Luck Plays No Part in Diplomacy

[RoganJosh]

Let me be more precise: Over the last 30 hours, you are the only one that has used the term "luck" in here.

[Squigs44]

The game was not designed to have coin flip optimal decisions and will never have coin flip optimal decisions. If you think diplomacy has luck in it because of coin flip optimal decisions, you misunderstand the game, or humans in general.

The game of diplomacy has randomness in its optimal strategy. In certain positions, your best choice is to flip a coin. That's not an opinion, that's simply how the game is designed.

And again, this is where we disagree. As Restitution just said:

An optimal strategy is fundamentally *not* the same as the "best strategy".

I believe that a coin flip is not your best choice. It might be "game theory optimal", but it is not your best choice. If you want to prove that your point is not an opinion, explain how Jmo's 75% success rate is inferior to flipping a coin - and work with real people, real situations, not some game theory perfection situation. I already admitted that in a game theory perfect situation, a coin flip is as optimal as you can get.

[RoganJosh]

I'm pleased to see that you have stopped objecting to "coin flip is optimal" and now only object to "in certain situations coin flip is best." It should be said, "best" and "optimal" are almost always different.

But, again, I have to disappoint. While it is very uncommon that they are the same, there are situations in which both "optimal" and "best" is to flip a coin.

The situation I described a while ago is such a situation. This situation happens when a pure 50/50 guess between two high level players, both of which are fully aware of the situation, will determine the game. In this case, you can not do better than flipping a coin.

Maybe you don't play by Optimal Strategy. Maybe you have a preference for Berlin. And you are playing jmo. And he is so smart that he is gonna try to exploit this.

That means that jmo is not playing by Optimal Strategy. Which you can exploit, by playing Munich against him.

But that means that you are not playing by Optimal Strategy. Which jmo can exploit by predicting that you are gonna try to exploit him.

But that means that jmo is not playing by Optimal Strategy. Which you can exploit by predicting that he is gonna try to exploit that you are gonna try to exploit him.

The chain doesn't end. And if both you and your opponent are competent enough to see the whole chain, well... Then there really isn't any way of exploiting your opponent. And the only way you can make sure that you are not being exploited is to randomize your choice.

[Squigs44]

I am going to work on two axioms, and if you disagree with either axiom, okay, then we disagree.

Axiom 1: No two players are of exactly the same skill. If two players were told to outguess the other (restricted to not use a coin, etc), one of the two would have a higher than 50% success rate (even if that rate is 50.0000000001%). We will call this success rate X.
Axiom 2: For each opponent, there is a nonzero chance that they will try to outguess me rather than flip a coin (even if that chance is 0.00000001%). We will call this chance Y.

If I am the better player, and I flip a coin, then my success rate is 50%.
If I am the better player, and I try to out guess my opponent, my success rate is (50% * (1-Y)) + (X*Y) = 50% + (X-50%)*Y > 50%.
Thus, given any two players, one of the two has a strategy that has a better success rate than flipping a coin.

Even if either axiom breaks down, then the success rate of trying to outguess my opponent is equal to my success rate of flipping a coin, and so in no situation is the coin flip a better decision for both players. It can be an equally good decision, or better for one and worse for the other, but it will never be a better solution for both players.

[jmo1121109]

I'm pleased to see that you have stopped objecting to "coin flip is optimal" and now only object to "in certain situations coin flip is best." It should be said, "best" and "optimal" are almost always different.

But, again, I have to disappoint. While it is very uncommon that they are the same, there are situations in which both "optimal" and "best" is to flip a coin.

The situation I described a while ago is such a situation. This situation happens when a pure 50/50 guess between two high level players, both of which are fully aware of the situation, will determine the game. In this case, you can not do better than flipping a coin.

Maybe you don't play by Optimal Strategy. Maybe you have a preference for Berlin. And you are playing jmo. And he is so smart that he is gonna try to exploit this.

That means that jmo is not playing by Optimal Strategy. Which you can exploit, by playing Munich against him.

But that means that you are not playing by Optimal Strategy. Which jmo can exploit by predicting that you are gonna try to exploit him.

But that means that jmo is not playing by Optimal Strategy. Which you can exploit by predicting that he is gonna try to exploit that you are gonna try to exploit him.

The chain doesn't end. And if both you and your opponent are competent enough to see the whole chain, well... Then there really isn't any way of exploiting your opponent. And the only way you can make sure that you are not being exploited is to randomize your choice.

2 good players will never end up in this situation. Please stop making up non realistic diplomatic scenarios to prove some game theory point. It's actually harmful for any SoW students to see people arguing that flipping a damn coin is their best option in scenarios in games. I have already explained the multiple other factors in a game that can and should be contributing to order decision making. There are so many factors that nearly nobody sees the entire chain of options and factors that can influence it. Calling it a single factor determining situation is just ignorant.

[jmo1121109]

I explained in detail why the 1 actual game example provided was not a 50/50. Does anyone have any others for me to correct?

[jmo1121109]

I think this is the main point of disagreement. I don't think randomness is the optimal strategy, I think outguessing your opponent can yield better results. Jmo's 75% success rate at alledged 50/50s is a case study of this point.

If Jmo's opponents would have been smart enough to use a random number generator, then Jmo's success rate at 50/50s would have been 50%.

If jmo only played non anonymous games this might make any sort of sense.

[jmo1121109]

Let me be more precise: Over the last 30 hours, you are the only one that has used the term "luck" in here.

Errr, you are aware that the entire conversation is about luck in diplomacy...per the OP and the articles? If that's not what you're discussing then maybe start a different topic?

[swordsman3003]

So Pterodactyls are dinosaurs and a tomato is a vegetable and RPS is random.
[. . . ]
https://en.wikipedia.org/wiki/Linguistic_prescription

I am well-aware of criticisms of linguistic prescription. That concept is so far-afield from our conversation that I am not sure if you are aware of what the phrase means (or if you even read the article you linked).

The Wikipedia article you linked is about style guides, non-standard dialects, split infinitives, preposition stranding, and grammatical concerns like that. Linguistic prescription -- and the challenges to it -- are not about how words correspond to concepts (that is, definitions of words).

Let me quote back to you the article you asked me to read:

A frequent criticism is that prescription has a tendency to favour the language of one particular region or social class over others, and thus militates against linguistic diversity.

A second issue with prescriptivism is that it tends to explicitly devalue non-standard dialects.

Another serious issue with prescription is that prescriptive rules quickly become entrenched and it is difficult to change them when the language changes.

Although prescribing authorities may have clear ideas about why they make a particular choice, and their choices are seldom entirely arbitrary, there exists no linguistically sustainable metric for ascertaining which forms of language should be considered standard or otherwise preferable.

Finally, there is the problem of inappropriate dogmatism. [. . .] A classic example from 18th-century England is Robert Lowth's tentative suggestion that preposition stranding in relative clauses sounds colloquial.

What does this have to do with the meanings of words? This article is quite clearly irrelevant to this conversation. I'm not offering style tips today...

Pterodactyls are dinosaurs.

When people say "x is random" it's up to you to use your brain to figure out what they meant to say, and not try to pick apart their use of language as if that constitutes a refutation of an argument.

You are asking me to read a volume of meaning into your (possibly ironic?) three word statement about pteordactyls. Your argument is left almost entirely to implication. If you become upset by my efforts to understand your cryptic, vague, misleading -- and I'll get to this -- stubbornly false statements, you only have yourself to blame.

But I'll try anyways:

The question "are pteorsaurs dinosaurs?" is not presented as a question of language, but as a question of fact. By providing the false answer "yes" to this question after I provided you with information sufficient to ascertain the true answer, I must take you to be implying "I am answering falsely in order to make a rhetorical point. I could have admitted the truth, but decided to give the opposite answer in order to demonstrate my loyalty to the principle that words have no fixed meaning. If enough people agree with a statement, then that statement because "true" from a certain point of view."

In other words, your answer is "dinosaur is defined by its manner of use; it does not necessarily correspond to any particular concept." Isn't that right?

So -- how is this possible? How is it possible that the statements "pterosaurs are dinosaurs" and "pterosaurs are NOT dinosaurs" can both be true at the same time?

Because you are actually saying "People commonly describe Pteorsaurs as Dinosaurs" and I am actually saying "Pteorsaurs are not part of the taxonomomical clade Dinosauria." These two statements can both be true at the same time. So we're both right (which means I am wrong in another sense, because I claimed that only I was right).

Restitution -- this is what you were demanding of me when you said that I needed to use my brain and charitably interpret your statements, isn't it? I think so, anyways.



But I have more to say.



You already knew that I was trying to communicate about clades, and I already acknowledged as part of my argument that people describe pteorsaurs as dinosaurs in everyday speech.

Thus, your answer trivializes the scientific question, the interesting, the factual question. We know everyone says X, but is X true? Yes, everyone knows that people commonly say pterosaurs are dinosaurs, so, necessarily, few people are aware of the truth that pterosaurs are not, in fact, part of that clade.

When people talk about dinosaurs, they understand that there is a clade (even if they don't understand what a clade is in such terms, they are aware of the concept of common descent and that this is how life is taxonomically described) called "dinosaur" to which certain organisms belong and others do not. They know that megalodon, titanoboa, 20-foot crocodilians, mammoths, and other extinct megafauna are not dinosaurs, even though some of them may even be giant, extinct, reptiles. That is to say, your average person is vaguely aware that snakes and crocodiles are not in the dinosaur clade.

If asked "Does a giant extinct snake count as a dinosaur?", I think the average person usually knows enough to correctly answer "No, I don't think so." They know you're asking them about the clade, and are trying to answer correctly. The thought-experiment proves that "giant, extinct reptile" (the only definition of dinosaur that makes sense for your statement, Restitution) is not the meaning of "dinosaur" used by anyone -- not in science, not in kindergartener's dino-books, not in their own minds, and not in everyday common speech. "Dinosaur" refers to a clade and not a vague concept (the way creature, beast, and monster do). Damn near everyone knows this, and you. do. too.

This is why the statement "Birds are dinosaurs!" is not just intelligible, but laden with meaning. The insight "Birds are dinosaurs!" is not a linguistic insight, it is a scientific one.

More than 10,000 species still roam the Earth. We call them birds.

The National Geographic is not guilty of linguistic prescriptivism; the National Geographic is trying to inform the reader of a difficult-to-understand proposition that took decades of research, the insight of so many people, and innumerable man-hours to realize: birds are in the dinosaur clade. Incredible!

Bird's don't become dinosaurs once everyone starts calling them dinosaurs. They were dinosaurs all along, and nobody appreciated this fact.

Nobody lives in a scientific conference and most people have never been to one.

Neither have I, but I want to elementary school where I learned about dinosaurs and the concept of evolution by common descent.

When people say "Pterosaurs are Dinosaurs," they mistakenly believe that Pterosaurs must be in the same clade that makes Triceratops and T-Rex "dinosaurs" because their intuition tells them that every creature appearing in Jurassic Park (or whatever) was probably a dinosaur. They're guessing that a huge, extinct reptile is probably a dinosaur, even though they know that this is not how dinosaur is defined. In other words, they are wrong.

Ignorance isn't converted into "fact" just because the majority of people agree with each other. That's the infamous non-sequitor fallacy "Argument from Popularity." I appreciate that you, Restitution, do not understand or believe that you are making an argument from popularity and NOT actually quibbling with me over words and grammar. But you are, in reality, making an argument from popularity. What you are saying amounts to (if it isn't outright) the assertion that the truth or falsity of a statement is to be evaluated according to common belief.

If you can believe this, I actually talk about dinosaurs (and clades!) as water-cooler/small-talk conversation, especially when socializing with children. The topic of "dinosaurs" is something almost every person can contribute to somehow, even the smallest child; nobody is excluded (at a minimum you can get them to say what their favorite dinosaur is!). With this in mind, I frequently discuss with friends, family, and strangers about how pterosaurs and plesiosaurs are not actually in the dinosaur clade. And sometimes I'll get a smartass kid who tells me her favorite dinosaur is Bald Eagle (delightful!).

Every single human being with whom I discussed the topic of dinosaur clades -- up until I encountered you, Restitution -- has reacted with something like:
"Oh yeah, I heard that somewhere."
"No way! I didn't know."
"Yeah! and it bothers me when people call chimps 'monkeys'"
"That's why we call them avian dinosaurs and non-avian dinosaurs."
"The more you know."

I have never in my life, until now, met a person who insisted that pterosaurs "are" dinosaurs. Your position so astounds my sensibilities that I currently believe you are taking that position out of sheer perversity.

-------------------------------

The case of whether "Luck Plays No Part in Diplomacy" is an immensely more challenging question than the question of whether "Pterosaurs are Dinosaurs." But I think we will never examine that question together because we cannot agree on what words mean. That's a shame.

But at least I learned that our disagreement is not exactly about the role of luck in Diplomacy, but more about whether and how communication is possible.

When people say "no, you are misrepresenting me, I meant X", you can't just be a snide nerd and say "But ackshually, the words you use have a formal mathematic meaning, and if I purposefully misinterpret you to be using the formal, mathematical defitiions of your words, you're wrong!"

My position is that the concept of randomness does not apply to Diplomacy (other than initial assignment of powers), and therefore luck plays no part. You can friggin' call this thing ("luck","randomness") whatever you want; I don't care!

Your contrarian position, as far as I can tell, is that the word "random" gets applied to Diplomacy in everyday conversation, therefore I am wrong. You are confusing the word "random" for the underlying concept of randomness, and calling that a criticism of my position. It's completely spurious. There is no semantic content to your argument. You are the one who is guilty of purposefully misinterpreting the words of others in order to misrepresent what they are saying.

When people say "x is random" it's up to you to use your brain to figure out what they meant to say, and not try to pick apart their use of language as if that constitutes a refutation of an argument.

And in your view, writing a 4,000 word essay with the introductory paragraph that reads:

Why do so many people deploy the word “luck” to describe Diplomacy’s gameplay?
I stand by my statement that luck plays no role in Diplomacy. But I must acknowledge that there is something going on with Diplomacy that motivates players to use the word “luck” to describe what they experience.
Since Diplomacy doesn’t involve randomized gameplay (other than the assignment of the powers at the start), what are the players trying to describe when they say that “luck” happens during a match?

does not satisfy your requirement? Or did you forget? This is why I accuse you of not reading my essays.

Meanwhile, there is a reasonable debate going on about whether the players are (or should be) randomizing their moves using computer RNG/dice/coins, and whether this refutes my "Solo Win Tip" advice to avoid the temptation to perceive Diplomacy as a game with luck. I think every other person participating in this thread understands that a player actually does need to use some kind of external apparatus to "randomize" their moves. This whole line of conversation assumes that choosing Diplomacy moves unaided is non-random.

[jmo1121109]

@Swordsman, and anyone that cares.

There is no such thing as "optimal" moves. There will be a bunch of reasonable moves. You will have to compare the strengths of these moves somehow. You should assign weights to them, to the best of your ability. (Is that trivial? Not at all. But that's a different question.)

Lol this is incorrect. There's always a moveset that best leads you to the goal of the game which is to win. If your move set allows for that without hindering it, it's optimal. If it doesn't...then it's not. It also has to appease all of the press happening in the game...but I'm explaining this to someone who clearly doesn't care about the press aspect of the game considering you've played 1 classic press game. All your arguments so far seem entirely concocted on the notion that Diplomacy doesn't matter in a game called Diplomacy.

[Carl Tuckerson]

Well, in fairness, the conversation does change substantively when gunboat is the subject. That's why I attempted to make clear caveats about gunboat vs press when posting.

[RoganJosh]

@jmo

You're just taking quotes out of context.

The situation described occurs at the end of most 1v1 games. It happens at the end of gunboat and full press games too, but not as often.

We are using "optimal" in the game theoretical sense here. Which is different from how you use it.

[jmo1121109]

@jmo

You're just taking quotes out of context.

The situation described occurs at the end of most 1v1 games. It happens at the end of gunboat and full press games too, but not as often.

We are using "optimal" in the game theoretical sense here. Which is different from how you use it.

If you're going to continue making the claim that it *EVER* happens in full press games you need to back that up with at least 1 example and explain how the Diplomacy aspect of the game has no impact on it. If you want to stick to 1 vs 1 or Gunboat I'll definitely give you the point that a 50/50 coin flip scenario is far more possible.

Some of those were entire quotes though...so they can't be out of context.

[RoganJosh]

It turned into a discussion about whether such a situation is even possible. And it is possible. Which implies that diplomacy contains an element of randomness, even if that element is very small.

Actually, my own opinion is that, while it is possible to have a situation where you best move is to flip a coin, the situation is so uncommon that if you are writing a blog with the aim to help people improve their gunboat skills, then you should ignore it.

Can you generate a list of all full press games that ended in a stalemate along the standard stalemate line?

[RoganJosh]

Is it so hard to imagine a situation where you have one solo contender and the remaining powers are trying to form a stalemate line? There is not much diplomacy left in those situations. Coordination of moves, yes, but not much of negotiations. And, then, a last 50/50 which determines whether the solo contender snatches the 18th center or whether the defense completes the stalemate line.

[swordsman3003]

Actually, my own opinion is that, while it is possible to have a situation where you best move is to flip a coin, the situation is so uncommon that if you are writing a blog with the aim to help people improve their gunboat skills, then you should ignore it.

<33333

Actually my blog is whatever I feel like writing about, but I do feel like writing about Diplomacy a lot...

Despite your suggestion, I think I really will examine this idea about coin flipping and come back next month with a new post. I'm not going to start out with any conclusion in mind; I'll see where what I'm learning takes me. I really had never ever considered anyone (myself or my opponents) actually using RNG to make a Diplomacy decision until I started this thread.

And just so everyone knows, I am keeping notes on the stuff you are saying re: whether to intentionally use RNG to decide your moves. I'm kind of giving up replying because I don't want to make up my mind until I've meditated on the idea for a while. I read your other posts RoganJosh, and I'm taking them into account for later even if I'm not replying now. (This applies to your thoughtful posts, Your Humble Narrator -- what you are telling me is what I'm using to outline my questions. Kremmen, you too.)

Okay. With that out of the way, I really want to poop on this point:

It turned into a discussion about whether such a situation is even possible. And it is possible. Which implies that diplomacy contains an element of randomness, even if that element is very small.

This is, like, a suped-up verison of Russell's Teapot, but unironic. You are saying "X is possible, therefore X exists."

The hypothetical or theoretical possibility of a situation existing does not mean that it exists. Jmo's challenge -- to prove that this has happened even once in a real Diplomacy match -- is a serious and powerful challenge to your claim. And the burden of proof is on you to show that the situation exists in reality, not just hypothetically, because you are the one claiming "diplomacy contains an element of randomness."

Is it so hard to imagine a situation[...]

Jmo's criticism, which I am adopting, isn't that the situation unimaginable, it is that the situation has not occurred.

I'm not saying you can't evidence your conclusion; I am encouraging you to actually do it. If you do, that may change how I think.

[jmo1121109]

Is it so hard to imagine a situation where you have one solo contender and the remaining powers are trying to form a stalemate line? There is not much diplomacy left in those situations. Coordination of moves, yes, but not much of negotiations. And, then, a last 50/50 which determines whether the solo contender snatches the 18th center or whether the defense completes the stalemate line.

Yes, with all the factors that go into a press game, such as everyone being online to coordinate, everyone agreeing on the best move set for the coin flip (1 person can't be responsible for flipping a coin if 2 powers are involved), no press being leaked, no previous game play compelling any of the defenders to use a different factor to make the choice, and everyone involved who's spent months on this game deciding to toss all their press work to get reads out the window to decide days of press work with a flip of a coin, is something that I just cannot fathom. If you can show me a game where this actually happened, I'd obviously have to reconsider, but I really doubt it exists. And if it does exist, I really doubt it was the best move.

[Squigs44]

I likewise have enjoyed this conversation and have reconsidered certain aspects of game play. It has always been interesting to me to find the balance between theory and practice in anything I am studying, but especially in game theory. I might start up a thread to do some Riddler type game theory games if there is interest in people participating. I also realized that I need to do some more studying on game theory so I can keep up with you all.

[Your Humble Narrator]

All we need for the example I provided to prove the possibility of luck determining an outcome in Diplomacy is for me, Italy, and Austria to all be equally adept at predicting one another's moves. If we accept that there is a cap on potential Diplomacy Skill (or whatever term we want to use to refer to the ability to predict an opponent's move accurately), this is especially easy to swallow: just imagine all 3 of us are as good as Jmo (PBUH).

If we reject the notion that there is a human limit on Diplomacy Skill, then a tie in Diplomacy skill is still perfectly possible--if a player's skill can transform a 50/50 guess into a 55/45 guess, why can't another player's skill counter that effect and bring it back to 50/50? To accept the possibility of the former is to accept the possibility of the latter. It's not the sort of thing for which we could ever identify a concrete example, because we don't have a means to measure Diplomacy Skill quantitatively. But, by the same token, we have no way of knowing for sure whether Jmo's above-average guesswork was made against opponents of lower Diplomacy Skill--if it turns out they were equal, then the guesses were luck, but we'll never know.

Luck comes into play when you run into a situation like my example and all parties involved are tied in Diplomacy Skill. The coin (or 60-sided die), i.e. luck, determines the outcome.

[Your Humble Narrator]

Oh, and even if you reject the possibility of two players ever being equal in Diplomacy Skill, then you could still attribute to luck every instance of the player of lower Diplomacy Skill guessing correctly against the player of higher Diplomacy Skill. Which I am CERTAIN you would all agree happens plenty often.