4-5 point 4-3 approach ogeima, contact

Contact play  

After the 4-5 point 4-3 approach ogeima, the contact play at may look like an obvious choice but it is not considered joseki ^[1]. White may expect a-b-c for a peaceful result but Black can cut at b instead, leading to a fighting variation which is in his favor.

This play occurred in game 4 of the famous sanjubango between Shusaku and Ota Yuzo, where Shusaku as Black won by resignation in 167 moves, despite being "tricked" by Yuzo's ogeima^[2].

Table of contents Variation a Variation b Push in

Table of diagrams Contact play Contact play LZ suggestion Variation a Variation a, 11–20 Variation a, 21–30 Variation b Variation b, 11–13 Push in

Contact play  

If the diagonally opposite corner has a ladder breaker may be played at directly and White can get a good result.

Otherwise, the moves here are almost forced. Now White may choose between a and b but neither gives a good result.

kmr - even if lower right corner has ladder breaker, LZ still likes to play at .After white 9 black has 3 possibilities a,b or c.

LZ suggestion  

Variation a

Variation a, 21–30  

( @ a)

White remains with only a few points of territory, against Black's influence.

Variation b

Variation b  

also leads to a poor result ...

Variation b, 11–13  

... White lives on a small scale.

Push in  

bugcat: It's worthwhile to note that the push in (2) seems to only be a pressured move, used when White needs to force for life on the top. I think that is because, in contrast to the keima position, (5) can more easily be at a rather than the usual b, since the easternmost stone is further away and thus less relevant to the various possible semeai.

Compared to when the keima is on the board, White might also have more follow up aji after the settlement following (5) c, which is a hypermodern move often suggested by AI in the keima joseki.

Also, even if the joseki from (5) b is played out, the eastern stone would still be better placed in the keima position, making a tiger's mouth in the settled shape.

kmr - in this case LZ suggest extension at "d" giving white around 71% of WR (empty board,chn rules, 7,5komi). "c" and "e" LZ are also in LZ good paths, giving around 70% WR. LZ doesnt like much "a" (63% WR) or "b" (59%WR).