Disclaimer
This post contains a ton of discussion about the impact the Tribute scoring system has had on the Windy City Weasel league over the past two months. Even among avid Diplomacy players, only a small amount of people will care about this topic. Here’s the short version. If you want a good score in our league (or really within most scoring systems), get 18 centers (solo). If you can’t solo, get more centers than your opponents (top the board). You don’t need to read the rest of the post unless you want to dig into the minutia.
Advice from someone who isn't even close to first place
If you are still reading this post, you are probably either playing in a league with its own scoring system or playing in or running tournaments. For this very small group of players, this discussion can be very fun! Tribute is a new scoring system. We’re the first club to adopt it and have only been playing it for two months. As a club, we are all still trying to figure out the best way to maximize our overall league scores within the system. We’ve played enough games that people are starting to develop opinions, but haven’t played enough that there is anything close to consensus. It’s an exciting time!
I think it’s also worth pointing out that league (and tournament) Diplomacy are variants. As soon as you bring in external standings, you are no longer trying to maximize your score within a single game, you’re trying to maximize your score within that league’s standings. This means that sometimes league players will make decisions based on how it impacts their standings instead of how it impacts the board they are currently playing. League players need to be mindful that non-league players may not care about the league standings, and some league players could care less about their score. Figuring out the motivations of your fellow players is a huge part of succeeding at this game.
As a club, we’ve had sporadic discussion about Tribute scoring after the past few games but the post game analysis of game 405 created a *ton* of spirited chatter about this topic. To me, the essence of all of that post game discussion can be distilled down to this: If you can’t solo or top the board, what is the best way to maximize your score within the league? I felt like there was enough engagement on this topic that it deserved its own post.
Sometimes it’s better to show than tell. Here’s the moment that sparked all of the discussion. In F1904 , Austria, England, Germany, and Italy are at 6. England and Germany have just stabbed their allies and will be at 8 centers, building two units with position to gain more centers in 1905. This is a timed game and 1905 will be the last turn.
Pretend you are France. Italy approaches you with an offer. There is some disagreement about what specifically was offered in game 405. For the purpose of this post, we’re going to explore a few hypothetical options.
Scenario #1: Give me your dots to help force a tie
Hey France, I screwed up. Austria is building two units and I don't have anything near my home centers. If we don't do something, he's going to have a huge score. Let's take chance out of this. Give me all of your dots I can take so we can try and force a tie.
In this scenario, you believe Italy is telling the truth. You decide to give up on Iberia and give it to Italy so he can build units to defend his home centers and attack Austria. Your theory is that Austria and Italy fighting means it is less likely for either of them to get a huge board top and might result in a tie.
In this offer, Italy takes Marseilles and has position to take Spain and Portugal, but these cannot be forced. France can choose whether or not Italy gets in to one or both of the Iberian centers. Assuming France follows through with the agreement, Italy is guaranteed at least nine centers from Iberia, can defend his home centers, and depending on how Austria and Turkey choose to defend, has a shot at picking up one or two Austrian centers. For this example, I’m assuming Italy can’t get anything else off Austria and Austria and Turkey don’t get anything off Italy.
The end game scores probably ends up looking something like this:
There is a very real chance that Italy manages to take an additional center and wins. There is also a real chance that Austria takes an additional center off Russia and creates a 3 way tie for the board top. France could also have blocked Italy from getting into one or both of the Iberian centers if they wanted to ensure Italy did not top which could have resulted in an English board top. I went with the spirit of the agreement which was France giving their dots to the Italian as part of a stop the leader alliance.
- French Points: 8.4
- Score difference between tied board toppers and France if there is a tie: 17.5 points.
- Score difference between Italy and France if there is an Italian board top: 35 points.
Scenario #2: Let me take Marseilles and I'll go fight Austria
Hey France, I screwed up. Austria is building two units and I don't have anything near my home centers. If we don't do something, he's going to have a huge score. Let's take chance out of this. Let me take Marseilles so I can protect my home centers and I'll leave you alone.
In this scenario, you believe Italy is telling the truth. You decide to give up on Marseilles and give it to Italy so he can build an army to defend his home centers and attack Austria. Your theory is that Austria and Italy fighting means it is less likely for either of them to get a huge board top and might result in a tie.
In this offer, Italy takes Marseilles but realizing that they can’t force the rest of the Iberian centers, decides to attack his biggest rival for the board top, Austria. Assuming France follows through with the agreement, Italy is guaranteed at least eight centers.
The end game scores probably ends up looking something like this:
Turkey is the big wildcard in this scenario, particularly if they work with Italy. Italy and Turkey working together could easily force a tied board top. For this example I am assuming a neutral Turkey who chooses to protect himself rather than help Italy.
- French Points: 10.4
- Score difference between England and France if there is an English board top: 26 points.
- Score difference between tied board toppers and France if there is a tie: 15.5 points.
Scenario #3: Give me your dots or else I throw the game!
Hey France, you can’t top but I have a chance at topping or tying. Give me your Iberian dots or England or Austria will win this. If you don't do what I ask, I'm going to give all my dots to Austria to punish you.
I’m not exactly sure how the diplomacy went down in game 405 (and probably differs depending on who you ask) but I’ve played with both the French and Italian players enough to believe it went something like the above. The actual conversation isn’t the focus of this article. Instead I want to explore the impact of not forming the “stop the leader” alliance.
In this scenario, you don’t believe Italy will stop attacking you and you believe he will give his centers to Austria if you don’t do what he wants. You decide to make sure Italy doesn’t get your centers because his request is crazy. Screw that guy!
In this scenario, Italy cannot nor wants to defend against an Austrian attack. No other powers on the board can keep up with or slow Austrian growth. Italy will not interfere with the inevitable Austrian win.
The actual end game scores ended up looking like this:
The Austrian score could have been even larger, possibly getting as high as 13 centers. Italian dot throwing didn’t actually make much of a difference. It was really just a single center. However, Italy not attacking Austria freed his units up to do other things which greatly changed the game.
- French Points: 9.4
- Score difference between Austria and France with Austrian board top: 40 points.
Analysis
Comparing the scenarios, here’s what we learned.
- Scenario #1:
- Point Difference: 17.5 – 35
- Probability of split board top: highest.
- Probability of an Italian board top: highest.
- Scenario #2:
- Point Difference: 15.5 – 26 points
- Probability of a split board top: medium.
- Probability of an Italian board top: medium.
- Scenario #3:
- Point Difference: 40 points.
- Probability of a split board top: lowest.
- Probability of an Italian board top: lowest.
First, it’s important to re-state that the analysis is different for players trying to maximize their score in the league than those trying to maximize their score in an individual game. Excluding personality conflicts, it is my opinion that league players who cannot win should try and keep the difference between their own score and the person who board tops as low as possible.
Other club members may disagree. There is a strong argument to be made that punishing another player who is ahead of you in the league to prevent them from getting a higher score is the stronger choice. I’d love to hear the thoughts of other players about this.
It is my opinion that league players who cannot win should try and keep the difference between their own score and the person who board tops as low as possible.
If your goal is to minimize the difference between you and the player who board tops, it’s clear that scenario #3 is not the best choice. Instead, I believe the Tribute scoring system strongly encourages the powers who don’t think they will be able to board top to form a “stop the leader alliance” to force a draw with either a split board top, or to keep the leader’s score as low as possible.
So what would this hypothetical “stop the leader alliance” look like? This is where it gets interesting. In this specific scenario, the highest chance for a split board top was definitely in scenario #1. France vacating the Iberian centers and allowing Italy to take them gives the alliance the most control over the outcome of the game. What is interesting is that this scenario also gave Italy the highest chance of topping the board himself so if you didn’t want a fellow member of your alliance to top, this probably isn’t the best choice.
I’d also like to point out that scenario #1 is also different than emotional dot throwing to “teach someone a lesson.” These are calculated exchanges of centers as part of an alliance to minimize the damage for all members of the “stop the leader alliance.”
Knowing the players involved in game #405, personally I would have been far more likely to go with scenario #2. France giving up on a single center that is difficult to hold to help prevent another power from walking into another player’s home center for a huge board top is absolutely the right choice for a “stop the leader alliance.” France only loses one point in exchange for potentially cutting the leader’s score in half or more!
Scenario #2 also doesn’t involve giving up as many centers voluntarily which just feels better.
Because of Tribute's huge bonus for topping the board, in most cases keeping the leader's score as low as possible has more impact on the league score than taking or keeping centers for yourself.
Conclusion
What I’ve concluded based on the above analysis is that there is very clear incentive within the Tribute scoring system for league players to form a “Stop the Leader” alliance and keep the board topper’s score as low as possible. Because of Tribute’s huge bonus for topping the board, in most cases keeping the leader’s score as low as possible has more impact on the league score than taking or keeping centers for yourself.
It’s far better for the “little guys” to coordinate than attack each other. The mechanics of how the alliance chooses to go about executing its plan is where things get interesting. Is it better to give dots to one single player to try to force a tied board top and risk that this player ends up topping the board themselves? Is it better for the alliance to eliminate some of its members to increase the survivor’s survival bonus? How much can you trust your allies to not stab for the board top themselves? How early should you begin exchanging centers to stop the leader?
I’m looking forward to discovering the answers to these questions with my fellow players over the next two months of playing within the Tribute system and beyond. The reality is that we are all human and will have different motivations within every game. Figuring out the answers to questions like these are what keeps the game interesting and keeps me coming back to play again!
This Post Has 5 Comments
After the game noted in the article, Christian had an argument about “sunk cost” that made me think. If I remember correctly he was saying if one person boardtops with a huge score it will keep everyone else down and consequently it benefits one in the league.
My first thought is this may make sense if we only played a few games in the league and having many people with very small scores may be beneficial, while we have a select few with crazy high scores.
However, I think in a league like ours that many games are played if every each game we crown a king with such a high score we will end up with MANY high score players not a SELECT FEW. This will make it harder to get to the top and won’t help the survivors of a game who are chasing the ones atop.
Additionally, I think for good players who can score high, having a big pool of average scores (as opposed to very low) won’t matter, even if there are so many of them. They will be making it to the top eventually.
Hence I think the “sunk cost” argument does not make sense in our league.
In my experience, Tribute scoring system has resulted in something new: a clear winner while it rids 2nd and 3rd place of relevance. It has created a seriously high incentive for wanting to boardtop and “win”. I remember with Sum of Squares scoring the boardtopper still got a better reward than the most sprinkled on top with some nice complements and congratulations. Now all survivors get seriously punished for letting someone else win. Anyone who’s not paving a way for themselves to boardtop is basically not playing the game.
I think it’s fair to say the new system has effectively made the motivations to play more multidimensional (“growth, boardtop, dominance, balance and survival”).
Maybe I’m missing something, but if the goal is to try to keep the highest score on the board as low as possible, would scenario #1 be the right agreement as France?
— There’s way more likelihood that the top is split 2 or even 3 ways, so the board toppers would have their individual scores very reduced.
— Even if Italy ends up topping, the center count with which they do is probably the lowest of any scenario (fine, #1 and #2 are close there).
— The risk of you being eliminated is low to non-existent in either case, so the difference in your end score for the game and the impact on your league standings is super marginal.
It depends on how you evaluate the relative likelihoods, and how much you think you can put your fingers on the scales to ensure a tie or not. But #1 has a lot to go for it. Why isn’t France doing that?
Also, maybe unrelated to this scenario (I hope at least!), I think that there is a potential reason to not want shared board tops, especially if you’re not one: shared board tops can increase more people’s league standings. For example, suppose three people are competing for a board top. Two of them you know are very good players, and will likely put up very good scores in other games, which means chances are for one of them that their shared board top this game could end up being their third highest score for the year, or even in the ‘average’ pile. That’s a rather good score to pull up their averages/offset any games in which they were eliminated. Meanwhile, the third player you think is getting lucky this game, and is unlikely to replicate this success: you think you will end up with more board tops than them in the end, and so they aren’t a real threat to your place in the standings even if they win this one.
That’s a potentially good reason to throw some dots to the 3rd player and see them through to the top, no? Especially if the shared top is not guaranteed and one of the “better” players manages to squeak out a win?
Wes I agree with you that scenario #1 is the most ideal for the “stop the leader alliance.” Honestly, I was surprised at this conclusion. What I’m unsure about is if this is a “in this particular situation” sort of thing, or if this will prove to be the right thing for the future as well. The reason Scenario #1 is so good in this situation is that France can let Italy into Mar and then has the option to bounce him out of the rest of Iberia. It’s a pretty low risk play with the largest upside.
I’m still not 100% sure I do it as France though. It’s probably the fact that i’m not used to this scoring system yet, but walking out of supply centers just feels wrong to me. I’m far more likely to accept scenario #2 and then maybe I can flip Spain on the final turn or something if it looks like it’s going to be needed.
What I think I’ve learned here is that in Tribute, unless someone is winning really big like Ariel in game 403, if the “little guys” have their act together it should be hard for someone to top the board.
Wes, I love your thoughts about not raising the score of your competition. Things like this is why I’m not super worried about this getting metatgamed too much. You have so many different ways of looking at this that no one is every going to reach a consensus. Even among the active league players, depending on where they are in the standings their motivations may be different.
At the end of the day, this just boils down to figuring out your opponent’s motivation just like it always has. The scoring system and league play doesn’t change this, it just adds other motivating factors to keep in mind.