Life In 19x19

Black 127 is a reverse sente worth about 7.5 points per excess move. White J18 - tenuki - J15 - H15 - K16 - H16 - K19 - L18 - M18 etc. explains why White J18 would be sente and Black 127 is a reverse sente rather than a gote. As long as Black is not guaranteed to get both F18 and F16, the extra intersection E18 contributes only 0.5 to the value of the move. White J18 should be answered by M18, I think. Therefore, Black 127 makes the new territory intersections M18, K19 and J19 and prevents White's new territory intersections H18 (two points), G19, F19 and (as mentioned) possibly E18.

White 136 has Black's privilege follow-up E1 etc., so that the white move makes 14 points in gote = 14/2 = 7 points per excess move. (Let us assume for the sake of simplicity that C1 already counts as Black's territory.) So J3 has a - slightly - smaller move value than Black 127. (If one goes into more details, namely White's follow-ups after a Black J3, one sees that actually the value of J3 is even slightly less 7 points per excess move, because Black J3 does not prevent all of the 14 new white territory points there cleanly.)

Presumably White knew that White would not have enough points to win if White 122 or 124 were already played at J3. White tried to catch up before letting J3 be as powerful as possible in the endgame position. Unfortunately, the catching up tactics did not suffice and eventually White had to take J3 nevertheless when probably knowing of still being behind with it.

EDIT: move sequence

If white gets J18, then later K19 is his privilege:

Click Here To Show Diagram Code

[go]$$Wc19
$$ +----------------------------
$$ | . . . . . C C C . 3 7 5 6 .
$$ | . . . . . C O B 1 X . 4 8 .
$$ | . X O O . . O O O X . . . .
$$ | . . X , O . . . X M . X . .
$$ | . X . . . . . . . . . . . .[/go]

Note that falls back because of the aji around the marked spot.

If black gets J18, then later H19 is his privilege:

Click Here To Show Diagram Code

[go]$$Bc19
$$ +----------------------------
$$ | . . . . . 4 3 5 S S S S S .
$$ | . . . . . 6 O X 1 X S S S .
$$ | . Z O O . . O O O X . . . .
$$ | . . X , O . . . X , . X . .
$$ | . X . . . . . . . . . . . .[/go]

This is a swing of some 14 points.

Robert mentions the following sequence after

Click Here To Show Diagram Code

[go]$$Wc19
$$ +----------------------------
$$ | . . . . . . . . . 5 . . . .
$$ | . . . . . . O X W X 6 7 . .
$$ | . X O O . . O O O X . . . .
$$ | . . X , O . . 4 X 3 . X . .
$$ | . X . . . . O 2 1 . . . . .
$$ | . . . . . . O X . . . . . .
$$ | . . . . . . . X . . . . . .[/go]

But it seems to me that is a mistake and should be at

Then again, with at L17 white can squeeze like this:

Click Here To Show Diagram Code

[go]$$Wc19
$$ +----------------------------
$$ | . . . . . . . . . . 9 . . .
$$ | . . . . . . O X W X 6 7 . .
$$ | . X O O . . O O O X 5 8 . .
$$ | . . X , O . . 4 X 3 0 X . .
$$ | . X . . . . O 2 1 . . . . .
$$ | . . . . . . O X . . . . . .
$$ | . . . X . O X . . . . . . .
$$ | . . . X . . X . . . . . . .[/go]

Click Here To Show Diagram Code

[go]$$Wc19m11
$$ +----------------------------
$$ | . . . . . . . . . 1 O . . .
$$ | . . . . . . O X O X X O 3 .
$$ | . X O O . . O O O X 2 X . .
$$ | . . X , O . a X X O X X . .
$$ | . X . . . . O X O . . . . .
$$ | . . . . . . O X . . . . . .
$$ | . . . X . O X . . . . . . .
$$ | . . . X . . X . . . . . . .[/go]

Though black may get to do some counter damage in the center later at a

I wonder about o4, is it better at n4? Perhaps it leaves some greater aji inside (I think I once saw Cho U play bizarre q3 attachment in such a shape), but with the way play went against the lower side black group it would have ended up in a more efficient place.

Overall, it feels like white played rather simply, perhaps assuming that as the stronger player he would win and not respecting black's centre potential quite enough. Often when a weaker player goes for the centre you can just let them have it and win anyway, but not so here. I agree 70 was slow and 84 should have done something about the centre.

Sequence 4 (which I imagined during the game) is better than sequence 3, with which I confused myself when trying to read on the small L19 game diagram. I think White prefers sequence 4 to sequence 1, and then the loss in Black's territory becomes unbearable. So I do not think it is just White's gote with privilege follow-up, but rather White's sente. Maybe Bill can clarify this for us? He always knows whether sente is ambiguous:)

During the game, I did not calculate the value on the top carefully enough. Instead, I simplified my thinking by removing the aji on the top, assuming White J3 and making the positional judgement that I would still be ahead in the game by about 2 points.

With N4 instead of O4, one enters constant worries about whether Black S3 works. IOW, N4 is not proper.

RobertJasiek wrote:This is such a nice aspect of European tournament go: there are even games regardless of amateur or professional ranks, and the professionals have the courage to accept occasional losses.

For now, with only two of them, anything else seems impossible. What will happen when there is, say, 12 of them? I can well imagine that there will be a number of closed tournaments.

RobertJasiek wrote:With N4 instead of O4, one enters constant worries about whether Black S3 works. IOW, N4 is not proper.

Click Here To Show Diagram Code

[go]$$B
$$ --------------+
$$ . . . . . . . |
$$ . . . . . . . |
$$ . X . O . . . |
$$ . . . O . . . |
$$ . X . . . . . |
$$ . . . a . . . |
$$ . . . c . . . |
$$ . . b . O . . |
$$ . . . . . . . |
$$ . . . , X d . |
$$ . . . . . . . |[/go]

I had a look for this shape in GoGoD and the results were: a = 6, b = 3, c = 3. In one of the c games black did live inside but that was helped a lot by a black stone at d which appeared during a ko fight. So I don't think the kosumi can be dismissed outright, but it will depend a lot on the rest of the board and who is likely to have sente to start actions here and what effect they have on neighbouring groups. In your game white does the q7 s8 combo immediately afterwards which slightly changes the s3 aji.

OC, positional context matters. However, here White has little use for a faster shape on the lower outside. - If White had combined N4 with Q7, I might have considered a more aggressive answer to retain the S3 aji, such as Q8 at R7.

In similar positions in my earlier games, a few professionals have commented on Q7 etc. being premature and helping Black more than White. Comments were like "Q8 punishes White for his unreasonable, premature invasion". I agree. I.e., the black thickness is worth more than White's added 6 points of territory.

Is q7 itself premature, or just the s8 continuation? q7 I see as a probe, getting black to commit to a choice early and although it could be aji keshi I often see it in pro games (thinking of the shape of the three black stones generally rather also against that white corner shape). s8 does seem too early as you managed to use the resulting group as thickness that contributed to your moyo. I have seen games where that group is attacked later as s8 takes eyespace as well as territory. So I think white should treat the q8 answer to the probe as creating a large gote move that for later but not take it until it is clear black can't efficiently use the thickness. There is a slight danger q7 becomes mochikomi if black later make the other one point jump from his group or otherwise swallows it.