Interestingly, a typical practice in face-to-face (F2F) games is to allow players to go off and discuss in groups of any size, but I think most diplomacy websites (at least all that I’ve personally played on) restrict private communication to only one-on-one. In F2F, it’s anyways too hard to enforce anything different, since even when the group is all together, one can’t ensure that everything publicly said is heard by everyone. There has also been significant participation in postal correspondence diplomacy, where the practice of using snail-mail naturally limits the communication to one-on-one.
1 Like
This is very subjective (note that the concepts of winning/last place are equally arbitrary and imaginary as virtual prize money). For me personally I think a virtual prize would work fine, but of course it depends on what all the players think, so I’m interested to hear more opinions 
I like the idea of promoting alliances more, but I’m not so sure about the proposal as it stands, since it seems like it often might punish having made more territory. Like if 3 players make 28, 27, 26 points on the 9x9 board, the only loser will be the player who made the most points. That just doesn’t feel right to me. I would prefer the endgame to still be about securing as many points as possible for yourself, not having to worry about all the possible sums of different players scores and optimizing for having a “useful” score to combine with others to come as close as possible to 41 points. Maybe I’m not understanding the strategic implications correctly, but to me it seems this changes the game a bit too much from go
3 Likes
Personally, I’d love to give this a try!
2 Likes
I tend to disagree. Ranking higher or beating some opponent is in itself a reword, while winning slightly more / less but not managing to beat another player does not feel much like a reward at all, it’s all relative in the end.
Virtual prize money would work again if score is being summed over a lot of games, which again gives an opportunity for players to compare themselves to others.
Exactly: it’s much more about cooperation, not about selfishly trying to get the most territory. And there is still a benefit of being the strongest player, if you can make sure another player has a very small number of points: getting 28 points, and another player can be limited to have only 13 points, will mean you get a bigger share of the prize at the end, and the player with 13 points will accept to be your team mate, since it gives them the largest share.
It’s still about maximising your score, but not in total, just within your alliance. You have to balance getting the most territory with having the right friends.
There’s also additional strategic considerations, such as ditching one of the teammates because the alliance grows too strong, making it feasible to get more than half with one member fewer.
1 Like
I think @Vsotvep’s proposal is interesting, but it seems that those that are trailing in score may also wind up with not having a direct incentive to continue play, except to potentially influence the payouts to others.
It also seems odd that maximizing one own’s score could wind up being detrimental.
1 Like
Seems like on both points we have differing subjective opinions, maybe we’ll just have to try both and see how they play out 
Even then we might of course disagree on which sort of game we enjoy more… But that’s far in the future, let’s not worry about it now 
3 Likes
I think simpler concepts might be perceived as more meaningful. When a game only just declares winners, losers, and drawers, then the significance of those are very clear. Someone can always say that they “won” a game and feel good about that, or feel good about making their arch-rival “lose”, or at least console themselves with making it into the “draw”. However, the relative psychological impact of received X virtual dollars instead of Y virtual dollars is probably a lot less meaningful to many players.
2 Likes
I agree. We could keep the notion of a winner as the one who makes the most money, and players who are doing well are free to aim for “winning” if they want to. But it seems like most players agreed during the previous game that a more fine-grained score would make the game more fun while you’re “losing”.
3 Likes
With a less crowded board, most players will probably manage to live, making competition for second, third place etc. a valid goal as well.
Only having one player win seriously disrupts the fun I had near the end of last game. I honestly felt I had a chance at winning, but it was blocked by Haze and Martin, who had arbitrary other goals that had nothing to do with the actual ruleset. Granted, I could’ve played differently, but the thing that is annoying to me is that there is no way to reason with others if you’re playing for different goals.
If I myself was in a losing position, I don’t know what I would’ve done. Probably I’d have passed, as I don’t feel like creating artificial personal goals is much fun.
It’s a bit like those Go AI that aren’t allowed to resign, and then just start playing crazy moves, because winning becomes impossible.
1 Like
To be honest, when I play diplomacy and find myself “losing”, I find more fun in making those responsible for my loss also lose.
In Diplomacy, the game is a bit different in that a winner is declared only if the leader controls a strict majority of the board, whereas our scoring perspective in Diplomatic Go recognizes a win based on plurality. However, with the money-making objective, one might pathologically be in a situation where it is better to “lose” (not be the leader) in order to make more money than one might get while going down another path where they “win”.
2 Likes
I wouldn’t go so far as to declare that.
I think Haze’s end game play can be understood as one that had been reduced to a most likely losing position and then choosing to disrupt a rival’s (Vsotvep’s) plans to win. Similarly, I think Martin’s play was based on a strong perception that a draw was not feasible given his belief about the priorities of others. The ambiguity of other player’s preferences is a psychological part of the game. The game features both imperfect and incomplete information.
2 Likes
I understand what their goals were, that’s not my point.
My point is that those goals are arbitrary personal goals, not something set out in the rules, and not something that increases their chance to win / improve their position. They could’ve equally chosen to resign, to fill up their own liberties or to let a random generator play for them; it wouldn’t have changed their outcome of the game, only that of the other players.
I’m fine with people blocking my goals if that is in their interest, but when it is not, it’s just annoying to me. I can’t strategise about what the best move is if I have to account for arbitrary goals. And I can’t judge if somebody is playing well either.
It removes a huge part of the strategical aspect of the game.
1 Like
Could I play?
3 Likes
I believe @le_4TC has mentioned some excellent arguments in this post as to why it’s actually a good thing that people in a losing position are still allowed to influence the game however they want, even if it doesn’t increase their chances to avoid a loss. After thinking about it, I agree with le_4TC.
Generally I believe that for different people, different aspects make a game fun. My natural pessimistic nature tells me that the winning chances in a 5-player game are far lower than in a 1v1 game to begin with, so I would advise everyone to approach these games expecting to lose most of the time. Therefore a casual mindset (as opposed to a try-hard approach) is helpful to have fun and avoid frustration. At least that’s my opinion.
2 Likes
There is nothing artificial about assigning significance to the outcomes of others. The overall outcome of a game is the finishing result of all of the players. One can meaningfully prefer “I lose and my rival also loses” to “I lose and my rival does better than losing”.
In two player games, like Go, Chess, etc., the outcome of the opponent is complementary (in a one-to-one fashion) with your own outcome. You simply aim to win, which means that your opponent must lose, and if you lose, your opponent wins, and if you draw, you both draw. However, with 7 players, much more combinations are possible, and even though you lose, you still might care about seeing someone else lose or not.
Further, it’s a valid strategy to follow through on punishing your rivals, even if it might be perceived as actually carrying out a non-credible threat, since 1) one might care about how others place, 2) it can be seen as reinforcing a reputation (in the context of repeated games). Note: repeated games does not necessarily mean across two separate games (which is probably undesirable from a fairness and meta-gaming perspective), but the concept of tit-for-tat retaliation can also occur at a smaller scale within one game. For example, when I attack Bob, I should not expect him to leave that attack unaswered and focus on the best point-making plays elsewhere, but instead I should prepare for the possibility of retaliation, even if that might be viewed as suboptimal when his attention could be focused elsewhere.
The game is called “Diplomacy” for a reason. It’s not enough to just attack others and exert one’s will by force, one must also maintain their diplomatic standing with others, since even the losers have a role in choosing the winner.
3 Likes
Maybe I just don’t like diplomacy, then. I want a game to have a clear unambiguous goal for each player, so that I could reason about what they’ll most likely do. Players straying from that goal in order to achieve something else, have a bad strategy in my opinion.
When players don’t have any goal at some point in the game, but still have the possibility to play, it just ruins it for me, since I can’t judge how well they’re playing, or what they ought to play.
The beauty of Go, chess, and even things like poker, or Settlers of Catan, is that everybody always has a clear goal, and that you can expect players to behave a certain way. There are still many strategies, and we have no idea what the best one is. The current rules of Diplomatic Go are however incomplete, since losing players have no strategy; they have no goal; they can’t be reasoned with; they can’t be predicted.
This is also debatable. Punishment only makes sense if it is to improve the overall situation. If I lose anyways, there is no actual reason to take my rival with me, except if it is to make my strategy known over the course of multiple games.
Punishing a player just because they did not honour your agreement, is not inherently rational. Especially not if it goes against your actual goals.
4 Likes
On this point it seems like you and I agree strongly 
But I also recognize the validity of the opposite position. Like I explained in an earlier post, I’ve come to the belief that there is a trade-off between predictability and diplomacy. I (and it seems @Vsotvep too) value predictability over diplomacy, while others might value diplomacy over predictability. Of course we need some of both for the game called “Diplomatic Go”, it’s just a matter of how much of each.
Taking the “opposite position” to the extreme, we could play a game without any specified incentives at all. Each player can decide on their own goals. Maybe one is trying to get five-in-a-row, while another is trying to connect two sides of the boards, while still another is trying to make territory. This actually sounds like a fun sort of social experiment, and diplomacy would play an important part (you could collaborate with players whose goals are completely compatible with your own). But it’s not as fulfilling as most other games, since your not necessarily working “against” anybody, so there might not be much conflict. This is not an argument for any side, just a fun thought.
6 Likes
But if I know you will never do this, your power to make threats is greatly diminished. This makes you a weaker player. Thus a rational player must be able to follow up on threats. Whether they actually follow through is another question.
Personally, I think it becomes very weird if we don’t think in terms of repeated games. If we enforce some sort of rational play for an isolated game, making threats becomes completely impossible in most cases (every player should always play the best they can from the current position, without caring about the past). The only case where a threat could work would be if you are going to have an option between two, for you, equivalent outcomes.
So I think we should always think in terms of repeated games. Then following through on threats makes sense (but it’s still also possible to bluff and then not follow through, of course).
6 Likes
Exactly, that’s why it’s only a valid strategy when the punishment gives you actual power, either because your punishment can be made in the same game without getting you in a losing position, or because in a round of multiple games (or your history as a diplomatic go player) it is known that you follow up threats even when it is not beneficial to you.
I personally would not punish a player if the punishment itself brings me into a weaker position. But, again, we bump into the problem of having a goal: if I would be losing, and there is no competition for second place, it may not be clear whether punishment brings me into a weaker position or not: I’d have literally nothing to lose.
1 Like
The problem with not imagining repeated games is that you can’t prevent other players from doing so. If you play 100 games with other people, and you insist on always playing “rationally” in each isolated game, then you will just have a lower winrate than the other players who don’t limit themselves in this way.
If it could be enforced in some way, we could consider this a different way to play. But since it’s impossible to enforce, we can’t say in the rules “Don’t play in this way, which will allow you to win more often”.
(Note that there is no need to “build a reputation”. Making threats would work even in a huge pool of players, where no two players ever meet again after a single game. You could argue that there is then no reason to follow through on a threat, and thus ignore any threat. But then some of your opponents will follow through on those threats anyways, and once again you will end up with a lower winrate than players who instead judge the validity of each threat on a case-by-case basis, and sometimes take them seriously.)
6 Likes