Thick plays in the endgame

    Keywords: EndGame, Strategy

These occur in the early endgame[1].

Example given by Bill Spight, moved here from WhatWasTheHighestMoveNumber (now thickness versus endgame).


In Ishi no Kazoekata (How to Count), Sakata says that simply making plays based on counting territory is not enough. You also have to play thick yose.

Thick yose are plays that aim at aji. Even though they don't look big, later on they make a profit. For instance, if a ko breaks out later, thick yose generate ko threats.
Sakata gives the following example (13th Honinbo league, 2nd April 1958, GoGoD 1958-04-02a, position after move 128).

Iago: this game can be found in gobase at [ext] http://gobase.org/games/japan/titles/honinbo/13/game-l04.sgf

[Diagram]
Shobu Te  

Black was Fujisawa Hideyuki, White was Sakata Eio. Black seems to be lagging (KataGo agrees: white is about 2 points ahead), and makes a shobute with B1.

Having made his kikashi on the top, Fujisawa switches to the large yose in the bottom right.

[Diagram]
Thick yose  

This W1 is a thick yose. Sakata preferred it to W2, protecting his territory in the bottom right. W1 eliminates some black aji in the top and makes W3 sente. In return, White allows Black to penetrate deeply into his territory in the bottom right.

[Diagram]
Black to play  

Jared: I would play W1 here without hesitation, and this page hasn't really convinced me otherwise. I guess I just don't see what is Sakata afraid of?

Dave: Unfortunately Sakata does not give any examples of play on the upper side if he answers in the lower right. However, note that Black can choose something like B2 and B4 to expand the center. As far as the right side is concerned, he clearly states that ending in gote in the lower right is unsatisfactory because there is the large question in the upper right about whether Black or White gets to play at a.

xela: KataGo agrees with Dave's line (as one of a few equally good options). It's nothing fear-inducing (especially for us amateurs who can't count so accurately), it's "just" the difference between white being a couple of points ahead (in the "thick yose" diagram) versus allowing a half-pointer (this variation). So a big deal for players of Sakata's and Fujisawa's calibre :-)

[Diagram]
Game continuation  

Dave: Sakata used his sente to play W1 and W3. Now the marked stone comes into its own. A White push at a cuts Black's corner off from the center while threatening to pull out the two captured White stones. Black answers in the upper right but that leaves White with sente again so he was able to play W7. This does not capture any stones because of the weakness of White's lower left corner (how to save Black's stones is left as an exercise for the reader :-) but it reduces the points that Black makes here.


[1] Bill: Well, this example is from the middle endgame. ;-) Obviously, such plays cannot occur in the late endgame, after the board has divided up into small, independent regions.


Thick plays in the endgame last edited by xela on January 10, 2024 - 00:27
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library