Non-local move versus a local move

  Difficulty: Beginner   Keywords: Tactics

This page needs wiki master editing.

Compare the following two situations (grossly oversimplified):

Black tries a ladder  


Black tries a net  

The way I see it (25-20 kyu), the second move is better because Black does not have to worry as much about the rest of the board.

The ladder depends on the absence of white stones in the ladder-breaking-area (marked points - possibly I'm a few lines off here :-) ) while the net nearly always works.

Correct. Kageyama writes:

  1. Check if you can capture a stone in a net.
  2. Check if you can capture it in a ladder.

I would like to comment on your diagram, though. In the position on the diagram both moves are small, because the white stone is not important (non-cutting stones). If you capture it you'll spend a move for two points, which is too little with so much empty space. As soon as the black group is alive, everything in this corner is endgame.

I believe that one of the reasons low kyu players lose their games is because they make too many small moves. Get rid of the moves that capture a single unimportant stone, and your strength will increase considerably.

To illustrate the same principle I would suggest a following diagram that is logical.


Here, the marked stone may run away, splitting Black in two groups. Each of them will have to live separately. Now B1 not only captures a stone, but also connects two groups and is likely to form an eye; which makes it a very good move.

12 kyu or so chiming in with a question. In the example just above where the white stone cuts, the net is clearly a fine move. In addition to the reasons above, it leads out toward the center.

However, as black in the original position I wouldn't even consider the ladder and probably would not play the net either. I would quite likely tenuki.

But what about ...  

If I did play in the area, it would be at a, b or c (roughly in that order of preference with no other stones nearby) or their mirror images on the other side, aiming to expand my moyo. These moves also threaten the white stone a bit, but that does not seem important.

Or should one play the net first, then expand from a stronger base?

Kageyama also says: if you can capture a stone firmly, don't capture it loosely.

Applied to your diagrams this means:

The net  

The net leaves forcing moves at a and b.

The "ladder"  

The ladder captures the stone more firmly, and actually it's no ladder at all, because after W2 Black can play geta with B3.

(In fact, as a net, it leaves kikashi at a, b and c... ;-)) --AvatarDJFlux)
: Surely the same would be true of the other net. But in this case, White has to part with another stone for the dubious benefit of the kiksahi plays.
:All the same, the example is not good, because the Black groups are not split, both players are playing too close to thickness.

I'm not saying that this is the optimal move; I'd play the net move at B1 myself, but one should always think about options and (gasp) think some moves ahead.

-- Arno Hollosi


Closing the net  

This is the best enclosing technique I can see. Having this option is clearly preferable to having only a ladder aimed out into the wild blue yonder, but it still has to reach out a significant distance, and is also vulnerable to spoilers nearby. The early net seems much more decisive and reliable.

Uh oh...  

The above attack is dependent on the rather artificial isolated corner position. In my humble opinion, a shape like this would be farther out and much less isolated. Furthermore, it also has the above-mentioned forcing points.

I've played around with this out in the middle, and it's messy. I'm not sure, but I think White can live, even without any help around.

It's seems to be a complicated situation. Can anyone point out a foolproof capture I missed?

AvatarDJFlux (3k): Yes, you spotted the right sequence in the first diagram, except that White will never play it (especially W5...), because we hope White can read that there is no way out. Instead White will keep the moves in reserve as aji, for ko threats.

Of course the sequence works because you can drive White to the edge. So in your "Uh oh" diagram it does not work, if there are no edges or friendly stones in the right places nearby.

But: even considering that position in isolation on the Goban, White will have anyway a hard time trying to save the eyeless and shapeless string of stones, caught in the middle of two strong black positions!
In fact, as already mentioned by Arno, the whole discussion is a little off the mark: if the white stone is not a dangerous cutting stone, and sits all wilting and almost dead against a strong wall of black stones, White won't try to save it, as adding moves there just makes the resulting group heavier and weaker, a perfect target for attack; neither will Black add moves to 'overkill' it.
Unless of course the surrounding position changes dramatically in White's favour...

TheDullBlade: Ah, thanks. My point was that the "turn the ladder into a net" trick does not produce a neat, small widely-applicable net as the initial net does, but is dependent on odd circumstances. Usually a wall like that will exist to wall something off, in which case White can probably escape, either by reaching around or running out when Black blocks the reach. In the process, she would destroy the eye point preserved by the initial net and threaten to surround and isolate Black's wall.

This is my second encounter with Kageyama's "choose the firmer capture" principle (the first at geta). Both have lead to rather controversial advice. Is this preference for ladders really what he meant?

DJ: Not exactly: the idea is to capture the cutting stone(s) in a way to minimise the possibilities of threats of pulling it (them) out. In other words, the firmer way is the one that gives the opponent the smallest number of kikashis (peeps, ladder breakers, etc.)
Please bear in mind that very often the combination of net and a loose ladder (as in your first diagram) is the only way to accomplish the capture of the cutting stone(s)...

Non-local move versus a local move last edited by tapir on June 26, 2008 - 12:53
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library