tam1998 v. GoDave89

Published 22 Dec 2020 by antti (last edited 30 Dec 2020)

Made publicly available on December 30, 2020.

Download kifu


Comments (3)


tam1998 wrote 3 years, 3 months ago:

Hi Antti,
thanks for the review, it was very interesting!
Regarding move 111/113, at the time i was actually considering myself ahead (not by much, but anyways^^) and i thought the only way i could lose this is if w makes a considerable amount of territory in the upper center. So i tried to avoid that here. Is that reasonable or did i make the game too close like this? I tried to be 'ruthlessly pragmatic' here (to cite Jeff^^).
Also, you didn't specifically mention it in your commentary, but is it in general reasonable to treat the lower left black group in a bot-ish way (e.g. i just leave it alone a lot, attach and take sente as much as i can and in the end it still lives)?


antti wrote 3 years, 3 months ago:

You’re welcome!

re 111/113: To me the game looks way too close to call it either way, and so I’d restrict my own thinking still to regular play without giving a mind to winning, losing, or ‘securing’ either of them. A quick count for me gives a bit less than 60 points for black, 40 points for white (not including the komi), and then we have the white centre influence which will certainly materialise some amount of territory. To call ‘black ahead’ is to say that you think the centre isn’t worth more than 15 points for white, and I’m sure I’m unable to estimate that.

If, as you say, your main priority was the centre, then 111 to me seems a bit strange; why don’t you then play one line further at n15, forcing white n17, and then lean on white with a move such as l15 to develop into the centre? Of course, this way white will form some territory on the upper side, which also closes some of the 15-point gap.

One more way you could maybe have tried to get the ‘best of both worlds’ could have possibly been to start with m17. Imagine then white m16, black l17, and white n18, and you can then play somewhere around n13.

Regarding the lower-left corner: I think I mentioned at least one good tenuki timing at 59 in the commentary. After that, I would in general be a bit afraid to leave the group alone; probably, if for example instead of 71 black played elsewhere, the group would not die, but then white c7 will become a severe attacking move. To efficiently play elsewhere, black at least needs to make sure that white has no obvious good way of attacking black, and I saw fairly few opportunities like that in this game.


Sanderbaduk wrote 3 years, 2 months ago:

This game and the move at A (at move 266 –Antti) actually revealed a bug in KataGo's cleanup phase in japanese rules, where black loses a decisive point by having to play an uneccesary move which is why it is so confused.