An idea I've been playing around with mentally, but am having some difficulty properly formalizing in a rules context is Auction ko. Basically the idea is instead of fighting with a ko threat, after the initial ko capture the other player makes an offer of some number of points. The person who made the initial capture has a choice:

1: Reject the offer - The player rejecting the offer hands his opponent that number of captured stones, and gets to play a second move in a row (presumably completing the ko).

2: Accept the offer - The player who made the initial capture is given the offered number of stones, however now the opponent gets to make 2 moves in a row (one of which must retake the ko, the other presumably used to finish the ko).

The idea is that a ko fight will no longer favor the person who has more ko threats, but will instead favor the person who can more accurately value the ko (regardless of the board position). I could see it potential making ko a very powerful weapon for those who are good at counting, it also would reduce the strategic effect of winning one ko then immediately starting another (because you know your opponent has no threats).

Any thoughts on how it might affect a game? Any ideas on how to properly codify this?

I think there was a ruleset by BillSpight that used cards places aside the board with different point values, that you could take instead of making a move.
I think it was an educational roleset about the concept of temperature. It was posted in Sensei's but I don't remember the name now. It should be linked from or to ButtonGo, I think.

The concept seems similar to yours, which is why I mentioned it.

_________________
a1h1 [1d]: You just need to curse the gods and defend.
Good Go = Shape.
Associação Portuguesa de Go

It seems inelegant. Why not simply say that the n-th time a board is recreated, the player recreating gives summation(I)(from I=1 to I=n) prisoners? The con (from your point of view - I like it) is that ko threats still have a value. The pro is a simple bidding mechanism that doesn't require introducing consecutive moves to the game.

Edit: I forgot to say that either way, it seems hard to find a rule for this that won't demolish the difference between taking first and taking second.

Yes, I believe it's called environmental go (I'd try to link but I'm on a phone). Perhaps I should have called this environmental ko?

Your suggestion is interesting (and exactly the kind of feedback I was hoping for!). It is fundamentally different though, as you say, in my version there is no value to capturing first, while in your version the first "repeater"cannot hope to profit (eliminating throw in ko). Actually in practice I think your version would end up reducing to basically a regular ko fight.

I don't think it would reduce... not necessarily. It basically allows you to negotiate a price at which you buy extra threats. You both retake, retake, retake, bidding up the ko, and then your opponent can play a ko threat, cashing in on the N points that he has netted and now leaving you the choice of whether or not to answer the threat.

I think where this would see the most action is in a case where clearly no threat is bigger than the ko. But in any case where there is a reason to bid on the ko rather than use ko threats, going first is no longer a real advantage.

I guess you could say that I've proposed a way of bidding for extra moves in a ko, whereas you are looking for a way to bid directly on the outcome of the ko itself - but, imho, the latter route will have to be kludgy, since you're trying to make a rule about one of the emergent phenomena of the game (a ko fight) rather than one of the rules (not being allowed to repeat a position). Right? That aspect of what you're looking for seems as bad (bad as in, inelegant, not bad as in sinful) as offering a rule allowing both players to bid on whether a stone can get out of a ladder, or escape a net. Next we might as well roll 3d6 on the Combat Resolution Table to determine whether that monkey jump ends in sente or not!

So to understand this a little better, what will happen here (a double ko situation from the large avalanche joseki) when White captures at ? Black is not allowed to capture the double ko at "b" but must bid for the life of the group?

Click Here To Show Diagram Code

[go]$$W Double ko life
$$ --------------
$$ | . X . . . . .
$$ | O O X X X . .
$$ | b O O X O X .
$$ | O X X O . . .
$$ | X . X O . O .
$$ | 1 X X O . O .
$$ | X O O X X . .
$$ | O O X . . . .
$$ | . . O . . . .
$$ | . . . , . . .[/go]

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

One simple example:

Click Here To Show Diagram Code

[go]$$Bc 0 Value Throw in ko
$$ ------------------
$$ | . O X X . . . . .
$$ | O . O X . . . . .
$$ | O O O X . . . . .
$$ | O X X X . . . . .
$$ | a b X . . . . . .
$$ | O X . . . . . . .
$$ | O X . . . . . . .
$$ | . X . . . . . . .
$$ | X X . . . . . . .[/go]

If black attempts to profit by playing A, he creates a position (let's call it position 1). White captures with B creating a different position (position 2). Black recaptures making position 1 again and must give white a stone (first time position 1 is recreated). White recaptures recreating position 2 and gives the stone back (first time position 2 is recreated). Black recaptures and gives 2 stones (2nd time position 1 is recreated), white gives 2 stones back (2nd time position 2 is created)...There's never a time when black is profiting from this exchange, if white wants he can always just give back the stones black just gave him. in order to make it a profitable exchange black has to play a ko threat to reset the counter and change who captured first.

I think for the effect you're looking for you would want to say something like "On the n-th consecutive move that recreates a prior board position the player recreating gives summation(I)(from I=1 to I=n) prisoners" that would allow closer to the auction style (but still have ko threats).

As I mentioned before, I like the idea behind what you're saying because it tries to find a simpler / more elegant way to achieve what I'm thinking of, but as you note it's still note quite the same.

To my rule as stated/intended -- Black has to bid on the ko at 1, if white rejects black is captured. If white accepts the first bid (white is given, let's say 8 points), black gets to capture 1 and play B, white has the option of bidding to recapture B. If white bids lower (let's say 7 points), black will presumably just capture (otherwise white has been given free points and you're back where you started). White also wouldn't bid higher (because then you are just giving black free points and you're back where you started). If I'm analyzing it right, the swing value of your ko is going to be approximately 3 times the value of black's initial bid (because either black will get N points, or white will get 2xN points). To make things even more confusing though, just to create this position white has probably already been forced to bid on playing the 1-4 stone.

Doesn't this then come down to the same counting exercise that is required to determine whether an effective ko threat exists? You must determine the value of the two moves in a row plus the amount in play in the ko. It seems to me there is no new/different skill involved. What is involved is new restrictions on play that would narrow the options available. Not only things like double ko but also simply ignoring the ko altogether would not be possible. Or do I misunderstand the proposal?

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

You're right, the skill is effectively the same, the two main differences are now 1: You just need to know how big the ko threat has to be, you don't actually need to have the ko threat available on the board, and 2: You need to know exactly how big the ko threat needs to be, not just a ballpark of "Threatening this other group is bigger than the ko". I guess in actuality what it does it almost completely remove the value of ko (assuming you have a game of two perfect counters), because you will get the same value as the ko in return for your opponent finishing it.

What about mutli-step ko's where taking the ko is essentially worth three moves in a row for an opponent?