Seki or not?

[Verrius]

It doesn’t affect the game; but I was wondering whether my attempt to make a seki worked here or not:

Tournament Game: Through the Years: Long Correspondence (59567) R:1 (Verrius...

The computer counted the white stones in the bottom right as captured, and at the time that seemed right; but afterwards I started wondering whether I might not have succeeded after all. White can take the stone at T4 without any danger, but they can’t take the stones at S1 and T2 (I think).

[polar-bear]

Sure looks like seki to me!

[BHydden]

Correct. Black can never play at S2 without putting themself in self atari, so that corner would be rightly scored as seki… our current score estimator struggles with identifying seki

[Atorrante]

If one of the players places a stone, the result is that the other player will capture a group. If both players sit still and don’t move, they will both live in seki.

List of Go terms (Wikipedia): Seki (セキ) is a Japanese term for an impasse that cannot be resolved into simple life and death. It is sometimes translated as “mutual life.” For example, a capturing race may end in a position in which neither player can capture the other. There are numerous types of seki position that can arise, characterized as cases in which neither player adds a play to groups that do not have two eyes. The area remains untouched; at the end all groups involved are deemed alive, but no points are scored for territory.

Score estimator is often wrong

[Verrius]

Thank you for the replies. I’ll try to remember this if it comes up again.

[shinuito]

So with Japanese rules, black should capture the one stone at T4 with T5 to get a point, because eyes won’t be counted in seki right?

[yebellz]

Technically, yes. However, I think most players would simply accept that stone as dead without the formality of capturing it.

Odd Cases in the Japanese Rules

Compulsory Capture of Dead Stones In most situations, the Japanese rules compel players to pass before capturing all strategically dead stones, since playing additional moves to actually remove dead stones would reduce one’s points under territory scoring. For example, consider the following position. [dead%20stone] In this corner, black has 4 points (3 from territory plus 1 for the white prisoner). However, if black were to play an additional move here (at the end of the game, when they shou…

[shinuito]

I guess it’s just a difference in applying procedure. In an over the board game maybe you can just remove the stone as captured in the counting phase and ignore that corner for counting purposes later.

On OGS, I don’t think I’ve too much experience marking seki in my games so far, but I presume it’d be better to just capture the stone? I wonder if you mark the stone dead will it auto-count as two points even if you mark the other points as neutral? (The score estimator is fairly awful to change individual points and stones in my experience whereas the score agreement phase calculator seems to be nicer.)