I played a bit better, but still lost.

[tekesta]

Hello again. I bring you yet another handicap game in which I play as White. I played a bit more cautiously this time, but looks like I did not do enough globally. (Perhaps I should've invaded one of Black's corners.) I thought I won the game, but ended up losing by 11.5 pts., despite killing a large group in the lower right side of the board. Much obliged in advance for your kind comments.

[billywoods]

A few comments.



By the way, skip to move 70 or so and estimate the score, assuming black's centre group lives.

[Dusk Eagle]

One comment on an issue that's small, but will show up in every single game.

Beginners like to play moves like :w207:, descending down to the first line. They do this, I believe, because if they just connect at N18, they see that black can play N19, forcing them to block at M19, and their territory is one point smaller. What is missed in this analysis is that it is gote for black to play N19. The correct move for white is to connect at M18 instead of descending. The reason is that this leaves white the sente followup of O19, which costs black two more points than the descent to N19. So, by connecting at N18, black either has to play a very small one-point gote move when there are almost always bigger moves on the board, or black will play elsewhere and white will get the opportunity to play the O19 sente move. So in conclusion, it is better to connect solidly at M18 than to connect at N19.

Like I said earlier, this shape comes up in virtually every single endgame, so it's a good thing to learn and remember.

[Bill Spight]

I decided to try to find the game losing move. Maybe there was none, since you gave a handicap. However, killing a large group is usually enough to win a three stone game, and indeed, when you do that the game looks close enough for White to go on and win.

But fairly soon Black has regained the advantage of the handicaps. Starting at move 167 White devastates the bottom left corner, but that is not enough to win.

One candidate for the losing move is White 135 at N-06, which kills a dead group. Obviously that move gains nothing. White has an extension on the left side to C-14 that gains at least 11 pts. It bolsters the White group there and threatens to slide into the corner. An extension to C-15 may also be playable. But Black has a big reply to either extension, so that is probably not enough to win, even assuming that Black plays as badly in the bottom left corner.

Another candidate is White 111 at L-06. that also kills a dead group. It allows Black 112 to play a turn at J-11. That move is bigger than it looks. It bolsters Black's only weak group, the four stones in a row in the center. In fact, White goes on to allow Black to build a central wall, turning Black's liability into an asset. Even though the wall has cutting points, by move 122 White's chances of winning are slim. White 111 should attack Black's weak group. The capping play at G-10 comes to mind.

Another candidate is White 103, the atari at M-11. White can capture the Black stone in a net at L-11. Black ran with the M-10 stone while White pushed from behind. It was that stone that became the central Black wall. Capturing that stone securely would have retained winning chances for White.

[EdLee]

Beginners like to play moves like :w207:, descending down to the first line.
They do this, I believe, because if they just connect at N18, they see that black can...

They don't think that far; that's not what they're thinking.
(By they, I mean we, most of us as beginners.)
Yes, many beginners play the N19 descend. Another popular move is M19 tiger's mouth.

That a beginner replies at all to :black:206 means they have learned that otherwise B can capture the N18 W stone.
We play N19 descend or M19 tiger's mouth probably because we saw others do it -- that's all.

Most of us as beginners simply don't consider connecting directly at M18, let alone what happens afterwards.

In this particular shape, there are five possible local moves to save the M18 W stone.
The two most popular beginner replies are N19 descend and M19 tiger's mouth -- I don't know the percentages.

[tekesta]

Well, thanks for the kind commentary. What Dusk Eagle mentioned is true. Had I played solid connection at M-18, I would not have had to waste another move playing there anyhow after the descent to N-19. Also, I should've expanded my White groups on the left side; this would've given me better chances at victory. Finally, I should've fortified the top W group and used it to attack the Black groups White 11 @ R-17 would've probably been a good way to begin attacking the Black shimari in the top right. Although then Black would've just sealed White in and, while letting me have the corner, have good outward influence.

Well, I'll remember - or try to remember - this information in my next handicap game when I play as White.

[Bill Spight]

Normally the descent, W207, is wrong. But the position on the top side is special, and in this case it could be the right local play.

To show that I have set up two difference games, where the playing area is on the top side or bottom side, but not in between. As we can see, neither the solid connection nor the hanging connection dominates the descent, because the first player can do better in the difference game.



I did not include a difference game between the hanging connection and the solid connection, because it appears that the ko matters, and in that case the difference game comparison is invalid.