Failed attack on a surroundable but apparently safe group

[hendrik.falkenberg]

Usually I analyze my games with AI and the knowledge I got from courses, videos and books and find out approximately what the problem was. This game I am still confused after the engine review.

But in this game, my beginner-level understanding is that the group White has in the top/top-right had enough of my stones nearby that it was weak and surroundable, yet the engine criticizes all my moves and would love to play somewhere else like 3-3 in the top left, and somewhere entirely different on the board after that. I think I understood which groups are good targets for attacks and which not totally wrong.

After I realized his group is totally safe and my last ~20 moves have been wasted, I resigned. Maybe a little early as my attempt to surround also gave me some influence in the middle, and in 15kyu games there are blunders all the time so I could have played on for a bit in hindsight.

[ft3g0]vs[V240348213]_20251020_15K_15K.sgf (584 Bytes)

The game was on Fox and I don’t think those have true weblinks so I uploaded the sgf, if any other way to view it is possible and more comfortable for the reviewer I can accomodate

[square.defender]

[Uberdude]

What is the earliest point you can now identify the white group as unconditionally alive? How about during the game? Where are its eyes?

[hendrik.falkenberg]

Thank you. I uploaded the game to ft3g0 vs. V240348213

[hendrik.falkenberg]

On move 46, when my one stone was surrounded, I thought that this will be an eye and there is most likely another way to make an eye on top, so then I played more in that area for containment than killing the group, but admittedly also due to stubbornness.

In retrospect, while I think of this point still as “unconditionally alive” as one eye is basically already there, already at move 40 White has many possibilities to try to make eyes and I don’t think I can stop enough of them. On the top row there is at least one eye, P17 could always make a eye on top if one is needed and White could also threaten cuts against my diagonal top 3 stones, and after move 40 there is also the possibility to expand top row eye space even further. With all these factors, I don’t think I have good reason to keep attacking the group

Edit: about White’s P17 to make an eye. I just looked at the variation where I play P17 myself to prevent this and it doesn’t work: bP17 wP16 bQ17 wP15 and the O15 stone dies and White gets an eye

[jlt]

If you don’t have a good attacking move, it’s better to play elsewhere, like create another group or make points.

[Uberdude]

@hendrik.falkenberg

On move 46, when my one stone was surrounded, I thought that this will be an eye

Yes, good, n13 is likely to be an eye, and it definitely is at move 48 (without that white move locally speaking black can make it a false eye with o12 o13 m12).

@hendrik.falkenberg

and there is most likely another way to make an eye on top

@hendrik.falkenberg

On the top row there is at least one eye, P17 could always make a eye on top if one is needed

Yes, not just likely, but definitely an eye already, even if black plays first you can’t stop white getting an eye at p19. So white has had 1 eye already since move 32, so has only needed to run away and make 1 more eye in the centre to be alive and the attack is over. This can be traced back to white getting the slide to q18 on move 28, this makes some eyespace on the edge, where it’s much easier to make eyes than the centre. So if you wanted to attack white’s o17 approach, the standard idea is to kick at p17 to prevent white getting that slide and the eyespace. Your n16 move isn’t terrible in itself if you want to build the centre, but it’s not an attacking move because it doesn’t stop white going in the direction of the corner which is the easiest place to make eyes.

As to why c17 is important, it’s big for points and base of both group in that corner. Once the white top right group is settled (and it was never in big danger), the black group at L16 doesn’t actually have eyespace of its own though it’s jumping out to the centre, but a way to use this influence is to pressure the d17 group: after c17 you are threatening f17 to steal the eyespace and make it into a weak group, as well as making your c16 stone surely alive, a multi-purpose move.

[JethOrensin]

@hendrik.falkenberg

I think I understood which groups are good targets for attacks and which not totally wrong.

I do not have anything to add to the already great advice you got on the issue, so if you do not mind a similar suggestion on this could be adding the thought of “what do I want from this part of the board?” in parallel to whatever other opportunity comes along on the board.

For example, when you get out of the opening, this is what your moves declare that you want:

But when White approaches, you keep changing plans and you try to achieve multiple conflicting goals.
For example, now you suddenly want this:

Right afterwards you are back to wanting this:

And then you are back wanting the corner as well. In the end, you are also trying to attack the White group and you end up losing your main initial goal, for not enough compensation:

If you want to decide to attack or defend or gain something, remember to look around and utilise your stones. If you want to attack, then attack by using the stones you have.

A simple example:

Now all your triangle stones are working together, if it was a joseki then you are already one move ahead from White with R14 already in position, so if White does the usual extention then:

…then White could end up with a stick with no base and not many places to run. It might find a way to live, but you will gain a lot of moves while it struggles. Try to visualise the same board without the triangle marked stones. It would be ludicrous. But the existence of these stones give you the strength to play those moves, because the stones work together.

Last, but not least, even if you find yourself straying from your goals, you can still return to them, but at some cost. The more moves you lapse away from your goals, the more the cost, however, sometimes it is still all good. For example, even if you played the same exchanges initially, you could still rotate back to building in the center:

White is building a wall while pushing from behind, and that wall is facing your stones. Meanwhile you are building towards the center, your are ahead in the pushing and you can start thinking of setting up some sort of attack on white’s stones on the top left or, even better, reducing/invading at the bottom side.

So, simple moves and a simple thought: Decide what you want and try to get it, but also try to avoid changing goals, on the same local part of the board, with every couple of moves. Have fun!

[uzaya]

I am not as good (ps- not just as good but like… nowhere remotely near) as the other people commented on that but my initial idea upon seeing the O17 approach was attempting to split it from white’s left upper group with a move like L17 or K16. I am unsure if it is tactically sound or strategically feasible, but at least I reduce the possible northern area of white at the top and even if I couldnt kill the O17, if I succeed at splitting it then I can potentially limit the area it steals from me.

[hendrik.falkenberg]

Thank you all for your helpful answers!

[JethOrensin]

@uzaya

I am unsure if it is tactically sound or strategically feasible,

It is very sound, indeed: