Diplomacy Dojo Episode 9: Reading the Board27 min read

Intro 0:00  

The Diplomacy Dojo is a weekly discussion, led by Your Bored Brother, about Diplomacy tactics and strategies. Let’s listen in on what our players are discussing this week.

BrotherBored 00:16 

Okay, Paul’s here today at the Diplomacy Dojo; how’s it going, Paul?

Paul 0:19  

Oh, it’s going alright. The first question I had is just to talk about signalling in Diplomacy. Like, are there any do’s or don’ts that you have in mind when sort of figuring out how to do it and, like, the best way of getting the other people on the board aware of the intentions you want them to be; and for like coordination in Gunboat? Can you do it in a way that doesn’t cause you to, say, waste the entire season?

BrotherBored 0:53  

So is this a question about Gunboat Diplomacy only or Diplomacy in general?

Paul 0:58  

I mean, you can talk about Diplomacy in general but I guess people use their press, if they want to make signals there. I guess the focus is less on the moves being your signals.

BrotherBored  1:13  

Okay, I understand that we’re talking about Gunboat, then. What else?

Paul 1:18  

Maybe sort of ideas about like, when you’re, let’s say when you’re spectating a game, like, what are you looking for? I mean, what are you trying to understand when you sort of— if you were to play through, say, a high-level Diplomacy game, which you aren’t playing in?

BrotherBored 1:37  

How to spectate a match? That’s fascinating.

Okay, so let’s begin by talking about signaling in Gunboat Diplomacy. That’s a topic that I’m very interested in; and I think I know as much about it as anybody can know. oT put it another way, I’ve played so much Gunboat Diplomacy, and at a high level, that if there’s a form of signaling that I’m not aware of, then I’m not really sure it counts as signaling; because it means that an experienced player wouldn’t understand what you’re trying to communicate.

So, the app you’re playing with matters.

Paul 2:16  

Yes; because you either are allowed to signal with illegal moves, or that’s not possible.

BrotherBored 2:23  

That’s right. So let’s first talk about more confined— I’m more familiar with Web Diplomacy, and the other ones that are programmed similarly, that do not allow you to enter totally illegal moves, moves that would never ever work.

So when I’m playing a Gunboat game, and I’m thinking about how to send signals, the first is that there are regular moves you could make that send signals. For example, attacking a player signals that you want to attack them. That’s not to be underestimated, the value of signaling what we’re doing by doing it.

For example, let’s say that you are Turkey and you open by moving an army in Smyrna to Armenia. That is how you begin an attack on Russia in 1901. But you’ve also communicated to the other players, particularly Italy and Austria, that you are going to attack Russia, and they may decide not to attack you. And that’s usually the thinking, right? “Okay, if I begin by attacking Russia, maybe those players will decide to do something else other than attack me. Maybe Italy will do Lepanto or something.”

Now, they might actually do it, but the point is, that every time you take an action, it does signal your intentions; because you have taken those actions. That doesn’t cost you anything extra; it doesn’t cost you anything tactically.

The next least costly form of signalling is to issue orders, pieces—issue orders that you think will fail or not matter with pieces aren’t doing anything anyway. A great example of this is in 1901, in Spring, Germany often moves Kiel to Denmark and in Autumn maybe bounces Russia out of Sweden, or maybe doesn’t. There is a subtle difference in whether Germany simply allows Russia to enter Sweden, or whether Germany issues “support Russia’s fleet to Sweden.” The outcome’s the same, there’s no way Russia’s move to Sweden can be stopped by another player. Allowing Russia into Sweden suggests, “I’m not particularly friendly towards Russia,” and supporting Russia into Sweden says, “Hey, I want to work with Russia. I’m going to help Russia in the future.” Germany could have issued either order; so why issue the support order, unless you want to signal some kind of extra level of friendship towards Russia?

Other examples include support hold orders that don’t matter. You can issue a support hold order on a neighbor’s piece; it won’t make any difference whether you issued hold or support hold but when you see, “Okay,” the other guy sees, “I see you. You did a support hold.” That signals some kind of friendship.

So in 1901, Italy could just hold with Venice, which indicates a quite icy view towards Austria, or could issue Venice support hold to Trieste, which doesn’t do anything different. The outcome is going to be the same, there’s no way that the support hold is going to matter. But the latter version says, “Alright, yeah, I haven’t moved Venice, but I would like to be friendly to Austria.” Now, of course, everyone can bear in mind that you can send signals that are misleading. They could be false signals, but yeah, it’s a signal nonetheless.

My personal favorite version of using a signal with a piece that’s not doing anything, is if you have an army that’s just going to hold, but it has some convoy opportunities. On the Web Diplomacy interface, this is actually maybe the best way to communicate clearly, because you could convoy that army using opponent’s pieces. It will fail for sure, but it draws a line. It could draw a line very far because you can chain together a lot of convoys. So this convoy will fail. But you could do something like draw a line from Norway to North Africa, all the way across the map and send a very clear signal of something. What you want, the players to move towards this direction, or back away from you, or attack somebody, something like that.

And then finally, there’s the third kind of signaling, which is to make a signal that forgoes doing something useful with the piece. And that could be something like, I really wish I was moving this piece around, but I want to communicate to another player that we are friends. So this could be something like, something—It’s not a move I would do, but this is an example. In Spring 1901, France supports London to English channel using Brest. That gives up using Brest as a move, but shows that you anticipated England would move to English channel, and you’re cool with that and you want to work with England. That sends us such a loud signal like, “England! Be my ally!” is what that’s trying to say. Because you thought England was going to move to the English channel, and you didn’t bounce it, in fact, you supported it. Whether England accepts that offer of alliance, I don’t know.

Something I actually have seen a few times in a game is Russia supporting Turkey to Black Sea in Spring 1901, to say, “You know, I’m not going to attack you. Please, don’t attack me, let’s be friends! Juggernaut, let’s make it happen.” That’s something that’ll happen, and it actually gives up a move. Russia could have bounced Turkey in Black Sea and gives Turkey a considerable tactical advantage by not making that move, but maybe it’s worth it to send the signal.

When I’ve played as Turkey, I have sometimes gone, “Hmm, the Russian player wants to be my ally that badly and understands this game well enough how to send the signal, maybe I do want them as an ally, that seems pretty cool that they did that.” It works on me; it won’t work on everybody.

So those are the three kinds of signals. Before I go into further detail, what do you think about these points?

Paul 7:53  

I think I can at least see where this is coming from. I mean, for instance, I guess you can also, like, move troops away from a common border. I think that’s probably also sending a signal even though… Like, let’s say you have an Austria-Russia Alliance, and you want, you would probably want to make sure that you know, Galicia is under control, but you might move your army to Budapest, just because it’s, you know, you’re showing that you’re willing to keep an alliance with Russia and you trust them to not fill in the space.

BrotherBored 8:26  

Maybe there’s like a halfway point where you’re trying to go, say, like when a dog or a wolf rolls overexpose to their belly to say like, you know, “I’m not a threat.” There are ways to do that as well. Yeah, like, well, maybe that’s a better way of characterizing what you’re talking about. Just moving everything away from another player, even if that’s unnecessary or makes you vulnerable, but you want to communicate it to them, “I’m not going to attack.”

Paul 8:48  

And what about builds? I guess you can also make signals with your builds, although maybe that also falls into the first category in your opinion.

BrotherBored 8:59  

Yes, but I think it’s worth pointing out, because I think in Gunboat Diplomacy, the builds communicate perhaps the most of any other normal move, because in deciding where the pieces are built, and what type of unit it is, the player has made a likely permanent choice on what their offensive capabilities are. For example, France in 1901 builds an army in Paris and a fleet in Marseille, then that communicates pretty strongly, “I’m not going to attack England.” France could have built in Brest and didn’t. Okay, that’s a big communication. Or whether Russia builds a fleet or an army in St. Petersburg in 1901 has immense implications for Russia’s tactical capabilities on offense and defense and it’s probably a permanent decision. You’re making that decision strategically, but it also communicates something. So I would say yes, in the first category.

A great example of this might be, if Germany is building a fleet in 1901, Germany could build a fleet in Kiel, or a fleet in Berlin. And where that fleet is built communicates quite a lot because a fleet in Kiel or Berlin can both move to the Baltic Sea, but a fleet in Kiel can also move to Holland or other anti-English things. So the fleet build in Kiel is ambiguous, whereas, the fleet in Berlin is crystal clear, “I’m going to attack Sweden,” no other interpretation of that build.

So if you make the fleet build, you should be thinking about what it will communicate to the other players because you’re sort of locked into what you’re going to do. And some builds don’t do this, some builds are sort of obvious or as a matter of course. Turkey almost invariably builds a fleet in Smyrna in Gunboat Diplomacy. Doesn’t really say anything. It’s just so common; it’s hard to read much into it. Austria often builds two armies, or all armies. Doesn’t necessarily mean that much.

I think that, as far as your “do’s and don’ts” aspect of it, my threshold is to figure out which players are capable of interpreting these signals; because if you’re dealing with a player who does not seem capable of interpreting the signals, there’s not much value in sending them those signals. So if a player is sending signals, they are sending signals. “Okay, this other player is issuing orders that only make sense as signals, of the latter two kinds.” They’re using their extra moves to issue orders, or they’re even sacrificing a piece’s move to send signals. They’re going out of their way to send signals. That implies that they would understand if you sent them a signal, and you can send them signals right back. So if they’re doing something like, the French player moves to Burgundy in 1901, and then support to Germany into Belgium. Okay, regardless of which power I am, I learned in 1901, the French player understands the idea of like, “Okay, I could support Germany to Belgium in order to convey that I want to be allies with Germany.” That’s what that means, and then I would anticipate that if—regardless of which power I am—that if I sent signals that I wanted France to understand, probably France would get the meaning of it.

The next bit of advice I have is that I have played Gunboat games where afterwards, the players tried to explain some signals that they were sending, and nobody understood this. Because the players don’t get to speak in Gunboat Diplomacy, to understand the language of how players are signaling means you need some experience with what players consider to be signals and how they interpret them. To put it another way, if you’ve made up this signal yourself, and you’ve never had other people send you the signal that you’re trying to send, you might be trying to speak in a private language that other people aren’t going to get what you’re trying to say. And I think that does come up in Gunboat Diplomacy, where players think they’re sending a signal, but they could have appreciated that that signal was not going to be heard, because it was just too unusual, and the other players didn’t understand what it means.

So based on that, if you want to signal in Gunboat Diplomacy, I say spend some time talking to players and ask them what their signals meant after the match, and ask them if they understood what your signals meant after the match. And you might learn “Oh, you know, I said that I was trying to communicate this and other people didn’t understand it.”

In a high-level Gunboat Diplomacy match, you can communicate some pretty subtle stuff. It is possible to communicate things like, “I’ll vote draw if we eliminate this other player.” It’s possible. In order to communicate something that subtle, you need to be playing with players who are capable of understanding that; so in a low-level Gunboat Diplomacy game, maybe like a pickup game, or you don’t really know who’s playing it, then I wouldn’t count that much on communicating a lot with your signals. But in a high-level game, you can go pretty deep.

Any other thoughts come to mind something, maybe something particular in your experiences?

Paul 14:07  

So I guess one thing which wasn’t discussed is also like, choosing to disband a unit instead of retreating it or you know, things of that sort, they can, that can send a signal that, I don’t want you to worry about that this unit could come and harm you later on or in the Winter, so you don’t need to base your winter build around my unit still being there.

BrotherBored 14:32  

That’s a great example of a specific tactical situation where there’s a lot going on there. I agree that you could choose to retreat your unit or you could choose to disband it and rebuild it, and giving the other players information that you are for sure, rebuilding it, don’t be afraid of this, can be immensely useful.

That particular situation I find is good when you’re trying to get somebody who you’ve been fighting with to change sides. You know, “Hey, let’s stop fighting, let’s join together. Someone else is a bigger threat.” Laying down your arms, so to speak, can go a long way in communicating that; because even if the player isn’t paying attention, let’s say the other players like, I’m just going to kick your butt,” they’re not paying attention to the big threat, they’re going to go, “Why the heck did this guy who I’m fighting lower their defenses?”

Something else is a term that I learned from, a concept I learned from learning about bridge bidding: Whether a bid is “loud.” Some signals are so unambiguous and leave—don’t leave much room for other players to send any countervailing signal. And there’s a time when you want to say something very “loudly.” And that can apply in Gunboat Diplomacy. For example, let’s say that I am Italy, and Austria is reaching 13 or 14 centers, and I have, maybe I have Spain or Marseille. And I think, you know, “England is encroaching way too much on my area, in Liberia, and I think I’m going to get shut out of the draw somehow because if I go down to just my three home centers in Tunis, I’ll be expendable. I won’t be necessary to form a stalemate line.” Alright, you know what? I want to shout. I want to scream into their ears, “If you dare come one step further, I’m throwing the game to Austria.” And the way that I might do that is to send every single one of my pieces towards Iberia. All of them. Even if it’s not really technically necessary. Even if it means Austria is going to be able to conquer centers from me. It’s so loud, there’s no other interpretation. And that can be very effective! “Hmm, maybe we’ll cut you out of the draw—” “No! No way.”

So I think it is advantageous to send signals early and often to demonstrate that you are capable of doing this, as that may make you an attractive ally to players, and may induce them to send messages to you. And if there’s a player in the match who’s not really understanding what’s going on with that, then you gain some mutual advantage over that player.

I also think that if you end up playing for a draw later, there may be another player who’s in a position to decide whether to destroy you or to work with you. Because in Gunboat Diplomacy, you can’t talk to other players about what the stalemate line is, and a player who has been sending and responding to signals implies that they are capable of forming a stalemate line. It implies you know how to play; and those players may decide to work with you when the time comes to form a stalemate line. It’s advantageous to communicate that

You might not necessarily have to communicate so much immediately to get away with that, but that’s what I personally try to do. I want to show that about myself; because even in a high-level Gunboat Diplomacy game, using signals to get somebody’s loyalty as an ally can be really useful in creating a board state from which you could actually solo win; because they might give you some space to cross the stalemate line.

The other thing I will say is, don’t lie too much with your signals. Where you support hold somebody to be their friend, and then immediately attack them, stuff like that.

Well, maybe we could go to the next topic: How to spectate a match. That’s very interesting. That’s an interesting question. And I have observed so many Diplomacy games, because players asked me for advice, and so I follow along with games. I think I have some insight into this. Are we talking about Diplomacy in general, like press Diplomacy?

Paul 18:51  

Yeah, I mean, both press and Gunboat are fine. Obviously, you can see less in the press game, because you’re probably aren’t privy to all the conversations unless you’re somehow the GM of the game or spectating.

BrotherBored 19:06  

Yeah, that’s where I was going next, is that in a Gunboat game, you don’t know anything more than they do. Or anything less, I mean. You don’t know anything less than the players because all their information comes from just looking at the game board.

So I think it’s pretty easy to spectate a Gunboat Diplomacy game for that reason. You can just pop in and figure out what you think the players are saying and stuff. This is really helpful when I’m trying to coach players on how to be better. Because in a Gunboat game… When they’re playing a press game and asking for help, the player who I’m helping has to summarize, but in Gunboat, that’s not the case, where they say, “Well, it looks like Russia wants to be my ally,” or whatever. I can say, “Well let’s look harder at these signals. I don’t really think that’s true,” or something like that. I help them interpret the signals because it’s the same information, and it’s not very much information, it’s just the pieces.

I think that observing can be very illuminating. Especially a high-level game where you can watch players send signals, and try to interpret what they’re saying. In order to go, “Well, I think England’s going to do this based on what I see, so let’s find out.” In press Diplomacy, it’s… Maybe it’s a little intriguing to spectate because you don’t really know what’s going on. Where like, okay, Turkey and Russia look like they’re fighting, but not that much. Are they fake fighting? And if you were participating in the match you might be able to figure that stuff out, or just directly ask, you know, “Hey, Turkey, are you fake fighting, or is that real fighting?” Or something like that.

I guess the conventional wisdom on spectating press games, is that, it’s a lot more speculative. Where you think, “Okay, that alliance looks pretty strong,—oh, I was wrong!” or “That alliance looks like it’s falling apart. Nope, they were just faking it.” That sort of stuff. I think a relevant concept here is the idea of “reading the board.” Looking at the pieces on the board, and trying to glean as much information as you can from the pieces. It’s an “action speaks louder than words” kind of thing.

And Gunboat Diplomacy, which we just talked about, so much, is great for learning how to do that. Reading the board is the only form of communication. But that still applies in press Diplomacy, and you can still read the board. And some players, I think, what they do is, they use their press to try to interfere with how you might read the board. And so I think that has helped me many times see through the deceptions and manipulations of other players. Where I say, like, “Nothing you say can convince me that you’re not hostile, because these moves are just so hostile! I need you to move differently. And I don’t care about what your words are.”

And so spectating a press game of Diplomacy can be fascinating, because of how much there is that you don’t know. And you’ll be surprised and delighted when you see what the players do. The players are a lot more unpredictable than they are in Gunboat, in my opinion. Like, in regular Diplomacy, press Diplomacy, you’ll see players simultaneously backstab each other, and crazy stuff like that. That doesn’t happen as often in Gunboat, the players are a little more slow-moving in their strategic trajectories. Do you have other thoughts on spectating press games?

Paul 22:32  

Do you have any sort of guidelines, like, for instance, when you read the board, what are you looking at? Like, first of all, what are you trying to sort of pick up from your, say, first glance?

BrotherBored 22:48  

I measure what a player does against my metagame expectations. So, for example, if I see an opening move where Austria held with Trieste, that’s really different. I am not saying that’s a horrible move or anything, I’m not saying it’s unplayable, just… I would expect most players to either move to Albania, or if they’re so afraid of Italy, to move Trieste to Venice, and do a hedgehog opening, which would be even less common. But to just hold with Trieste, is either… it’s weird. It’s very weird, and so to see something like that, I go, “Okay, this player is either very experienced or very inexperienced because that move is not really that common, and not considered terribly good.” So what’s going on here? I want more information, so I would look to the next couple of turns and say, “Okay, this player is otherwise making reasonable moves. They might be a pretty good player who’s just got an idiosyncratic style of play or something,” or whatever. And if they continue to make unusual and bad moves, I would think, “Alright, this player is not very good, that’s probably the reason why I’m seeing what I’m seeing.”

I look to the players’ builds a lot. As we were talking about earlier, I think the builds communicate the most. Because at some point, the players become incapable of pursuing certain strategies if they’re so heavily built in one direction or the other. For example, if in 1901, Germany builds all armies, let’s say Germany builds two armies, or three, then that is… I least can say, “Okay, this Germany is not attacking England because they would have to have another fleet if they wanted to attack England early on.” I could presume that that was maybe negotiated with England. There’s a lot I could infer from a build like that when Germany builds armies in 1901, when they could have reasonably built fleets. That’s probably the easiest thing to understand.

Paul 24:53  

So I guess then what you’re saying is you’re sort of trying to sort of understand, like, the obvious things on the board, like what sort of deviates from the norm, and then—well, depending on how sort of familiar the board state looks, then you have an idea about the sorts of plans that you might expect the various powers to have in a given situation. And whether they follow that or not, will give you maybe a sense for their strength or experience.

BrotherBored 25:25  

Yeah, that’s right. One of the things that is hard about just spectating is that the players often make really small deals or make small lies that you wouldn’t necessarily be able to infer from how they move their pieces. For example, if you see Italy move Venice to Tyrolia and Rome to Apulia in 1901, what was the basis for that? Did Italy lie its way in there? Did Austria agree to this? Did Germany agree to this? Did they discourage Italy from doing so, but let it happen, or what? You know, because it’s unusual for Italy to make such an opening. You wouldn’t be able to figure it out, you wouldn’t. That’s one of the limitations of spectating press Diplomacy, is there is some stuff that’s just quite challenging to interpret.

That probably explains why after-action reports are, you know, a perennial source of interest in Diplomacy players. They want to read and find out what really happened, what actually happened behind the scenes. Journaling of specific matches, which is something that… I don’t know that I necessarily invented it, but I certainly started doing in-depth journals a couple of years ago, and now a few other players have done things like that, including the media Wars game that’s being published as we’re recording this.

Like, the only way you could ever really get to the bottom of it, or get to the heart of the Diplomacy game as a spectator, is you would have to have access to the press. And not just the press itself; because the press is an artifact of the gameplay. The Press is a kind of move in the game, and not necessarily what the player was thinking. Because maybe I send press to everybody, and it’s all lies. So just reading the press wouldn’t tell you what I was thinking. It would have to come from me somehow. I think if you really, truly, like, “Ooh, I just want to understand a press Diplomacy game, really understand it,” you got to depend on after-action reports, and these journals that exist out there.

Paul 27:39  

I mean, the ARRs are going to sort of get you closer to the truth, although I mean, certainly, it’s possible for players to sort of embellish to make themselves look better after the fact.

BrotherBored 27:52  

You’re very wise. That is why I personally valued in myself creating contemporaneous journals so that you get as close to what the truth was, at that moment. I can’t use hindsight. I have accused some Diplomacy players of lying in their ARRs in order to make themselves look better as a player; because they want people to think that they don’t lie, they don’t backstab, so  they say, “It wasn’t a backstab,” in the ARR, but I’m like, “It was. Why do you gotta do that?” I have a personal policy of being 100%—I attempt to be—100% truthful in my ARRs. I say, “The match is over.” I’m not playing a long game of portraying myself as this player who just never lies, never backstabs, although maybe that’s good. Maybe playing that long game could help you win championships. I don’t know, but I don’t do it.

But there’s also the problem of just not necessarily understanding yourself; just because you played in the match doesn’t mean your assessment of what you were thinking four weeks ago, in a drawn-out online game, is you’ve remembered it accurately. I mean, I can’t remember what I had for breakfast sometimes, let alone what I was thinking on a particular day, five weeks ago. So I agree with you, I appreciate your bringing that up. It’s just somebody’s afterward explanation of what was going on.

Also, AARs tend to focus on moments that were really important in hindsight. Like, “Now that the match is over, I see that this was a turning point,” but at the time players might not necessarily have been thinking like that. And so the hindsight bias can really affect how you’re thinking about the game. So for example, in my media wars video journal, when I was very close to getting—I was getting my rear kicked in and I thought I was close to elimination, you can see it! You know, here’s BrotherBored, just on the verge of despair. I don’t really know what else I can do. But I later initiated a solo win run. In an AAR, I could say, “See, I’m just a genius. Even though I was getting attacked so early on and losing, I never let my eye off the ball and I was trying to solo win the whole time.” That wouldn’t have been true. I try not to think like that. Is that the kind of story you’re getting at, Paul?

Paul 30:28  

Yeah, and I mean, also, I feel like people try to make an attempt to sort of downplay their own diplomatic skills. I’ve noticed this a few times, whenever you have interviews with winners, they’re like, “Oh, no, that’s all everybody else.” I mean, they don’t want to be targeted. It can help. I mean, definitely you have to take, understand what is said and also like, maybe what’s not said is left to the reader.

BrotherBored 31:08  

The, “Oh no, don’t mind me; I’m a simple, country Diplomacy player. I’m not up to anything,” is part of the schtick?

Paul 31:20 

Yeah.

BrotherBored 32:21

Okay, I think we’re coming to the end of it, and we’re going to play Among Us on the server, I think!

Outro 31:35  

If you enjoyed this episode, remember to subscribe and review the podcast. To learn more from Your Bored Brother, and to participate in the dojo, visit the blog at brotherbored.com. Thanks to loyalty freak music for the theme music, “it feels good to be alive too.”