Hi,

I decided today that I want to study the situation when the 4-4 point is approached from both sides.
I started with the double low approach (two keimas).

May starting point was Sensei's Library:

http://senseis.xmp.net/?44PointDoubleLowApproach

The most basic continuation deriving from this situation seems to be the extend and attach joseki played out like this:



The explanations at Sensei's leave some questions unanswered for me.

1) White 14 seems a little bit cramped to me. I would have expected it to be placed one step further at K17.

2) Black 9 seems to threaten a cut at c13, but can't white play 10 at 11 in order to go deeper into the corner? Or at least: What kind of reinforcements would white need nearby so that it could handle the cut and therefore ignore the threat.

3) What happens if white enters the corner at B17 and black had not played E18?

4) What happens if white enters the corner and black has already played E18?

5) Can someone point out a couple of games in which this variations was applied (if possible with a link to sgf files ?

Thanks for any help!

1) If white plays K17, black can cut at G17. The G14 stone allows black to capture G16. Whilst white can struggle and not just die horribly, a much better result is to just play J17 and protect the cutting point.

2) I think the problem is, if white plays B16, blck can immediately play C13 and extend to C12. Now white has to live in the corner at some point. When white does so, it will probably induce black D18, allowing black to get G17 (or potentially something harsher) and settle his group. So white would get a small amount of points, but black would get away too easily.

3) It will depend very much on what the players are aiming at, and what is nearby. For instance, black can answer B17 with A15, activating the cut at C13. Separately, this is quite an active position, so I think usually the players will immediately start testing one anothers weak positions and the game could go in many directions.

4) If black has already protected at E18, white will get cut off and die in the corner, I suppose. The same possibilities as in (3) also exist, of course.

5) Milanmilan just today on kgs:

Thanks to both of you!

A few more comments about white's threat in the corner. If black doesn't defend at e18, white can steal the corner territory, but as black separates by taking a liberty from the 3 stones with a15, he can cut afterwards and, unless white has support on the side allowing him to play at , he will sacrifice them. This will generally be good for white (notice black doesn't have 2 eyes yet).

Click Here To Show Diagram Code

[go]$$Wc Taking corner
$$ --------------------
$$ | . . . . . . . . . .
$$ | . . . 5 . . . . . .
$$ | . 1 3 X O O . . O .
$$ | . X 4 X . X O . . ,
$$ | 2 O X 8 . X . . . .
$$ | . O O X 7 . . . . .
$$ | . 0 6 O . O . . . .
$$ | . . 9 . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . , . . . . . .
$$[/go]

If we move j17 to k17 then this sequence doesn't work for white, due to black's clever cut at g17, making a and b miai.

Click Here To Show Diagram Code

[go]$$Bc k17 doesn't threaten to take corner
$$ --------------------
$$ | . . . . . . . . . .
$$ | . . a O 5 b 2 4 . .
$$ | . O O X O O 1 3 . W
$$ | . X X X . X O . . ,
$$ | X O X . . X . . . .
$$ | . O O X . . . . . .
$$ | . . . O . O . . . .
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . , . . . . . .
$$[/go]

Another important point in the 1st sequence is the timing of the atari at , it mustn't be played earlier or white destroys his threat to take the corner. Black descends to a16 now for liberties and then has a nice belly-attachment tesuji at to keep the corner. If white c16 was atari he could cut there in sente, now he is short of liberties and dead.

Click Here To Show Diagram Code

[go]$$Wc Not taking corner
$$ --------------------
$$ | . 8 . . . . . . . .
$$ | 7 4 5 6 . . . . . .
$$ | . 1 3 X O O . . O .
$$ | 2 X . X . X O . . ,
$$ | . O X B . X . . . .
$$ | . O O X W . . . . .
$$ | . . . O . O . . . .
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . , . . . . . .
$$[/go]

Another point here is if white already has k17 in place with the double approach, he can play considerably more severely (black could lean on the other stone, but it would be more normal to lean on the stronger stone).

Click Here To Show Diagram Code

[go]$$Bc k17 already in place
$$ --------------------
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . 5 4 O . . . O
$$ | . . . X . 1 2 . . ,
$$ | . . 7 a . 3 . . . .
$$ | . . O . . . . . . .
$$ | . . . 6 . . . . . .
$$ | . . . . . . . . . .
$$ | . . . . . . . . . .
$$ | . . . , . . . . . .
$$[/go]

If black plays , thinking he is playing joseki, white has a nasty surprise for him: the wedge in at a which makes miai of cutting the centre stones or capturing which was in fact an overplay and should have been bamboo at a. The reason k17 makes this possible is black's cut at g17 is much less effective with k17 already in place as white can atari underneath to avoid giving black e16 in sente. Without k17 white would be miserably crawling on the 2nd line, so would probably have to capture g17 with atari at h17, but this gives black e16 in sente so then he can come back to defend c15. Assuming black doesn't want to get cut at the following could happen:

Click Here To Show Diagram Code

[go]$$Wcm8 continued...
$$ --------------------
$$ | . . . . . . . . . . .
$$ | . . . . . . 3 5 . . .
$$ | . . . X O O 2 4 . O .
$$ | . 9 7 X 6 X O . . , .
$$ | . 8 X 1 . X . . . . .
$$ | . . O . . . . . . . .
$$ | . . . O . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . , . . . . . . .
$$[/go]

White captures 2 stones on the side and has sente, and black has 1 eye and a few points in the corner and captures g17 for an eye and thickness in the centre.

Click Here To Show Diagram Code

[go]$$Bcm17 ...continued
$$ --------------------
$$ | . . . . . . . . . . .
$$ | . . . . . . O O 8 . .
$$ | . 3 5 X O O X X 7 O .
$$ | . O O X X X O . . , .
$$ | 6 X X O 1 X . 9 . . .
$$ | . 4 O 2 . . . . . . .
$$ | . . . O . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . , . . . . . . .
$$[/go]

This double approach joseki was actually the focus of a recent game of mine, and I did have k17 already in place (well, position rotated) and black didn't make the safe bamboo. For some reason I chickened out of punishing him with the wedge. As you can see in what happened I played the corner attachment, but his clever cut at c13 meant I could not take the corner and hang on to the outside stone so I got a bad result. Instead of q10 I could have defended my group at l17, but as I had allowed him to take r17 in exchange for a double approach on his corner but he ended up getting a good result I felt I could not afford to reinforce, so white attacked at k17. You can see that if I had followed the variation above I would have captured 2 stones and although not 100% alive would have been much stronger. I couldn't let him play k17 in sente, so made some sente exchanges to reinforce my group, and then tenukid again. The focus of the entire rest of the game was how much damage white could do to my group at the top. Instead of n13 I could live with j19, but thought I would lose that way, so tried to kill his stones in the centre as my group was still a ko. Unfortunately my opponent quit OGS and resigned before the situation was resolved: it was a very exciting game and dominated by the implications of this joseki and both of our mistakes in it .

Wow, that is A LOT of input

I guess I will work through that betwenn christmas and new year's eve

Thanks!