I am following with interest the game, and I’m curious about how it will end.
Then I got ideas for a variant, kind of king of the hill, maybe could make the game more dynamic?
The goal could be to control the tengen for a few rounds, meaning to have the most stones on the 9 intersections around for X turns (3 turns?).
It could create a new and interesting way of playing, keeping intact the rules already established besides the goal.
I think that king of the hill concept is a neat idea to explore.
A variant similar to your idea could be to only have controlling tengen and other hoshi count for points. Thus, the star points are like the “supply centers” that the players would contest. This would lead to a much more quantized score than area scoring, but it would be very clear who is ahead.
Another consideration is whether the victory objective (for area scoring) should be to achieve a plurality in points (as it is in this game), or if a strict majority should be required to win. Switching the objective to a majority condition would make winning harder, since one would need not only the most, but at least half of the board (41 area points on 9x9), and if no one achieves this score, the game is declared a draw among all non-eliminated players. However, maybe this would cause endgames to become a bit too bland with a lot of draws.
This looks like to go back to nearby the ancient rules (Tibetan go), funny.
I dunno I was hoping to keep simple and mostly hoping that corners will keep some importance enough (like providing life and assistance) but now I’m not so sure 
I had another idea for how to set up the player incentives, building on this very natural idea by @martin3141 :
I really like the simplicity of this, and would prefer to combine it with stone scoring. For instance, when every (non-eliminated) player passes, the game is over and each player is given 1$ for each stone they have on the board.
If the result of filling out remaining territories is obvious in advance, a player can propose a certain distribution of payouts, and if every non-eliminated player as well as the referee agrees the game can be ended without playing it out.
I think this does for a good game in itself, quite similar to what we’re doing in the first game but in my opinion a bit simpler to explain. I also think it’s better that a player in second place on points can choose to be content with their points, rather than having second place be equivalent to losing. @yebellz seemed to want to avoid this in the original rules:
so this might be up for debate. Personally I don’t see any problems with allowing this. The player in second place still wants to maximize their own points as much as possible.
But let’s now take this a step further to see if we can improve this to incentivize players which expect that they will not surive until the end of the game. What if, in addition to the payout at the end of the game, we made such a payout every time a player was eliminated? Specifically:
If N players are eliminated in a round, all the remaining players get N$ for each stone they have on the board.
This makes it so that, even if you see no way of your stones staying alive until the end of the game, you should try to stay alive until some other player is eliminated. It is directly in your interest to help capture some other players big group, even though this might not open up a chance for you to live, since this makes it more likely that they will be eliminated before you.
Probably the last payout (when everyone passes) should be K$ per stone, where K is the number of remaining players in the game. Then the amount of money paid out during the game is bounded by the number of players times the number of intersections on the board.
Of course I’m not advocating playing for real money, this is just an easy way of talking about incentives. If you want to think about where the money comes from, I think a practical solution is that each player pays 81$ (or whatever the board size is) up front, and then the referee takes what’s left at the end of the game. But again, the money is imaginary.
What do you all think about this idea, any obvious flaws?
I think these are interesting ideas, and I definitely want us to explore these questions further, especially for potential adjustments in future games.
I would like to comment with more detail about incentives and strategic options, but I don’t want to personally influence the decision making in the current game. So, I’m purposely holding back on some thoughts for until the game is over.
Instead, I would like to mainly focus on some general game theoretic discussion and comment on the proposed variations which are quite different from the current game.
However, first let me just remark on this:
That specific comment quoted from me in the rules is more of a strategic clarification/interpretation. Ultimately, however, the players can try to shape the outcome of the game however they wish, and their personal preferences might be more fine grained (and based on many things that transpire during the game) than the simple distinctions between win, loss, and draw as defined in the rules. I do not intend to enforce any sort of prohibition against “playing for a strong second” or “throwing a game”, and such a thing cannot be objectively done anyways.
The concept of playing for as large of a score as possible vs playing for the largest score significantly changes the nature of the game. Analogously, standard Go is also drastically changed by adjusting the objective from maximizing probability of victory to maximizing score. Of course, human players do ultimately seem to have some preference toward maximizing their score (in standard Go), but this is typically a secondary and unofficial objective. If one actually changes the objective in standard Go, then considerations about risk and strategy are profoundly changed.
For example, with the standard Go rules, if one is uncertain about the security of a local position, but confident of having a substantial lead, then one might play a slack but safer move in order to feel more secure about preserving the lead while even ceding some of it. In another example, if one is trailing in expected score, standard Go incentivizes bold, even risky, plays that aim to take the lead, even if such moves might come with an increased risk of falling behind even further. Changing the objective of a two-player game to maximization of score greatly impacts the strategic considerations in situations like these examples.
Stone scoring introduces yet another wrinkle with the concept of group tax, which adds an incentive to create fewer and larger groups, and to fill in as many eyes as possible. However, with the multiplayer angle of diplomatic go, where life and death is affected by the possibility of multiple eyes being filled simultaneously, the consideration of how many eyes to leave in a group (which impacts stone scoring) is affected by diplomatic considerations. For example, a player with a large group with X single-point eyes might wish to fill in more of those eyes to gain more points, but has to consider whether if doing so changes the life and death status of that group, which ultimately depends on the other player’s willingness to attack a potentially vulnerable group. If each player’s own preference is to only maximize their own score (and not care about any other player’s score), then maybe there is no incentive to help try to kill a particular group (that is not adjacent to any of their own), and maybe such a hostile action might be diplomatically discouraged due to concerns of damaging retaliation.
Regarding payout design and general game theoretic considerations: A game might assign numerical payouts to various outcomes, but it is important to note that such payouts do not necessarily correspond to the utility functions that are implicitly defined by player preferences (even if those preferences generally correspond to the objective of maximizing one’s payout). The expected utility perspective of rational behavior requires knowledge of the utility function in order to understand/predict behavior, however, such behavior depends on the individual’s risk tolerance/aversion, which is not just a function of expected payout, but also depends on higher-order moments (like variance) and, generally, the entire distribution over the payouts (when dealing with the potential uncertainties at an intermediate stage of the game). A beautiful foundational result in game theory establishes that (under mild assumptions of player’s preferences being self-consistence and non-pathological) a utility function does indeed exists such that the maximization of expected utility (with respect to this function) captures all of the nuances (including risk) inherent to a player’s preferences. However, explicitly determining this utility function from a player’s individual preferences is not at all obvious, and generally impossible without a deep understanding of those preferences. Such preferences might also depend on various externalities and psychological factors that arise from interaction with other players as well. Thus, while a game can establish payouts for various outcomes, which will have an impact on player’s preferences and their implicit utility functions, it cannot hope to fully dictate those utility functions.
Agreed on each point 
Let me start by saying that the extra subtleties introduced by stone scoring is an absolute plus in my mind, but of course opinions may differ. I thought it was a good choice in conjunction with my proposal which essentially scores the game in the middle of the game, where area scoring is less meaningful. You could conceivably use stone scoring in the middle of the game and area scoring at the end, it’s just slightly less aesthetically appealing to me.
The objective of maximizing score indeed changes the game fundamentally, however it seems to me that maximizing score in Diplomatic Go is more well-behaved (with regards to players having clear objectives at all times, even when behind, which is important to make the game fun), and might even “feel” more similar to normal go than maximizing winning probability.
Of course players can always play however they want, in any game. When I sit down to play a regular game of go, no one is forcing me to try to win. But it definitely makes the game more fun if the player’s utility functions are aligned.
The payout model is merely an attempt at making the players agree as much as possible on what is a “rational” way of playing. In my mind, it makes it quite clear that I’m trying to maximize my expected payout (while assuming that the other players are doing the same). How to treat variance is still up to each player, but even with that uncertainty you would be able to predict alot about what the other players want.
This is my main focus with this proposal: I want the players to as much as possible be able to reason “Player A shouldn’t want to do X, because then player B would want to do Y, which wouldn’t be advantageous to A” etc.
It’s ok if there’s not always a single theoretically correct way for each player to play (this is almost definitely impossible to achieve), but it’s boring if multiple players in the game have no preferences where to play at all (both boring for them, since they don’t care about the game, and boring for the opponents, since their impact on the game can’t be reasoned about in a good way).
This is why I value so highly to give each remaining player a clear achievable goal to play for.
I agree with both of you (although I didn’t understand everything @yebellz wrote, I could tell it was smart), and would like to add some observations I’ve made. First I want to say that the current rules are very well thought-out in my opinion. I personally have the bad habit to change the rules of board games even before I tried the original even once. So I definitely don’t mean to criticize anybody when suggesting a rules tweak.
Player incentives for multiplayer games is a difficult topic in my opinion. I believe most multiplayer games define winning criteria or how to determine the winner, and oftentimes give the players additional incentives and goals to work towards. I fear that with the status quo (where the one goal is to control the most points), players who are close to eliminated might feel discouraged to keep playing. At the same time, from the perspective of other players, any non-eliminated player’s decision is very important, because they can still greatly shape the future of the game. Playing on a less crowded board (for example 5 players on 13x13) could help in this regard too, because avoiding elimination is probably easier.
Discussion in diplomatic multiplayer games are delicate and tend to lead to interpersonal tensions that some people might feel uncomfortable with. I understand this is part of the game, and probably one just needs a thick hide for such games (especially since backstabbing is supposed to happen). But maybe additional incentives could eleviate some of that. It’s easy to get frustrated when someone attacks / betrays you seemingly without reason, so understanding why they made this decision might help avoid this. And, like le_4TC said, it’s more fun for players that are behind, if there’s still achieveable goals that you can work towards. At least I think so.
Lastly I would like to throw another rules-idea at you:
Whenever a chain of stones gets captured, all players who have adjacent chains are rewarded prisoners equal to the number of captured stones, each being worth 1 point (This would perhaps imply that the game is scored by territory at the end, rather than area scoring). This way players who expect to not be able to secure territory for themselves could still try to capture stones to earn points. At the end of the game all players receive a rank according to their points, and the goal is to achieve as high a rank as possible.
I believe giving points for capturing stones is a bit problematic. For instance, imagine that there is a simple ko shape between two players who are behind on points. They could keep trying to take this ko back and forth at the end of the game, both gaining points. This is not a fool-proof plan to get them in the lead, they have to avoid repetition, and other players can interfere in various ways. But it still seems like a weird thing to give out points for moves which capture back and forth without progressing the game.
Edit: More generally, players could help each other by throwing stones inside eyes. You could basically spend your move to give one point to another player, they don’t even have to spend a move to capture it if it’s a 1-point eye. With enough eyes there’s no risk of accidentally capturing something, and also it would be unfeasible for other players to block these throw-ins. They would have to guess which eye you’re going to throw into, and if they guess wrong they’re just helping giving more points to that player. So this is a very realistic way of generating free points for another player. A pair of players could agree to do this for each other indefinitely, and all they’re risking is that the other player gets one more free point than them, since they can stop the move after the other player stops.
Right, I did not think this through. Maybe this could be fixed by equally decreasing the points of the player whose stones get captured, but that doesn’t feel right to me either.
@martin3141 I did suggest territory scoring earlier Diplomatic Go - #106 by shinuito
I still think you can just prevent/ban (intentional) suicides. I suppose it’s more complicated with simultaneous placement that maybe you could unintentionally suicide a stone or group of stones, but requiring one new stone (of a different colour) be added to fill the last liberty of a group to capture it might cover this. As in, in suiciding a group you fill your own last liberty, but you could play a move to fill one of your liberties provided it isn’t your last one, which could of course help someone else capture it during the same turn.
In theory we could also argue that japanese rules themselves are silly because you can start a triple ko and the game could end up annulled, in which case what was the point in the 0.5 in the komi.
You could also see view this less as a problem and more as strategy to win near the end of the game. I think at least one test game would figure out if this is a problem and how preventable it is.
So a silly mockup example then. Red is winning with 10 points, Black has 7 points and White has 6 points. Suppose there’s no captures in the game for some reason (or maybe we can simply amend the shapes so that everyone has an equal number of stones.)
The problem is that in theory white and black gain points from the ko. So what is the sequence that ends this game for example?
Black needs to allow a white move at A5, but Red will not allow that since presumably they understand this too. So Red and white will always collide at A5 and Black will never fill it?
I think of area scoring and territory scoring as just complementary ways of counting almost the same thing.
Even in standard Go, both systems count territory and then you either add your living stones or subtract your dead stones. If both players play the same number of stones, the two counting system are completely complementary (except for the rare edge cases involving sekis, kos, and life/death complexities), hence often they either agree or differ by just one point, and they don’t substantially alter the general strategy of the game.
So what happens if someone those end up ordering a suicide? Their stone play is cancelled?
I think using territory scoring just adds so much more additional complexities.
It’s not just kos, but general intentional feeding of captures to an opponent that’s a problem.
However, I think the biggest problem with territory scoring is how to resolve life and death disputes? How do you decide which stones are alive and which are dead at the end of the game? Telling players to “play on” and capture all dead stones will distort the score.
In japanese rules, I believe this is solved by the players playing on during a separate phase in which the only valid ko-threat is a pass, and in order not to distort the score, playing a pass costs one point. Is this correct? Unfortunately I don’t know how this could be implemented into this variant though.
If suicides are allowed, black and white would just play into each other’s eyes. Disallowing suicide introduces multiple new complixites, but let’s suppose we manage to do that.
Then the problem with the ko shape is that if black and white knows that red is going to try to block it, then black could play there to collide first, and then white gets it as a second choice. So then red shouldn’t play there, forcing back to fill… Basically it comes down to a 50-50 prediction of what the other players will do. It’s not gamebreaking, since black and white will not win that coin flip more than a few times in practice. But isn’t it annoying that the game comes down to such a coin flip, when it wouldn’t have in area scoring?
And the ko shape was just the simplest example I could think of. With larger eyes you would also have extra points from playing into each other’s eyes (interestingly a 2-point eye is not big enough since one throw in and one stone to capture cancel each other out with territory scoring, perhaps this was intentional in your example), and I’m sure more pathological examples could be constructed, especially with more players on the board.
Edit: For instance, let’s say black and white have two or three such ko shapes between them. Red can’t actually remove any of them (even after being allowed to play in one of them, the black/white player can take on the next move and continue the ko), and black and white can choose randomly in which one to play. Every now and then black or white might have to insert some random move elsewhere to avoid repetition, but I think they could get a lot of extra points this way (whether they can get an arbitrary amount of points would depend on whether passes lift ko bans and other subtleties). Basically it’s very messy and not in a fun way, in my opinion.
In Japanese rules, the hypothetical play to determine life and death is only hypothetical and used only to determine life and death, but then the board is restored to the final position (after the initial passing) for the purposes of counting.
However, there also a lot of other complexities in figuring out life and death, dealing with kos, sekis, etc. I really don’t think the Japanese rules can be easily adapted to multiplayer, simultaneous action games.
This concept is called “pass stones” and is not used in the Japanese rules. It’s actually a feature of the AGA rules, which is an area scoring rules set. Pass stones are effectively a bookkeeping measure that allows one to get the area score via territory counting mechanics.
One could similarly adopt pass stones into Diplomatic Go, by counting each pass (or move that fails due to three collisions) as a negative point, and also counting captured stones as negative points for the player that was captured, but this is just effectively the same thing as the area score.
On a tangent from the rules discussion, here are some thoughts just about logistics
It was mentioned earlier that playing this variant game in a live session (or sessions) could be interesting. I think this could be done quite effectively using some modern streaming and chat tools, like Twitch and Discord.
I think a non-playing arbiter would still be needed, but they could set up a Twitch stream to broadcast a live feed of the current board state, countdown to each deadline, and the movement adjudication at the end of each round. The discussion could happen via Discord among the players, each also having a private channel with the arbiter to submit moves.
Basically, although we’ve been playing a correspondence game via these forums, and it’s been very effective for that purpose, a live game could be facilitated using very different technologies, omitting the forums altogether.
Unfortunately, although it would be awesome to do something like this live, it would be very difficult for me to set aside a solid chuck of time to participate in something like this.
I’ll throw out another scoring idea I just had (I’m not saying it’s necessarily better than previous ideas, it’s just another option to discuss).
Instead of submitting a move, a player may choose to drop out of the game. Their final score is their area1 on the board at that time2. Naturally, once a player has dropped out they are no longer allowed to participate in any way, but their stones remain on the board until they are captured.
1 Tromp-Taylor-style scoring, where all stones on the board assumed alive.
2 Before the other player’s moves from that round are made.
Less important sidenote: The primary goal is to maximize your own score. That being equal, we should define whether it’s preferable to improve your own “placement” in score, or your score as a proportion of the total score of all players, or something else. I have no strong opinion here, but I think we should make some secondary goal explicit in the rules.
That’s it! There’s no need for a separate scoring phase after passing. At the end of the game, each player simply “drops out” when there is no way of improving their own score.
The main way this affects the strategy is that if you think a big group of yours is about to be captured, you should drop out now, unless you will be able to make up that territory later. I suspect that not everyone will like this aspect, but I think it could improve the player experience for everyone.
It’s probably more fun to “cash out” last minute with a decent score, than to stay in the game with a few stray stones and no clear goal. And for the remaining players, it’s no fun to have an unpredictable participant in the game without clear motivations, so it’s nice to instead get confirmation that a player has dropped out.
What do you all think about this? This diverges from normal go a little bit more than earlier proposals, but if we stay openminded about how the game “should” behave I believe this could actually play out very nicely in practice.
Edited to add: To me the weirdest part about this proposed rule is “premature” drop-outs in cases where you actually could have improved your “real” score on the final board, but you can drop out for more points. For instance, your group is completely alive, you can still gain 2 points extra in small endgame, but there are 3 of your stones about to be captured on the next move. Here you drop out to get 3 points for those dead stones, instead of making 2 more “real” points.
We could accept this as part of the game, but we could also drastically reduce the occurence of such situations by giving more points to stones alive at the absolute end of the game. There are multiple way to implement this, either by reintroducing a normal scoring phase (worth double points, but only for players who haven’t dropped out earlier), or maybe by adding a rule that if none of your stones are captured after you drop out, your score is doubled. Something like that.
I think that’s a very interesting proposal to consider.
The timing of cashing out could be very interesting. A player with a bunch of strategically dead stones could cash out earlier to lock in some points, even though they might likely be completely eliminated in subsequent play. Further, upon cashing out, it’s not immediately clear what proportion of the total points one would ultimately get, since some stones that one gets cash for would eventually be captured (allowing others to get credit for that area as well), while other stones might remain on the board preventing others from getting credit for those. The potential immortality of remaining stones would depend on how many other player are eliminated or cash out, so ultimately, when deciding to cash out, one has certainty about how many points they will get, but probably be uncertain about the size of the overall pie.
I think how players might perceive the value of the payout is vitally important (and definitely not just a sidenote) in understanding how they might behave under such a system.
Is the absolute maximization of score most important? Or should a player care about their own score relative to the total score of everyone? Should players care about what order they place relative to other players?
For example, we could use the hypothetical distribution of prize money in various imaginary scenarios to understand this scoring mechanism.
There is a fixed pot of prize money, say $1000. After the game is over, each player is awarded a fraction of that pot equal to their score divided by the total of everyone’s scores.
There is a flexible pot of prize money, with the total money distributed being variable depending on the actual scoring. After the game is over, each player is awarded $10 per point of score that they have. This type of game is definitely not zero-sum, since all of the players could hypothetically cooperate to maximize the total payout, by taking turns covering nearly all of the board, before cashing out to let another capture them and take their turn to cover nearly all of the board.
There is a fixed pot that is entirely distributed to the player with the highest score (if it is unique), or split evenly among those with the highest score (if there is a tie).
There is a fixed pot that is that is distributed according to some predetermined schedule that only depends on the relative rankings of the players based on their score. For example, first place gets $500, second place gets $250, etc. (and maybe with ties averaging out the payouts across the placements).
Yes, “less important sidenote” was the wrong choice of words. It was more the fact that my main idea was the “cashing out”-part, and I didn’t have a specific proposal in mind yet for how to define the player incentives.
I still really like the idea of territory scoring. Even a player that will not survive could still take part in a capture and gain points from it.
I think territory scoring and probably counting eyes in seki (contrary to japanese rules) would probably be ok.
There was the ko problem that there could be a strategy to gain points.
It’d be nice to have a perfect ruleset with no unwanted strategies, but the cashout game above
feels like it incentivises me to just make a big dead dragon and then drop out. It’s probably fairly nice not to have to worry about life and death problems and just play as many stones as you can and drop out before you get captured 
