Life In 19x19

I have been going through a Chinese endgame collection which is in two volumes, low dan and high dan, and I it occurred to me that that the problems delve into lot the complexities that make endgame difficult. This appears to sometimes be intentional. The book presents solutions on the next pages (usually only one) after the problems and there is sometimes a paragraph explaining what the problem was about. Other times it appears unintentional or alternatively that something was intentionally left out of the answers. It is understandable if the author wanted to stay on topic and avoid tangents, so I don't fault the book for this. When I solve problems I try to be pretty much certain that I have worked out the entire solution and then the use the answer section is mostly just to check that I'm solving the problems at all. Sometimes I find that problem books are harder than I expected or that I'm not able to get into the right mindset and end up tricking myself into thinking that I have the right solution. Getting to the point, I think some of the problems raise interesting questions about endgame.

For example this one, you might want to try it yourself so I will continue in a hidden section below...

Click Here To Show Diagram Code

[go]$$B Black to play. $$ +------------------+ $$ | . . . . . . . . .| $$ | . . . . . O X . .| $$ | . . , O . O X X X| $$ | . O . . O X X O .| $$ | . O . O O X O O .| $$ | . X X X . X O . .| $$ | . . , . X X O . .| $$ | . . . . . X X O .| $$ | . . . . . . . O .| $$ +------------------+[/go]

I hope I didn't make a mistake in my solution but I'm really interested to know who considered that white could hane with and how they approached that position.

I am going back some 50 years and so the memory is hazy, but I recall there was a book (by Kageyama or Rin Kaiho?) which illustrated this concept in some detail. It caused quite a stir among dan players in the London Go Centre, and it acquired an English name. I've forgotten it for the moment but the idea behind it was something like mutual destruction, this being seen as being better than than repeated submission (i.e. gote).

The intense interest it aroused may have been a sign of the times. There was then a lot of talk about the theory of (nuclear) mutual assured destruction in the political and military spheres.

kvasir wrote:I have been going through a Chinese endgame collection which is in two volumes, low dan and high dan...

Is it the sort of problem book that people can follow without reading more than a tiny amount of Chinese? Would you like to share the name of the book?

John Fairbairn wrote:I am going back some 50 years and so the memory is hazy, but I recall there was a book (by Kageyama or Rin Kaiho?) which illustrated this concept in some detail. It caused quite a stir among dan players in the London Go Centre, and it acquired an English name. I've forgotten it for the moment but the idea behind it was something like mutual destruction, this being seen as being better than than repeated submission (i.e. gote).

Nowadays I think the usual English term is mutual damage.

For this problem, I had a different (and less interesting) mutual damage line in mind:

xela wrote:Is it the sort of problem book that people can follow without reading more than a tiny amount of Chinese? Would you like to share the name of the book?

Yes, since it is a problem book you don't really need to read the text. That said, the text is sometimes easier than one might expect. I'm not really trying to read it, maybe because I have more interesting Go books in Chinese when I want to study Chinese.

I think it is enough to know that the problems are all black to play and black wins by 1 point and there is no komi and no captures. Then you need to be able to tell the right solutions from the incorrect examples. I guess this is enough information but there are some conditions stated up front in the book. It is maybe not unexpected that the book counts eyes in seki as territory. And, yes, it is always counting territory, at least so far. I don't think Chinese make as much fuss about area vs. territory (at least usually) as many people seem to think

The name of the books are:

收官之旅——趣味官子练习（低段篇）
收官之旅——趣味官子练习（高段篇）
by Li Junkai / 李君凯

I guess the title in English is something like "Endgame journey: an interesting endgame exercise book (low/high dan volume)" or even fun instead interesting.

I found these links that don't require login and didn't time out:
https://www.books.com.tw/products/CN11963780
https://www.books.com.tw/products/CN11963961

As mentioned above. There is chapter 11 "Endgame Pointers" in Kageyama's Lessons in the Fundamentals of Go. Which is a surprisingly valuable book to read despite how hard it is to relate to the constant chiding of what must be the book's most likely readers. It helped me to accept the book when i realized that he is also chiding some of his fellow pros that have done things like dedicate whole chapters in books to tricks to avoid reading ladders to the end and other such things.

The first 3-4 pages discuss how black should not play when white plays some sente looking moves in the endgame. It is a 5 stone handicap game and black doesn't have many if any weaknesses. Kageyama says it "may have been unavoidable" to defend after white's first move but after this there is a huge difference in terms of territory between black defending everywhere or playing good moves elsewhere. Then this short chapter turns to how to evaluate the size of endgame moves and finally there are some examples of different sorts.

I'm not sure what I think of his advice in the context of 5 stone handicap games. I think white is likely to try strong arm tactics to make something happen. Since it is a handicap game there are slightly different dynamics at play but if black always makes the right tennuki or the right defensive move then it is always futile for white anyway. The position in the book is possibly setup to not lend itself easily to white's shenanigans.

This is the position discussed in Lessons in the Fundamentals of Go.

Click Here To Show Diagram Code

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

Click Here To Show Diagram Code

[go]$$Wm11 $$ +--------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . .| $$ | . . . X O . . . . . . . . . . O 5 6 .| $$ | . . . X . O . . . . . . . O . . X . .| $$ | . . . X . . . . . , . O . . . X . . .| $$ | . . . X O O O . . . . . . X . . . . .| $$ | . . . X . X . . O O . O . . . . X . .| $$ | . . . . . . X . X . . . . . . . . . .| $$ | . . . X . . . . . . . . . . . . . . .| $$ | . X 2 X . . . . . . . . . . X . X X .| $$ | . O X O . X X . . X . . . . . , . O .| $$ | . . 1 . O O . . . . . . . . O . O . .| $$ | . . O . . . . . . . . . . . . . . . .| $$ | . . . . . O . X . . . . . . . . . . .| $$ | . . O . . . . . X . . . X . O . O . .| $$ | . . . . 7 . X X O O . . . . . . . O .| $$ | . O . X 8 X . O . , . . X X X X X X 9| $$ | . 3 X . . . X O . . O O X O O . . . 0| $$ | . 4 . . . . X O . . . . O . O X . . .| $$ | . . . . . . . . . . . . . . X . . . .| $$ +--------------------------------------+[/go]

...and so on up to .

Click Here To Show Diagram Code

[go]$$B $$ +--------------------------------------+ $$ | . . X O O . . . . . . . . . . . . O .| $$ | . . X X O . . . . . . . . . . O O X .| $$ | . . . X . O . . . . . . . O . . X X .| $$ | . . . X . . . . . , . O . . . X . . .| $$ | . . . X O O O . . . . . . X . . . . .| $$ | . . . X . X . . O O . O . . . . X . .| $$ | . . . . . . X . X . . . . . . . . X .| $$ | . . . X . . . . . . . . . . . . . . X| $$ | . X X X . . . . . . . . . . X . X X O| $$ | . O X O . X X . . X . . . . . , . O O| $$ | . . O . O O . . . . . . . . O . O . .| $$ | . . O . . . . . . . . . . . . . . . .| $$ | . . . . . O . X . . . . . . . . . . .| $$ | . . O . . . . . X . . . X . O . O . .| $$ | . . . . O . X X O O . . . . . . . O O| $$ | . O . X X X . O . , . . X X X X X X O| $$ | . O X . . . X O . . O O X O O . . . X| $$ | O X X . X . X O . . . . O . O X . X .| $$ | . . . . . X O O . . . . . . X . . . .| $$ +--------------------------------------+[/go]

Black responded directly to every threat and the game starts to look close. Well, the book says white is ahead by 10 points, which doesn't sit right with me, and he says black lost 30 points.

Then there is the remedy, which looks very simple.

Click Here To Show Diagram Code

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

But is this so simple? We could try to help white. How about the following...

Click Here To Show Diagram Code

[go]$$Wm5 $$ +--------------------------------------+ $$ | . . . . . 4 . . . . . . . . . . . . .| $$ | . . . . X 3 2 5 . . . . . . . O 1 . .| $$ | . . . X . O 6 7 . . . . . O . . X . .| $$ | . . . X . 8 . . . , . O . . . X . . .| $$ | . . . X O O O . . . . . . X . . . . .| $$ | . . . X . X . . O O . O . . . . X . .| $$ | . . . . . . X . X . . . . . . . . . .| $$ | . . . X . . . . . . . . . . . . . . .| $$ | . . . . . . . . . . . . . . X . X . .| $$ | . . X , . X X . . X . . . . . , . O .| $$ | . . . . O O . . . . . . . . O . O . .| $$ | . . O . . . . . . . . . . . . . . . .| $$ | . . . . . O . X . . . . . . . . . . .| $$ | . . O . . . . . X . . . X . O . O . .| $$ | . . . . . . X X O O . . . . . . . O .| $$ | . O . X . X . O . , . . X X X X X X .| $$ | . 9 X . . . X O . . O O X O O . . . .| $$ | . . . . . . X O . . . . O . O X . . .| $$ | . . . . . . . . . . . . . . X . . . .| $$ +--------------------------------------+[/go]

Probably not the best moves but maybe enough to suggest that the complexity of the choices has increased dramatically.

I think black can only have lost 30 points in the main line if we think that black's defensive moves had no value, that only white made some good moves. That seems to me like a wrongheaded way to look at it. Black's moves did have value even though slightly less than white's. Based on this I wouldn't expect black to have lost whole 30 points in the mainline.

Maybe Kageyama glossed over the complexities here but this seems to serve his method for the book, which seems to be to present examples in a way that the reader will find memorable.

kvasir wrote: I think it is enough to know that the problems are all black to play and black wins by 1 point and there is no komi and no captures. Then you need to be able to tell the right solutions from the incorrect examples. I guess this is enough information but there are some conditions stated up front in the book. It is maybe not unexpected that the book counts eyes in seki as territory. And, yes, it is always counting territory, at least so far. I don't think Chinese make as much fuss about area vs. territory (at least usually) as many people seem to think :)

Yes, I did a bunch of 7x7 endgame problems on 101weiqi a while ago. The solutions were of variable quality (with some valid alternative solutions being marked incorrect), and the problems less interesting than the one you've posted here, so they don't make my "highly recommended" list. But they were also no komi and black to win by 1 point without exception. In most cases they were also B+1 by Japanese rules. But a few were jigo by Japanese rules, and the point was to find the variation which gave black the last move, for the extra point by Chinese rules.

kvasir wrote: I found these links that don't require login and didn't time out:
https://www.books.com.tw/products/CN11963780
https://www.books.com.tw/products/CN11963961

Thanks! If Google Translate and I are reading this correctly, the price on the first link of "定價：234元 優惠價：87折204元" is something like "normal price 234 yuan, special price 87% of that, pay 204 yuan", so about US$28 not including postage? It's around a tenth of that price on Taobao... but then again, ordering from Taobao outside China is a baffling ordeal.

Thanks also for the excerpts from Kageyama. I've been meaning for a while to re-read that book and ponder how much of it is overstated to make a [still useful for most people most of the time] point. I'm already aware that KataGo disagrees with the earlier chapter on nets and why you should make the tightest capture possible (I think there was a thread here on that a few years back).

xela wrote:Thanks! If Google Translate and I are reading this correctly, the price on the first link of "定價：234元 優惠價：87折204元" is something like "normal price 234 yuan, special price 87% of that, pay 204 yuan", so about US$28 not including postage? It's around a tenth of that price on Taobao... but then again, ordering from Taobao outside China is a baffling ordeal.

That would be 204 Taiwan dollars for the high dan volume which looks like it would be around 6 USD. I think this is with discount depends on some membership level. There is a note on the bottom of the page saying this could mean 21-30 working day delay for them to receive the books when not stocked. I saw on google that you can get both books as a set for ~5 USD on taobao. It is a question of shipping, convenience and other things. Anyway, the prices are listed in Taiwan dollars, not Chinese yuan, on the linked site.

xela wrote:Thanks also for the excerpts from Kageyama. I've been meaning for a while to re-read that book and ponder how much of it is overstated to make a [still useful for most people most of the time] point. I'm already aware that KataGo disagrees with the earlier chapter on nets and why you should make the tightest capture possible (I think there was a thread here on that a few years back).

These kind of Go books are bit like self-help books. You will be very frustrated with them if you think that you will find lot of information about how to play. For example in the chapter about ladders and nets. His point is what he says on the last page, that you can't really master even the most elementary things (ladders and nets) without reading. Reading this in a book could motivate someone to work harder on ladders and net, which would help them improve at Go, but maybe the examples and the impetuous for this new motivation isn't what is important. That people will remember this book over 50 years after it was published must mean that the motivation works.

There also seem to be deep layers of irony, or sarcasm, in the book. What do we make of it that the main example in the Ladders and Nets chapter is a game against Takeo Kajiwara, which Kageyama won splendidly, and the 'Lessons' are published the same summer (according to the introductions) as the Direction of Play? 'Direction' has a chapter about "the professional approach" which starts off by comparing professionals and amateurs, stating for example that professionals shouldn't laugh at amateurs and that strong players shouldn't lord it over weaker players. Then in the 'Lessons' there is this example when Kageyama beat Kajiwara and a quote from professionals who is berating him for his "asinine" play. Then he says he would probably have played the amateurish keima, instead of the professionally approved net, if he was black and something like that he is hoping to improve himself and be able to think like a professional.

I think that Kageyama's point about the net example is that people are going to disagree about such things all the time. The keima, to my understanding, looks like a maximum move. Maybe the professional attitude is that if black is already better, since white shouldn't let black capture those stones, then you can capture tightly to make sure they stay captured. Basically, a professional doesn't need to win by 10 points when 2-3 points will do. However, it is mentioned that Hashimoto Utaro approved of the diagram for white and this is sort of swept aside by the quote from the professional who disapproved. Maybe this opinion from Hashimoto should have been reason to scrutinize black's play, but it is dismissed as sarcasm. The professional approach, here, was maybe not so different from any other arguments when people quickly pick sides and fail to make further progress.

kvasir wrote:That would be 204 Taiwan dollars for the high dan volume which looks like it would be around 6 USD.

Ah, silly me. Of course Taiwan dollars and Chinese yuan use the same Chinese character, so I can't expect Google translate to tell me the difference. I should have figured it out from the domain name ending in .tw. Thanks!

kvasir wrote:There also seem to be deep layers of irony, or sarcasm, in the book. What do we make of it that the main example in the Ladders and Nets chapter is a game against Takeo Kajiwara, which Kageyama won splendidly, and the 'Lessons' are published the same summer (according to the introductions) as the Direction of Play? ...

Interesting context. I have both books side by side on my shelf, but didn't realise they were so close in time! The Direction of Play comes across as more sophisticated, but I never did buy into the "move 2 lost the game" mystique.

kvasir wrote:This is the position discussed in Lessons in the Fundamentals of Go.

$$Wc $$ +--------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . .| $$ | . . . . . . . . . . . . . . . O . . .| $$ | . . . X . O . . . . . . . O . . X . .| $$ | . . . X . . . . . , . O . . . X . . .| $$ | . . . X O O O . . . . . . X . . . . .| $$ | . . . X . X . . O O . O . . . . X . .| $$ | . . . . . . X . X . . . . . . . . . .| $$ | . . . X . . . . . . . . . . . . . . .| $$ | . . . . . . . . . . . . . . X . X . .| $$ | . . X . . X X . . X . . . . . , . . .| $$ | . . . . O O . . . . . . . . O . O . .| $$ | . . O . . . . . . . . . . . . . . . .| $$ | . . . . . O . X . . . . . . . . . . .| $$ | . . O . . . . . X . . . X . O . O . .| $$ | . . . . . . X X O O . . . . . . . . .| $$ | . O . X . X . O . , . . X X X X X . .| $$ | . . X . . . X O . . O O X O O . . . .| $$ | . . . . . . X O . . . . O . O X . . .| $$ | . . . . . . . . . . . . . . X . . . .| $$ +--------------------------------------+

Click Here To Show Diagram Code

[go]$$Wc $$ +--------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . .| $$ | . . . . . . . . . . . . . . . O . . .| $$ | . . . X . O . . . . . . . O . . X . .| $$ | . . . X . . . . . , . O . . . X . . .| $$ | . . . X O O O . . . . . . X . . . . .| $$ | . . . X . X . . O O . O . . . . X . .| $$ | . . . . . . X . X . . . . . . . . . .| $$ | . . . X . . . . . . . . . . . . . . .| $$ | . . . . . . . . . . . . . . X . X . .| $$ | . . X . . X X . . X . . . . . , . . .| $$ | . . . . O O . . . . . . . . O . O . .| $$ | . . O . . . . . . . . . . . . . . . .| $$ | . . . . . O . X . . . . . . . . . . .| $$ | . . O . . . . . X . . . X . O . O . .| $$ | . . . . . . X X O O . . . . . . . . .| $$ | . O . X . X . O . , . . X X X X X . .| $$ | . . X . . . X O . . O O X O O . . . .| $$ | . . . . . . X O . . . . O . O X . . .| $$ | . . . . . . . . . . . . . . X . . . .| $$ +--------------------------------------+[/go]

OK, I KataGo-d it. You're right: the more I look, the more interesting it is.

My first impressions before running up KataGo: I had black about 18 points ahead (assuming no komi since it's a handicap game). KataGo said I was too pessimistic: playing out some sequences, there's a lot of potential for black's top right to grow and white's top to shrink. So KataGo has black 32 points ahead in the starting position.

After Kageyama's 38 moves of black being passive, KataGo says it's B+12, so black has lost 20 points, not the stated 30 points. Near enough: in a 5-stone handicap game, you might still expect white to be the favourite to win at this point.

What really stands out to me, playing out some "mutual damage" sequences and asking the AI what it thinks, is the instinct I've retained from my kyu player days: of course you need to respond to white's move or else your group totally dies!. Especially for black's bottom right. But actually...

Click Here To Show Diagram Code

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

KataGo rates as a mistake. There's a bunch of complicated variations, but black's "dead" group will be able to connect out either to the left or above.

I suspect this confidence in your ability to fight your way out of a tight spot is a large part of endgame strength.

"there's a lot of potential for [one player's region(s)] to grow and the [opponent's region(s)] to shrink."

Interesting concept - thanks, xela!

xela wrote:After Kageyama's 38 moves of black being passive, KataGo says it's B+12, so black has lost 20 points, not the stated 30 points. Near enough: in a 5-stone handicap game, you might still expect white to be the favourite to win at this point.

The book says white is ahead by about 10 points. It seems to serve the example to say white is so much ahead. That is just how this book is.

xela wrote:What really stands out to me, playing out some "mutual damage" sequences and asking the AI what it thinks, is the instinct I've retained from my kyu player days: of course you need to respond to white's move or else your group totally dies!. Especially for black's bottom right. But actually...

$$Wc19 $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . 2 . 4 . . . . . . . . O . . . | $$ | . . . X . O . . . 6 . . . O . . X . . | $$ | . . . X . . . . . . . O . . . X . . . | $$ | . . . X O O O . . . . . . X . . . . . | $$ | . . . X . X . . O O . O . . . . X . . | $$ | . . . . . . X . X . . . . . . . . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . X . X . . | $$ | . . X . . X X . . X . . . . . . . . . | $$ | . . . . O O . . . . . . . . O . O . . | $$ | . . O . . . . . . . . . . . . . . . . | $$ | . . . . . O . X . . . 7 . . . . . . . | $$ | . . O . . . . . X . 8 . X . O . O . . | $$ | . . . . . . X X O O . . . . . . . 1 . | $$ | . O . X . X . O . . . . X X X X X . . | $$ | . . X . . . X O . . O O X O O . . 3 . | $$ | . . . . . . X O . . . . O . O X 5 . . | $$ | . . . . . . . . . . . . . . X . . . . | $$ +---------------------------------------+

Click Here To Show Diagram Code

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

KataGo rates as a mistake. There's a bunch of complicated variations, but black's "dead" group will be able to connect out either to the left or above.

My interpretation is very different.

• Black, in handicap games, should prefer to finalize his shapes and secure his advantage. If about 10 points lead is enough to win after the 38 move variation, then that is by far the easiest way for black. In case the lead isn't enough then there are ways to improve on black's play without risking having to find any special moves. Basically, black should be able to win simply.
• The main reason that all these variations work out for black is that he might be 30 points ahead.
• A secondary reason, well it is related to why black is 30 points ahead, is that white's groups and his territory is mostly weaker than black's. Once forceful moves are being played it is possible for black to ignore quite many of them before he is in trouble, however, for white it is not like that. White ends up with almost dead groups and smashed territories before he can really hurt black.

Maybe what I called a "a secondary reason" starts to get at the kernel? If I paraphrase what I just wrote it would be: When you are faster to damage or destroy your opponent then you don't need to defend.