Loser's Go

[mbv]

This thread is about a varient of Go which may or may not work.
The rules thus far:

Rule 1: Area scoring is used.
Rule 2: In order to win, lose by 30 points or less.
Rule 3: When you score the game at the end, count up the difference in prisoners. Up to that amount can be optionally added to your opponents score.

I believe it is now time for a little less conversation and a little more action. Are there any volunteers to test out this prototype varient? And can anyone think of a better name than "Loser's Go"?

The gauntlet has been thrown. Let the games begin!


Waring: very long and messy self-depreciating thought process is hidden below.

Read at your peril, this is a timesink and you may become quite bored. However I didn't think it was appropriate to keep this on another thread where it would be off topic.

I've sometimes wondered if it would be possible to engineer a Loser's Go variant somehow.

There is a go variant where the first player to capture loses. See http://senseis.xmp.net/?AntiAtariGo

The best idea I could come up with was this:
Rule 1- In order to win, you must lose by a certain margin (5 points? 10?).
Rule 2- Something like, you cannot play within your own territory if your territory is safe. If a play is made within your own territory and it appears to be unnecessary, determine whether the territory was safe at that point (or after the game?). Breaking this rule, you forfeit the game.

The rules attempt to keep some kind of structure in place so that the game at least resembles go, as opposed to a series of hundreds of silly moves. I'm not sure if it would work though. (If anyone is willing to try it then we could play it in a new thread).

What is safe? Is removing aji or a ko threat safe, and so illegal? Is it safe and illegal if you play an inside move too early?

Thanks for the reply. Those are some good points.

I will amend Rule 2 for clarity and add a third.

Rule 2 - You cannot play within your own territory if your territory is safe. If you play within your own territory and it appears to be unnecessary (i.e. you are not defending a potential weakness), determine whether the move was unnecessary (or after the game? I'm not sure when this would be more appropriate.). Breaking this rule, you forfeit the game.

Rule 3 - If the opponent places a stone within or against your territory, you may defend against it.

So now, if the opponent threatens you or appears to in accord with rule 3, you may defend.

If you can't read it out but there is a sequence that means that you don't need to play a move you played, did you just forfeit, despite playing in good faith?

I'm not sure how to address this. If there may be a weakness but you can't read it out then hopefully your opponent can't read it out either. There are a few useful questions your point raises. Why would the opponent want to invade and capture your group when the victory condition is to lose (per rule 1? By a set number of points?) unless they need to capture your group so that they can catch up?
Example: The victory condition is to lose by 10 points or less. You are losing by more than 10 points and soon enough, if the game ends, neither of you will win. So you try to balance the score by capturing a group. At least with rule 3 it's okay to defend in that case.

I'm not sure how much fun this variation would actually be though, even if it did work. Feedback? Ideas? Disappointment?

p.s. In my defence, all I know is that if you've played cribbage and the variation where you the first player to reach 121 points (normal victory condition) is actually the loser, you know it's a lot of fun. The game is completely reversed. I wondered if Go could remotely capture that fun.

[skydyr]

Read at your peril, this is a timesink and you may become quite bored. However I didn't think it was appropriate to keep this on another thread where it would be off topic.

What is safe? Is removing aji or a ko threat safe, and so illegal? Is it safe and illegal if you play an inside move too early?

Thanks for the reply. Those are some good points.

I will amend Rule 2 for clarity and add a third.

Rule 2 - You cannot play within your own territory if your territory is safe. If you play within your own territory and it appears to be unnecessary (i.e. you are not defending a potential weakness), determine whether the move was unnecessary (or after the game? I'm not sure when this would be more appropriate.). Breaking this rule, you forfeit the game.

Rule 3 - If the opponent places a stone within or against your territory, you may defend against it.

So now, if the opponent threatens you or appears to in accord with rule 3, you may defend.

If you can't read it out but there is a sequence that means that you don't need to play a move you played, did you just forfeit, despite playing in good faith?

I'm not sure how to address this. If there may be a weakness but you can't read it out then hopefully your opponent can't read it out either. There are a few useful questions your point raises. Why would the opponent want to invade and capture your group when the victory condition is to lose (per rule 1? By a set number of points?) unless they need to capture your group so that they can catch up?
Example: The victory condition is to lose by 10 points or less. You are losing by more than 10 points and soon enough, if the game ends, neither of you will win. So you try to balance the score by capturing a group. At least with rule 3 it's okay to defend in that case.

I'm not sure how much fun this variation would actually be though, even if it did work. Feedback? Ideas? Disappointment?

p.s. In my defence, all I know is that if you've played cribbage and the variation where you the first player to reach 121 points (normal victory condition) is actually the loser, you know it's a lot of fun. The game is completely reversed. I wondered if Go could remotely capture that fun.

I think it's quite interesting as a thought experiment, but I keep coming back to the strategy of starting at a 1-1 point and making a dumpling, which I'm not sure how to address.

[mbv]

Haha. So much for that.

[Polama]

The problem is that awful moves are trivial to find in go. So any variant where you try to lose is going to be plagued by groups filling in their own eyes, stones scattered through an opponents territory, etc. This is different than cribbage, where its hard to not make some points in a hand.

Here's how I'd fix it. First, I'd probably increase the margin of loss, say to 30. If you lose by more than 30, you actually lose. So you can win by either doing sort of bad, or much better than your opponent. Now any strategy involving truly atrocious moves (like filling in eyes of big groups) can be defeated by your opponent switching to sane play.

Use area counting. Now filling in your territory wastes a turn, but doesn't actually change your score. Rule 2 can be dropped.

Switching to area counting means playing dead stones in your opponents territory becomes valid. So I'd add one more change. When you score the game at the end, count up the difference in prisoners. Up to that amount can be optionally added to your opponents score. So, if I capture 20 stones and you capture 17, I can give you 0,1,2 or 3 points. If you were losing by 1 point, now you're winning. On the other hand, if I'm losing by 49, I wouldn't add the prisoners, so I'm not behind too far.

This means getting captured is always bad for you. You don't want to take a lot of territory, but if you overdo it, your opponent will beat you by over the margin and you'll lose.

[mbv]

Wow those are really great ideas! It sounds like those two simple rule changes you suggest achieve exactly the point of the varient. To reverse as much of the game (i.e. the aim, strategy and tactics) as possible while still keeping the incentive to play an aesthetically normal game. As you pointed out, in cribbage the extremely limited choices make it easy to change the game into a losing varient. Whereas the flexibility of choice in Go means you really need an incentive to play "normally".

Here's how I'd fix it. First, I'd probably increase the margin of loss, say to 30. If you lose by more than 30, you actually lose. So you can win by either doing sort of bad, or much better than your opponent. Now any strategy involving truly atrocious moves (like filling in eyes of big groups) can be defeated by your opponent switching to sane play.

Genius. It took some time to wrap my head around this but that seems reasonable. (edit: and I was sure there was something in the rule 1. The problems with a "dumpling" strategy or trivial moves is - in addition to taking a long time to finish - you have so little control over how the game ends.)

Use area counting. Now filling in your territory wastes a turn, but doesn't actually change your score. Rule 2 can be dropped.

Switching to area counting means playing dead stones in your opponents territory becomes valid. So I'd add one more change. So, if I capture 20 stones and you capture 17, I can give you 0,1,2 or 3 points. If you were losing by 1 point, now you're winning. On the other hand, if I'm losing by 49, I wouldn't add the prisoners, so I'm not behind too far.

So now,
Rule 0 - Area scoring is used.
Rule 1 - In order to win, lose by 30 points or less.
Rule 2 - When you score the game at the end, count up the difference in prisoners. Up to that amount can be optionally added to your opponents score.

Thanks for writing this out so clearly and explaining it
It sounds like with your ideas the game could actually work

[skydyr]

Following Polama's suggestion, I wonder if near the end of the game, the player that is ahead can start filling their own territory to get back behind, since their opponent will run out of territory to fill first and be forced to make points by filling dame.

I'm also curious about the strategy of making several small living groups that you can easily fill an eye for, with maybe a 20-25 point swing as a result. Maybe the spread should be less than 30 points?

[Polama]

Following Polama's suggestion, I wonder if near the end of the game, the player that is ahead can start filling their own territory to get back behind, since their opponent will run out of territory to fill first and be forced to make points by filling dame.

You're right. Once again, go's bias towards the player with more points reveals itself =). Free passes would fix that, but it may screw something else up. For one thing, nobody would fill dame, but that's important with area scoring. Too bad, the rules were so simple...

How about this? A player can pass for free (no prisoner given). Two passes ends the game and the players alternatingly fill dame. You cannot gain additional territory during dame filling (if you left a hole in a wall that's your prerogative, but during dame the potential territory must be filled in with stones, you can't close off the wall). You can play non-dame points (if, for example, filling in revealed a weakness that needs fixing) but doing so allows your opponent to pass that turn. You can capture stones during dame filling if the opportunity arises.

I'm also curious about the strategy of making several small living groups that you can easily fill an eye for, with maybe a 20-25 point swing as a result. Maybe the spread should be less than 30 points?

That's what rule 2 is designed to prevent. If a 20 point group is captured, the opponent now has the option of giving you back the 20 points in the end. So the dead group might be worth the same as a living group (minus one of the eyes), but you've also given the opponent the option of not giving you those points if he's close to making the margin. Except for an extreme case where you need exactly one point less and the capture won't let your opponent beat the margin, you want to keep your stones alive.

Thanks for writing this out so clearly and explaining it
It sounds like with your ideas the game could actually work

Thanks. I'll have to try it out some time and see if it's actually fun =)
Although that raises a funny question: how would handicap work? Do you want handicap stones or not?

[Archivist]

this topic is awesome, please continue!

[mbv]

Twisting my melon man. You know these ideas are twisting my melon man.

Following Polama's suggestion, I wonder if near the end of the game, the player that is ahead can start filling their own territory to get back behind, since their opponent will run out of territory to fill first and be forced to make points by filling dame.

You're right. Once again, go's bias towards the player with more points reveals itself =). Free passes would fix that, but it may screw something else up. For one thing, nobody would fill dame, but that's important with area scoring. Too bad, the rules were so simple...

I didn't know it was a rule to fill in dame in area scoring; I thought it was simply a desirable action to take with area scoring. If you don't have to fill in dame then there's no reason to in Loser's Go right?

How about this? A player can pass for free (no prisoner given). Two passes ends the game and the players alternatingly fill dame. You cannot gain additional territory during dame filling (if you left a hole in a wall that's your prerogative, but during dame the potential territory must be filled in with stones, you can't close off the wall). You can play non-dame points (if, for example, filling in revealed a weakness that needs fixing) but doing so allows your opponent to pass that turn. You can capture stones during dame filling if the opportunity arises.

If the dame filling is obligatory under different rulesets, then your suggestion here sounds practical. Since this is a variation though, would it be better for the sake of simplicity to make dame filling unnecessary? (Forgive my ignorance, I've only played with the Japanese ruleset).

Thanks. I'll have to try it out some time and see if it's actually fun =)
Although that raises a funny question: how would handicap work? Do you want handicap stones or not?

If we were to play, I would need a handicap in some form. That really is a tricky situation. How about, black (9k) plays several moves (let's say 3 as he is playing a 6k) and for white's first three moves, black chooses where white will play I love the sound of the Choice Go variation linked in the link Bill Spright shared at the beginning of the thread. Maybe the weaker player can choose the other player's moves (equal to his handicap?)?

edit: Thinking about this again, maybe the handicap could just work as in a normal game. Even if it's 9 stones, that simply means Black has 9 places to choose defend or abandon as they would in a normal game. Presumably white can keep her komi.

p.s. And does anyone have any name suggestions for the varient, if it works? Loser's Go is just a play on what I called Loser's Crib when I used to play it.

[Polama]

Twisting my melon man. You know these ideas are twisting my melon man.

lol

I didn't know it was a rule to fill in dame in area scoring; I thought it was simply a desirable action to take with area scoring. If you don't have to fill in dame then there's no reason to in Loser's Go right?

Good point. It's not actually necessary. Day saved =)

If we were to play, I would need a handicap in some form. That really is a tricky situation. How about, black (9k) plays several moves (let's say 3 as he is playing a 6k) and for white's first three moves, black chooses where white will play I love the sound of the Choice Go variation linked in the link Bill Spright shared at the beginning of the thread. Maybe the weaker player can choose the other player's moves (equal to his handicap?)?

edit: Thinking about this again, maybe the handicap could just work as in a normal game. Even if it's 9 stones, that simply means Black has 9 places to choose defend or abandon as they would in a normal game. Presumably white can keep her komi.

Yeah, I think you're right that handicap stones are still beneficial. Which is really counter-intuitive given that you're trying to lose, but it does seem that more stones lets you lose more precisely.

[mbv]

Yeah, I think you're right that handicap stones are still beneficial. Which is really counter-intuitive given that you're trying to lose, but it does seem that more stones lets you lose more precisely.

It's like learning to walk; a controlled fall