Yuta vs. Sakai Hideyuki just recently:



It's nice for us amateurs b/c it's much simpler than the fights of the usual large avalanche variations.

I think this variation looks quite OK for white. White has points and sente. Black has influence, but the cutting point is annoying. And since this is originally black's corner, White cannot expect too much anyway.

This avoids the complications of the large avalanche, but has its own. For example, an often played opening is white q14 @ r10, and black r13. A really hard to understand sequence follows.

Anyway, I think both are popular among professionals.


@Herman: I think you're right. At the club Koen and me both thought it was good for black, but I don't see how we could even think that right now. By the way, I won that game

Some more thoughts:

White could play N5 before playing P2, that looks like it increases the cutting aji.

White can think about playing O5 in response to P6, but that looks like a very complicated option

I think that might be a bit better indeed.



I don't like this, since white's ladder doesn't work and black is still on the shape point. I can imagine no good follow up. Even if it's complicated, it seems better for black.

I wonder whether worrying about the cut is an over-refinement. It's hard to imagine a scenario where the cut is powerful. There are only two (rather old) GoGoD databases games featuring this shape. In both cases the cut was not a factor at all. The outside players got large territories around their thickness (and FWIW won the games), whereas the inside player's group, while safe, took no further part in the games.

There is also a small question mark over the inside player's play in tewari terms.

John, could you give a diagram with the mentioned tewari?

Life is too short to mess around with L19 diagrams, but take off the three white stones in the middle of the bottom row of five (to cancel out the three black prisoners. The white sagari shape in the right corner then looks like a slightly inferior shape. Not enough to matter to us amateurs, perhaps, but a pro may baulk at it, and maybe that's part of the reason the line is so rare.

John, taking away the three lower stones to compensate for the captures results in an inferior position for white then the one on the board. For example, when black descends on the left site he has sente endgame. So I think you can't do something like that to judge the position.

Well, pros do. It's not a case of "when black descends" but "if black descends". You have to make a judgement about who is more likely to get to that point. It's always gote for Black but it's easy to imagine it being a sente play for White. Tewari is not a rigidly calibrated tool.

I'm really confused here.

I don't think you all realize that you are basically discussing one of the most basic joseki after black turn at S5.

Click Here To Show Diagram Code

[go]$$B
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . 4 0 . . |
$$ . . . . 8 1 2 . . |
$$ . . O a . 3 X O . |
$$ . . O X X X O O . |
$$ . 6 X O O O X . . |
$$ . . 7 9 b 5 X . . |
$$ . . . . . . . . . |
$$ ------------------+[/go]

The only difference is that A is exchanged for B, which is a bit of a minus for black normally, but it hardly makes a difference. As I mentioned before, there are some other variations for white to take, to make it a bit more complicated, but this result seems fairly even.

I found this thread interesting, because it shows that although databases are excellent tool for self-study, it can also be fairly "dangerous" because you might misinterpreted the result.

Probably the result is a bit better for black locally, but something a lot of people forget when they make an analysis is that.

1. It was blacks corner
2. White ends in sente

So considering these two things, white can also be satisfied with the result locally.

Hope this post clear up some confusion

I think the first question should be why does Black normally play at 1 below instead of the marked stone. First there is Herman's suggestion of before . While it is true that the ladder after does not work, the ladder after does! If Black defends, the stones underneath fall to , right?

Click Here To Show Diagram Code

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

However, perhaps more to the point, what does Black do against below? White cuts the two stones apart. Black can no longer atari down and give up the three inner stones, but if , White is happy to be pushed from behind. After Black takes care of the inner stones, White has sente to decide how to handle the outside. Can Black do something better here?

Click Here To Show Diagram Code

[go]$$Wc
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X X . . . . . . . . . . . . . . . |
$$ | . . O X . . X . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . , . . . |
$$ | . O . O . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . B . . . . |
$$ | . . . . . . . . . . . . . 1 . . O X . |
$$ | . . . , . . . . . , . . X O O O X X . |
$$ | . . . . . . . . . . 5 3 O X X X O . . |
$$ | . . . . . . . . . . . 4 2 . . 6 O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

This position is not basic at all. There are only four GoGoD examples, but more to the point it is discussed by Takagawa in "Final Summit" (page 58). There, he indicates that the opening (still new) was being intensively discussed at the time, and the big debate was whether Black (in Fredrik's version of the diagram) should interpose the cut at 10. Takagawa eventually said no, but then changed his mind. The implication of this (supported by the numbers) is that not only is the position is not basic, but it's not even joseki. Since, however, the version with the cut has only 9 games, we can rely only on Takagawa's judgement as to whether the cut version counts as joseki.

Fujisawa Hideyuki, in his Nadare book, avoids labelling it as joseki. He says IF it becomes joseki (this was around 1969) it will be a relatively painless variation for beginners (maybe basic in that sense), but he adds that it will take a very long time for it to become established as a joseki. In other words, even he was unsure at that point whether it gave a satisfactory result, and he seems to be tilting towards no. The virtual disappearance of the line since then seems to confirm his doubts.

Incidentally, the significant move White 4 was attributed to Yasunaga Hajime. I was reading something the other day about Yasunaga (once famous as the world's strongest amateur) being granted an 8-dan diploma. At the award ceremony Go Seigen made a witty speech in which he said that it was still unclear whether this now meant Yasunaga was 8-dan amateur or 8-dan pro. Presumably the diploma, which would normally show by choice of phrases which was meant, was written in a deliberately ambiguous way. And still we hear of people agonising over whether they are 8.456 kyu or 8.457....

I agree that it's not the most common joseki, as there are some slight deviations. However I did not want to add to anymore confusion as I think the result in the game was equal.

But as you bring it up John, I would be happy to show the small differences between the josekis for the ones that are interested:

Click Here To Show Diagram Code

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

This shape is basically the same as the one mentioned previously, (however a bit better for White as he can choose between using a or b as sente.

Additionally there is the option for white to play the atari at 1, making another variation:

Click Here To Show Diagram Code

[go]$$B
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . . 2 . . |
$$ . . . . . O 1 . . |
$$ . . . . b X O O . |
$$ . . O . . X X O . |
$$ . . O X X X O O . |
$$ . O X O O O X . . |
$$ . a X . . X X . . |
$$ . . . . . . . . . |
$$ ------------------+[/go]

Again leaving A or B as sente.

There are some subtle differences between all variations, however there is nothing drastic and although that the game line might be slightly inferior for white because not keeping all options available, it's also hard to say that white not exchanging 1 for 2 is the best.

At our level, and perhaps even at the professional level, it might be extremly hard to analyse the difference between these deviations. Therefore I think the result in the game is equal for both players.

Worth to notice is that after White plays at B, there is no way for Black to break out, which reverts it back to the original diagram except for the 1 2 exchange.

Edit: In addition, I would like to clarify what I meant with "basic" as well. (My first language is not English, so I'm sorry if I'm being misunderstood).

When White plays the double hane 1:

Click Here To Show Diagram Code

[go]$$W
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . 1 . . . |
$$ . . . 2 . X O . . |
$$ . . O . . X X O . |
$$ . . O X X X O O . |
$$ . . X O O O X . . |
$$ . . . . . . X . . |
$$ . . . . . . . . . |
$$ ------------------+[/go]

White has the option to play at 2.

There are several complicated variations after 2. I personally don't know them all. Depending on the order of moves, you can make mistakes afterwards that might leave some step-kos in the corner etc.

(The continuation is:)

Click Here To Show Diagram Code

[go]$$W
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . . . . . |
$$ . . . . . 1 . . . |
$$ . . . 2 . X O . . |
$$ . . O . . X X O . |
$$ . . O X X X O O . |
$$ 6 4 X O O O X c . |
$$ a 5 3 . . b X . . |
$$ . . . . . . . . . |
$$ ------------------+[/go]

After this Black has the option to play at a, b or c.

In that sense, I meant that the result in the game was extremly basic compared to this line. And although there are some small deviations, I think the result is equal for both players.

Hi Fredrik

In your first diagram, I would be extremely happy as White to trade "b" for a forced inside capture by Black, resulting in the basic variation you showed earlier on. The acquired influence to me is superior to Black's territory, even if this was White's corner to start with, a remaining cut in the influence and Black having sente. Being Black I would then cut straight away, for if White turns his influence into thickness with another move, I'd prefer that position even if Black has gotten three moves elsewhere. Well, put that way, I realize this may be a matter of style, since truly 3 moves is a lot. But then I still prefer Black separating White as in your lower diagrams. It boils down to me liking influence, I guess.

BTW, I think you write "white" on a few occasions where you mean "black".

Hi Knowilg, you are correct, I edited my point to reflect your last line. (I thought the color was reversed while I was writing the diagrams, sorry about that).

Indeed, for the eye, it looks safer to exchange "b" for an inside capture but professionals usually do not play such ajikeshi.

The reasoning behind it, is that white is already somewhat strong on both sides, and at the current moment (the opening) it is not that valuable for black to break out. The moment white thinks it would be worthwile for black to break out, that is the correct timing to exchange "b". This kind of sente move is very delicate, and require a lot of strength to choose the appropriate timing.

(In addition, if white deems that it would be small for black to break out, then he would invite black to play such a move thus leaving more room for your opponent to "mess" up. )

After the exchange of "b", I think it would be a big overplay to cut immediately for black. Even if you cut the outside, you have already no possibillity to connect back. Because of that, white can fully utlize his thickness to attack the cutting stone which means that their is little value of cutting. (Depending on outside stones, the cut can be possible in the future but without any supporting stones it can't be good)

Why doesn't black play move 31 at Q4?

_________________
Problems worthy of attack prove their worth by hitting back!