Life In 19x19

I've never noticed the use of an avalanche joseki with which the bots agree. Is there a game? I like the avalanche josekis, so I kind of hope so.

But I also got the impressions that humans play them less than before. Are they out of fashion, simply suboptimal or am I looking at the wrong games? I don't have a strong bot at hand so that I could check for myself.

Following the influence of AI, the high approach to 3-4 is much rarer now that a few years ago, plus when people high approach and attach under the outside hane is normally the best response, but you do see some avalanche occasionally but rarely the long complicated lines, e.g. Iyama did the solid connect vs Shibano in the Oza a few days ago: viewtopic.php?f=13&t=13751&p=250771#p250771

One other speculative thought to add to Uberdude's observation about the high approach: as I understand the complicated lines, the strategic** reason why the 3-4 player (let's just assume it's black) may prefer them rather than accepting a large solid corner is that, presumably, white chose the avalanche because it's building a wall in the right direction. So, black may not want to passively allow white to get a wall in that particular direction. But--this would need a lot of qualification and nuance, but generally speaking--I think post-AI we're all a little less certain that a wall facing one way as opposed to another is that important. In other words, there may be less avalanche in part because "direction of play" is a fuzzier/less useful concept than we once thought.

**"Strategic" in the strictly board-based sense. There's also the "strategic" psychological aspect--by choosing the complicated large avalanche variations you're saying you're not afraid of and maybe even relish a tough fight, you ain't no pushover, etc. Those messages can be strategic in their own right though in my experience the messaging is less strategic and more instinctive arrogance.

bayu wrote:I've never noticed the use of an avalanche joseki with which the bots agree. Is there a game? I like the avalanche josekis, so I kind of hope so.

Sure. I took a quick look and the first avalanche I found was a Takagawa-Sakata game, GoGoD 1950-12-20b. In the Elf commentary Elf prefers the outside hane, but only by 2% over the avalanche, well within the margin of error. OC, we can't expect every joseki variation to meet the bots' approval.

In that game, and in some others I have seen, however, White usually prefers to descend to the second line after Black blocks on the third line, instead of playing the 4th line hane (small avalanche) or the third line nobi (large avalanche). Sakata played the nobi, and Takagawa played the 4th line push for the large avalanche. However, Elf does not like the push, preferring the 2d line atari by 5%.

That being the case I doubt if we can consider the large avalanche to be joseki anymore. However, there may well be positions where it is fine.

Here is a game where the small avalanche is joseki.

bayu wrote:I've never noticed the use of an avalanche joseki with which the bots agree. Is there a game? I like the avalanche josekis, so I kind of hope so.
But I also got the impressions that humans play them less than before.

It doesn't really matter whether "it is joseki or not", probably if you like it you should play it, and try to lead the game into situations where it's more likely for your opponent to play it, but it is still up to your opponent to decide.

As for me, I stopped playing it even before AlphaGo made it come out of fashion with pros, just because it is complicated and I hate complex joseki.
I don't think there is any situation where I would feel that avoiding it will make me fall behind enough to care: if I am "on the inside", I just don't play the hane that starts it (either the small or large avalanche).

It would be interesting to see if any position would make me change my mind

sorin wrote:I don't think there is any situation where I would feel that avoiding it will make me fall behind enough to care: if I am "on the inside", I just don't play the hane that starts it (either the small or large avalanche).

It would be interesting to see if any position would make me change my mind

You mean the fourth line hane?

It may not give enough of an advantage to change your mind, but check out the Go Seigen - Maeda game above. At move 28, though, don't take the corner, but play the variation that crawls on the second line. This variation was new to me. I kind of like it.

Bill Spight wrote:

sorin wrote:I don't think there is any situation where I would feel that avoiding it will make me fall behind enough to care: if I am "on the inside", I just don't play the hane that starts it (either the small or large avalanche).

It would be interesting to see if any position would make me change my mind

You mean the fourth line hane?

It may not give enough of an advantage to change your mind, but check out the Go Seigen - Maeda game above. At move 28, though, don't take the corner, but play the variation that crawls on the second line. This variation was new to me. I kind of like it.

Oh, I mean I would avoid it much earlier, by move 28 it is already way too complicated for me
In this game, I would not play hane with white at 14 (to avoid small avalanche), but connect in 3x3 or crawl on the 3rd line; If I chose to crawl and black would press on 4th line, again I would not hane (to avoid large avalanche too).

nuff said
(Go Seigen and Alpha Go agreed on this move, the matter is closed )

Click Here To Show Diagram Code

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

Gomoto wrote:nuff said
(Go Seigen and Alpha Go agreed on this move, the matter is closed )

$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . a X X . . . |
$$ | . . . , . . . . . , . . . . O W X B . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------

Click Here To Show Diagram Code

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

(Diagram edit mine.)

Well, this is joseki, OC. But if we allow instead of a, Then it is likely that the small avalanche hane is also joseki. The large avalanche depends upon White pushing on the 4th line, which may not be joseki, going forword.

Bill, while I normally can follow you quite well, I do not get this argument.

This move denies the avalanche, or is there an oversight of mine?

(I think of white as the player who wants to start the avalanche.)

Gomoto wrote:Bill, while I normally can follow you quite well, I do not get this argument.

This move denies the avalanches, or is there an oversight of mine?

My point has to do with the margin of error, something that we are still a bit uncertain about. is normally considered to be inferior to the outside hane, but by only a small percentage, almost certainly within the margin of error. So While a bot may make the hane, because why not make its best estimate, is almost surely playable in most positions, and may be considered joseki. Likewise, even though the bots generally prefer the descent to the small avalanche hane, the percentage difference is normally small, and we may consider the small avalanche to be joseki, as well. The large avalanche, OTOH, depends upon accepting all of three moves as playable: Black's third line nobi, White's fourth line push, and then Black's fourth line hane. That's asking much more, and so I think that the large avalanche will no longer be considered joseki. The most doubtful play of the three is the White push, I think.

Thank you, for explaining the argument.

Thanks Gomoto for reminding me of one more way to stop the madness early

Gomoto wrote:nuff said
(Go Seigen and Alpha Go agreed on this move, the matter is closed )

$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . X X . . . |
$$ | . . . , . . . . . , . . . . O O X B . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------

Click Here To Show Diagram Code

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

You are welcome