:: Igo Hatsuyôron 120 (2022)

New Story Line 2022

Kikashi in the Hanezeki

Our presentation will follow the chronological development of the findings on the kikashi in the hanezeki.

Fujisawa Hideyuki 9p did not play these kikashi before White's decision point, which turned out to be a mistake for the Semeai Variation a few years later.

We originally played this kikashi immediately after the Tsuke of Yamada Shinji 6p, but then came to the conclusion that it would be better to insert these just before the completion of the Crosscut Sequence.

Finally, KataGo favours this kikashi very early in the game, following a typical characteristic of AI programs.

Referenced by ( 2022MainLine), ( 2026)


Let's go back to the dawn of time ...

: ( 2022MainLine)
After having exchanged for in the (temporary) seki, Fujisawa Hideyuki 9p continues with giving atari at , so starting the creation of the hanezeki in the lower right corner.

He will play Black's kikashi , and , much later, that is after White's decision point in the Main Semeai.

Almost four decades later in the world of KataGo ...

:
KataGo likes to play the kikashi in the hanezeki early, following a typical characteristic of AI programmes.

Please note that the first (upper) kikashi at is not mandatory after my Second Throw-in (, ), as a White move at would be gote.
However, having played this kikashi early on prevents some potential later entanglements of the KataGo era from the very beginning.

:
The second (lower) kikashi, the atari at , is not necessary at all so early in the game. It can be played at any time later.

Let's return to the era of Fujisawa Hideyuki, arriving much later in the game, just before an alternative semeai comes to an end ...

Fujisawa Hideyuki gives only the first four moves of his Semeai Variation.

Our assumption is an identical sequence in the hanezeki as in the other semeai between Black's large group in the upper right and White's bottom group, which he explains in detail.
The final stage of this semeai is displayed here.

:
Black plays the final oki in the lower right corner.

:
White continues with occupying outside liberties of Black's large group in the upper right.

White's corner group has only two liberties left, so her earlier kikashi / no longer contain any threat so late in the game, as Black would simply continue with taking her corner group off the board.
Adding stones at and / or after White's choice for the Semeai Variation would only lose points unnecessarily.

:
Black continues with occupying liberties of White's remaining group in the earlier temporary seki.

:
Black wins this semeai by one move.
If White captures at , he will also capture with .

The final result of the application of the sequence shown above would fit with Fujisawa Hideyuki's commentary on his Semeai Variation, which reads "Black wins by a large margin." (i. e. Black + 5 under the assumption of komi = 1).

Back to the main line again, much later in the game ...

This is the position in Fujisawa Hideyuki's line of play after White's choice of the Semeai Variation of the Main Semeai.

This sequence is not continuously enforced - in the sense that both sides would have several opportunities to play tenuki - so let's forget the left half of the board for the moment.

:
Fujisawa Hideyuki starts the endgame at the left side of the board with giving atari at Black's single stone at the upper edge.

Please note that he would have really done so as early as with in the diagram above, as all of White's moves are gote therein.

White will capture at , only after Black occupied another two outside liberties of her upper hanezeki group.
This implies that Black needed at least three surplus Main Semeai Liberties (so a total of ten) to bring his large group to life in the Semeai Variation as well.

Let's take a short detour to China, exploiting a weakness in Fujisawa Hideyuki's sequence ...

:
Without Black's kikashi , played, White will play two kikashi of her own in the hanezeki at the earliest possible moment, if she wants to capture Black's large group in the upper right. This is immediately after Black's atari at White's single stone at the outside of the forthcoming hanezeki.

Both and take a liberty of Black's group below, in sente, threatening to capture it.

Cheng explained this special feature of the hanezeki in his 1988 book.

:
Black is forced to reduce White's liberties, so keeping White from giving atari from above at this group.

:
Please note that this second kikashi is not mandatory - contrary to the first one - and could be played at any time later.

:
Both of Black's stones will become additional prisoners in the Semeai Variation.

Semeai Variation:

White gained five points in the hanezeki area (at right), compared to Fujisawa Hideyuki's assumed sequence (at left):

- White captured an additional two Black stones, so gaining four points.
- White got an additional point of territory.

The game will end in a draw now.

VAR

:
This move starts the correct classical solution (without my Second Throw-in).

In his 1988 book on Igo Hatsuyōron, Cheng found this early kikashi being an improvement for Black, providing him with an advantage of two points in the Semeai Variation (which at that time was better for White than the Capture Variation), see above.

We have already met this kikashi in the previous section.

:
White's connection is forced.

Sometime later before White's decision point ...

Even after Black's initial kikashi in the hanezeki, is still White's priviledge, if she wants to capture Black's large group above.

Again, Black is forced to occupy a White liberty with his move at , which will end up as an additional stone in White's prisoner lid.

Please note that this exchange will benefit Black in the Capture Variation, so White might utilise it as the initial moves of the Semeai Variation.

Semeai Variation:

In the hanezeki area, White gets three points of territory now (at right), which is two points less compared to Cheng's proposal of an optimal White sequence for the continuation of Fujisawa Hideyuki's aborted Semeai Variation (at left).

Black will win the game by two points now.

This is the same score as in Fujisawa Hideyuki's (assumed) Capture Variation.

Let's enter the world of amateurs now ...

:
Originally, we had played these kikashi in the hanezeki much later, immediately after Yamada Shinji's Tsuke with .

: (C 2027)
But Harry found that White would then be able to benefit from ignoring these with one point, so we had to adjust the timing of these moves.

Please note that KataGo's Immediate Crosscut after Yamada Shinji's Tsuke was not yet born at that time.
After Black's block with , White draw back with , instead of playing the cut of .

VAR

: ( 2027)
Harry's suggestion:
White ignores Black's atari in the hanezeki, going for the kill of Black's upper right corner, instead.

:
Black has to capture White's stones in atari, providing independent life for his large group in the upper right.

:
White kills Black's upper left corner, in return.

For further details, please refer to the history of our investigations ...
1092a

In the end, Black will have lost one point overall.

2027

Since then ...

... we played the kikashi direct before the start of the Crosscut Sequence in the upper left centre.

Let's return to KataGo's world again ...

, :
KataGo thinks that the kikashi in the hanezeki can be delayed at most until after this exchange in the Nakade Sequence, just before Black's nakade is completed.

; :
An "extremely late" version of Black's kikashi in the hanezeki (more precisely: in the temporary seki).

:
Now, Black completes the nine-stone nakade on the left edge.

:
White continues the solution sequence with capturing Black's single stone in the lower left.

2022MainLine

Copyright © 2022 Thomas Redecker.

Design by Jan van Rongen, modified by Thomas Redecker.

L e g a l   n o t i c e