Graveborn Musings – Politics, Complexity and Multiplayer Strategy (Finale)

A long time ago, in a blog very much like this one, I started the unexpectedly ambitious task of offering up a clear definition of politics. Just to make it clear where I’m coming from, I have always been one of those people who loves multiplayer more than duels, because it is so much more complicated, interactive and tricksy, but I’ve also heard a lot of people complain that multiplayer is no fun because it’s all about politics. We have all heard these complaints, and even many multiplayer enthusiasts have had games that left a bitter aftertaste because they ended when someone made a deal with their buddy and won when they didn’t deserve it. But the problem is this: if ‘politics’ is at least sometimes a dirty word, and nobody knows how to distinguish between multiplayer politics and multiplayer strategy, then multiplayer itself becomes tainted.

In other words, multiplayer gets a bad rap because some people think that politics determines the outcome of games, and the people who love multiplayer can’t defend it because they can’t offer a satisfying explanation of the difference between politics and strategy. That also means that if someone complains about it and asks for a politics-free game, then nobody will be able to agree what that means or even whether it’s a good idea. That’s why it’s important that people who enjoy multiplayer have a clear understanding of what politics actually means.

In the first article, I established three basic premises:

  • Politics is unique to multiplayer[1]
  • Strategy is still the central element of success in multiplayer
  • Politics is different from strategy

In the second article, I explained the third point at length, showing how some of the things that are almost universally regarded/reviled as multiplayer politics are really strategic. However, I’ve heard from a lot of people who disagree with that idea. Some people say – and I know I used to believe – that politics is just part of strategy, like a distinct subsection of Magic strategy that only develops in multiplayer. That seems like a valid point, and if you still think that when you’ve read this article then I want to hear from you in the forums. But the first thing I would ask is: how do you distinguish between the political and non-political parts of strategy? The more I looked, the more I came to see that most of what I used to consider politics fell into two clear categories. Some of it was just strategic thinking adapted to an environment with more opponents,[2] making it indistinguishable from the other parts of strategy; the other category is stuff that is just so qualitatively different from strategy that there doesn’t seem to be any reason to think they’re the same thing.

Let me put it like this. Imagine you’re in a three-player game and one of your opponents has a Blightsteel Colossus. You can think of three options:

  1. Play a Keiga the Tide Star to deter an attack
  2. Promise your opponent that you’ll counter the Swords to Plowshares that you know your other opponent has, so that the Colossus can kill him first, buying you a turn
  3. Reach across the table and stab your opponent in the windpipe with a fork.

All of them are things that you can do to increase your probability of winning, but that doesn’t mean that they’re the same type of thing at all – which is why strategy, politics and grievous bodily harm are different categories of multiplayer behavior.

OK, so maybe that example was kind of loaded. How about the kind of example The Ferrett gave, where the person who made the strong play got killed; is that about strategy, politics, or both? Well, if you make a strong play, you impact other people’s chances of winning – that’s more or less the definition of a strong play. What can you expect other players to do if they are playing strategically? Usually they will try to increase their chances of winning by neutralizing the threat that you just presented. If your strong play was a Jin-Gitaxias then Doom Blade on the praetor would neutralize the threat and return the game to the status quo ante. If the threat is Doubling Season, Elspeth Knight Errant, pop her ultimate ability and sit behind a wall of invulnerable soldiers, then killing you is probably the only way to deal with the threat; if your opponents have enough evasion or reach then that strong play might cost you the game.

But what part of your opponents’ reaction would you call political? Nothing – they are using their resources and playing cards in a way that they calculate will win them the game, which is the essence of strategy. Is it political if you hold back from making that strong play and wait for “the opportune moment”? Of course not – you are playing around the strategies of your opponents, which is strategic. Really, there’s no qualitative difference between holding back on playing Karn until after your friend pops their O-Stone, and leading with your second-best creature in a duel when you suspect your opponent has removal in hand. Like my friend with The Abyss found out, timing is even more of an issue in multiplayer, but it’s a strategic issue, not a political one.

Politics is…[drumroll please]

So if we accept that politics is different from strategy we can start to look deeper at where that dividing line is. Earlier, I presented a list of 12 ways in which the strategic environment changes when you added more players:

  • More targets for each spell
  • More targets for each attack
  • More threats
  • More answers
  • Scaling effects
  • More damage to deal
  • Greater variance in board positions
  • More hidden information
  • Greater variance in cards played
  • More uncertainty
  • More complex decisions
  • More factors in the decision making process

Some of you may have noticed that I listed both hidden information and uncertainty, which might seem redundant. Hidden information is at the core of most card games, whether Magic, Poker or Go Fish; unlike Chess or Monopoly, where all of your resources are out in the open for everyone to see, part of the strategy of a game with hidden information is to play around the possibility of what you or your opponents might have. In a Magic duel, most of your questions come from this: not knowing what they’re playing, what they’ve drawn (or what you’ll draw next), how they sideboarded and so on; once they play the cards, the duel gets much more predictable. You can never be sure which of their critters will swing this turn, or how they will use the Doom Blade that they revealed with Dark Tutelage, but once the hidden information is revealed, you know there are only a few ways it could go.

However, in multiplayer, there are more options and harder decisions, for you and for your opponents. In multiplayer, there is a highly significant amount of extra variance, or uncertainty, which comes from hidden intentions, a special category of hidden information that is almost unique to multiplayer.[3] Who will they attack next, how are they choosing their priorities, who have they decided to kill, who are they most scared of, who are they pretending to go along with but are willing to kill at the first sign that they’re getting out of hand? They may even have the kill on the board, but you can’t know for sure whether they going to use it against you (which is why it’s almost always a mistake to scoop).

Seeing that politics is a feature of multiplayer, and uncertainty about hidden intentions is the element of the game that explodes the most in multiplayer, it seems reasonable to think of the two as related. In a duel, you’d normally only show your opponent that you have an answer in your hand if you’re trying to show them that the game is over, but in multiplayer you might do it to manipulate them. Trying to manipulate your opponents is something that you can do just by playing cards (holding back a threat, as discussed above, or dropping a rattlesnake), which means it is at least partly strategic, but the manipulation of hidden information is qualitatively different. That’s why my first stab at a definition of politics in multiplayer is:

Politics is the manipulation of hidden information to increase your chances of winning.

Notice that this can involve both showing cards to others and/or telling them what cards you have (I’ve got a Furnace Dragon, so leave me alone and I’ll be able to take care of that Darksteel Forge) and telling people what your intentions are, either in the form of a statement (“Don’t worry about my 50/50 Lord of Extinction, I’m going after Steve first”) or a negotiation (“I promise not to attack you with the Lord of Extinction if you blow up Ashling to kill Steve’s blockers”) Notice that this covers all forms of explicit cooperation, because you can’t make a deal without giving away at least part of your intentions (my intention to Fling my fatty at you after I kill Steve is still hidden information, which is why it is dangerous to let politics replace strategic decision making).

Celebrity reader Andy left a very pro-politics comment on the forums:

I am totally unapologetic about playing politics. It’s like the only thing I’m good at when it comes to Magic…“Blow up that Pernicious Deed or I will hit you with Cruel Ultimatum next turn”, “I have a hatchet in my car and am perfectly willing to take your right fingers”, “If you attack me I will tutor Identity Crisis and cast it targeting you”, and the like are pretty common at local tables.

The first example is Compellence, the third is Deterrence, and the second is technically illegal but as long as it’s effective in your meta that’s all that really matters, but all of them involve revealing hidden information, either about the cards you have or about how you will use them. Deterrence can be done without revealing hidden information, although compellence is a little trickier, but the point is that what Andy is talking about takes strategy and adds a slice of something that is completely different from strategy.

Of course, if someone is playing politics at your table, then you’ve got to consider both the strategic and the political levels in deciding the right play. For example, attacking Andy may be the best move for me based on the current board state, but if I add the costs of getting hit by Identity Crisis to my calculation of the pros and cons of each move, then I might reach a different decision – or I might decide that Andy needs to die, and attack him regardless of the costs. On top of that, the player making threats still has to make decisions strategically in order to win. If I attack him with my monored deck, and then the black player taps out to play Balthor, the Defiled and the blue mage Braingeysers for 20 cards and then plays Reliquary Tower, then Andy is going to have to reconsider his stated intention of blasting me. Playing politics would also add an extra dimension to his calculations: the most effective use of his cards is to cast Identity Crisis on the blue mage, but if he doesn’t carry out his threat against me then he loses credibility the next time he makes a threat like that. Also, how would everyone else feel about him if he cast Identity Crisis on me, with my three-card hand and largely irrelevant graveyard and the blue mage then used their massive card advantage to combo out and win next turn?

I had a situation recently in a four-player EDH game that illustrates exactly how this definition of politics works, why it is different from strategy and why it often pisses people off. We were playing with the following generals in order:

It was Bosh’s attack step and he was wondering where to send nine points of pain. The two strongest players were Oros, with lots of life and lots of ways of slowing things down, and Rhys who looked like he was only a few cards away from locking things up and going crazy with unkillable tokens. My deck was acknowledged Vampire Tribal, and I’d already revealed that I was playing with such powerhouse spells as Explorer’s Scope and Urge to Feed; I was clearly less of a threat, and he knew it. In addition, I was going for Rhys (unlike the Gargadon, Mirri could get around the tokens) and had tried to signal that by leaving myself open, while Bosh was the only one capable of dealing enough damage to bring down Oros. Strategically, the right play was for him to hit Oros.

That’s why my heart sunk when he asked Oros, “Who are you going to hit with the Curse?” Oros jumped at the opportunity and said “I won’t hit you if you don’t attack me with the Gargadon.”

So let me get this straight: I’m not the greatest threat to either of them, but the Gargadon is going to be coming my way (promised not to come at Oros, can’t get through Rhys’ tokens), and there’s now a 50% chance that the Curse will hit me, even though Bosh has twice as many lands as me and has proven that he can do crazy things with lots of mana and a recurring Projectile of Ish Sah? Sure, that makes sense.

Fortunately I had a Hideous End in hand for the Gargadon,[4] but that could have been ugly – neither player would have used their strongest cards against their biggest threats, and a weaker player who was a) not an immediate threat, and b) a potential ally[5] would have been put at death’s door. Ignoring the issue of fairness to a player who might get crippled or killed purely on the basis of political wheeling and dealing, these types of deals often don’t serve the interests of either player.

Let me make one final point about this. I assume everyone knows roughly what the Prisoner’s Dilemma is – it’s a common scenario in game theory which illustrates how players can make bad decisions when they don’t know what other players will do. For example, imagine a three-player game where Andy can tap out to kill Bill, but if he does, then Carl will kill Andy and win the game, while Bill can tap out to kill Andy (but if he does Carl will win). Also, if Andy and Bill both attack Carl then they can kill him, and there is no other way to stop Carl from winning. In that case, neither Andy nor Bill can win the game unless they attack Carl, but Andy will be afraid to take the first step against Carl, because that leaves him vulnerable to Bill. In this type of situation, politics would solve the problem of uncertainty, allowing Andy and Bill to agree to take out Carl and then fight it out between themselves; through politics, both players increase their chances of winning the game.

Politics is also…?

Thinking about politics as revealing hidden information takes care of the stuff that is clearly different from a duel – without explicit cooperation, sharing and other forms of PDA[6] the strategic environment is nice and cut-throat. It also allows us to limit the role of politics (by agreeing to keep hidden information hidden if some players insist) while still leaving lots of delicious complexity for us multiplayer fans to enjoy. Perhaps most importantly, it lets us show people that we beat them fair and square with superior strategy, rather than politics. I for one would be happy to play without politics in my games, or at least to have it appear less frequently than it currently does, but knowing exactly what it is that people are doing when they play their political sub-games allows me to play the game better myself if I need an equalizer.

However, I realize that there are other things that don’t fit into this definition – in particular, your homework question from my last article. Saying “Hey, Phil – you do realize that Tim’s deck has a card that destroys the Elves you have on the table, right?” only counts as politics if the information you reveal was hidden. If Phil has never played against Tim’s deck before but you encountered a Tsabo’s Decree when you played against Tim earlier tonight, then that is politics, and if Tim has a Pestilence on the table then you’re not really saying anything of value.

However, the point of that kind of table talk is still to manipulate Phil into thinking that Tim is more of a threat and you are less of a threat. As someone who has been playing for a while I really enjoy that kind of back-and-forth banter, and it almost never affects my decision making, but it can swing the game occasionally. How do we classify the kind of table talk that merely emphasizes known information?

The only way to deal with the problem is to introduce a new distinction: high politics and low politics.

High Politics is the manipulation of hidden information to increase your chances of winning.

Low Politics is the selective emphasis of known information to manipulate your opponents’ perceptions.

Of course, “high” may sounds good and noble, while “low” sounds dirty and manipulative, but that’s beside the point. I chose the phrase “high politics” to indicate that it is more sophisticated, requires more commitment on your part and is more likely to impact the game, whereas “low politics” is simple, easy and usually has a limited effect on the game.

Low politics might involve reminding someone of life totals (“do you realize that if you attack me I’ll die, because he’s already done 15 damage to me with his Thoctar?”), pointing out a blocker (“don’t forget that he can use Puppet Strings to unblock his Thoctar and block”) and suggesting targets (“Elspeth can activate her ultimate next turn and blow up all of your critters”), but the more attention people are paying to the game, the less impact this will have. If I attack Steve with my Juggernaut and then someone points out that Steve has Shattering Pulse in his hand and I could do more damage by attacking Michelle, then that just compensates for the fact that I am a dumbass and I was probably having too much fun talking to a friend to notice the last time Steve used Shattering Pulse. In that sense, low politics serves to make multiplayer more like a duel, because in a duel people are more likely to be aware of all of the known information.

I also want to emphasize the difficulty in distinguishing between low politics and general socializing. Last week a friend played Keening Stone (this card flew right under my radar when ROE was released – it is a house, even in EDH) and I commented that Brian was the only one who was safe from it, because he had Kozilek in his deck and K-Bot beats millstone. Now, I had recently searched Brian’s library with Thada Adel, so I had seen Kozilek recently, but we’ve all seen Brian play him in that deck before, so I wasn’t revealing any new information to anyone else. Was I trying to manipulate anyone? No, because if the Keening Stone player realizes that Brian is a bad target, that increases the probability that he will target me instead, and for once I wasn’t playing a reanimator deck, so that would have hurt me. In effect, I was saying something that everyone knew, which either wouldn’t have any effect on the game, or would have a negative impact on me. If sarcasm worked online I’d say “Oh no, the horrors of politics,” but as it doesn’t, I’ll just leave it there.


Strategy in multiplayer is much more complex than it is in a duel; that complexity is hard for some people to deal with, but many of us find that it makes multiplayer game much more challenging and enjoyable than duels. However, the existence of so much more hidden information, and the increasing importance of hidden information about your intentions, enables politics to play a role in multiplayer games that it doesn’t play in duels. Politics is only one of many factors that influences the outcome of games; there are very few games where politics determines the outcome, and even fewer playgroups where politics regularly determines the outcome of all or most games. Moreover, politics is likely to play a greater role in those playgroups because those players actively enjoy that aspect of the game. The bottom line is that the role of politics in your playgroup is something for the group as a whole to decide.

Now that we can understand high politics as the manipulation of hidden information, I would argue that it doesn’t have to play a role in any particular game. If people don’t like that sort of thing, it is easy enough to cut out high politics in any group, which could actually have a beneficial effect by encouraging players to think more deeply about their actions. On the other hand, politics isn’t all bad. It does allow people to overcome some of the thornier problems of game theory, such as the Prisoner’s Dilemma or Stag Hunt, and it can help to break a deadlock and end a painfully long game. High politics plays a role in multiplayer, but it doesn’t need to be a decisive one.

Low politics is a different animal. If it gets out of hand, then it becomes a case of back-seat driving – there are few things worse than having one loudmouth point out every mistake that every other player makes because they’ve forgotten or overlooked some piece of known information – but it is in most cases harmless. If you find that you often change your plans when someone points out something you should have known, then perhaps you need to pay a little more attention to the game state, but on the other hand if someone frequently overlooks known information then perhaps you should allow them the freedom to make mistakes rather than constantly correct them (unless they’re new to the game, in which case low politics is virtually indistinguishable from teaching). Unlike high politics, low politics has a place in every game, because attempting to ban low politics will cost you a lot of the social dimension that makes multiplayer so enjoyable.

[1] Or at least it does not play a significant and consistent role in duels.

[2] “Opponents” in this case means people who are trying to beat everyone else, just like you. The strategic environment is obviously very different in an Archenemy situation where all of your opponents are totally focused on killing you, and while it may be harder to win in that situation, the actual strategic complexity is greatly reduced. With only one opponent to eliminate, decision trees are much simpler than traditional multiplayer formats.

[3] Almost. Recently I won a sealed duel by realizing that I didn’t have enough Infect creatures left to win with poison, and had to win by regular damage instead. Because my opponent didn’t realize that my intended path to victory had changed, I was able to get him to waste removal and blockers on my few remaining Infecters, which left him open to my real wincons.

[4] Full disclosure: I was so flustered by that negotiation that I didn’t even think about whether I could deal with the Gargadon or not. As I recall, I started sobbing “Why would you do that?! Don’t you like me?!!” and then cowered under the table shrieking inconsolably for about half an hour, and didn’t realize that I had the answer until after I’d come back for air. Do as I say not as I do, and keep your head in the game.

[5] Ally in the purely strategic sense that our interests would converge and they could benefit from acting in concert with me,

[6] Provisional Diminution of Aggression

PS: This is totally unrelated to politics or strategy, but I’m here to help. I recently bought my first Gaea’s Cradle – it was on sale for about $30 bucks and I wondered if that was a good price for a card I’ve always wanted. I checked the price on my phone and was shocked to discover that Star City had sold out of them at $60 a piece!

I don’t know why this happened, but I know it was much closer to the $30 mark last year. If you have the opportunity to pick up a few copies of this multiplayer classic at the old price, then I suggest you do so.


About Graveborn Muse

Daryl Bockett has been an avid Magic addict since Legends/Revised. He lives and breathes deckbuilding and casual play. "The more the merrier" is his creed! In those brief moments when he isn't playing, reading or thinking about Magic, he teaches at Rikkyo University in Tokyo. He has a Ph.D. in International Relations, which is basically only useful for helping him to understand the strategic interactions at a multiplayer table.
This entry was posted in Graveborn Musings. Bookmark the permalink.

8 Responses to Graveborn Musings – Politics, Complexity and Multiplayer Strategy (Finale)

  1. MrShine says:

    I just wanted to say that I’m really glad you managed to reach a clear and well thought-out definition of politics in Magic – no easy feat! The inclusion of, and separation between, levels of known information in politicking is essential; only by using our own power of communication are people able to share thoughts and strategies regarding unknown (or potentially unnoticed) information. This makes ‘politicking’ an invaluable tool when making tough decisions in a multiplayer game, and as such politics is always going to be relevant in any multiplayer game.

    Basically, I think you it the nail on the head 😉 Great article!

    • Graveborn Muse says:

      Thanks Mr. Shine! I was thinking about how to fit what you said in there, about it being out-of-game factors, but i wasn’t sure how to fit it – I think the category is too big to define. Certainly some of that stuff belongs. On the other hand, attacking somebody just because you don’t like him is kind of like the absence of strategy rather than anything else. Glad you liked what I came up with – I hope it’s a starting point for more discussion.

  2. Matthew Banister says:

    Ugh…this article was terrible. And not for any reason the content structure and points were fantastic. However, it is finals week here at my school. I have taken an Urban Politics exam, an International Organizations exam, I am writing a 20 page paper on Russian Military Reform After Communism, and have a Russian Politics exam later today. I currently hate everything to do with politics right now. Thats not your fault though lol.

    So again, actually a fantastic article, just the word politics may drive me to a breaking point soon lol.

  3. Estel says:

    I would say that politics is less about ‘hidden information’ (which has an in-game denotation that muddies things, I think) and more about letting your opponents in on context. Defining politics as the revealing of hidden info for strategic gain draws an arbitrary line between what is obvious and what is not. Politics involves making players see the appropriate (for you and not them, perhaps) context of a card or action. While a card like Pestilence obviously threatens to hose an Elf strategy, the presence of Future Sight and Helm of Awakening on the board might be a more subtle threat that requires the correct context to understand (the existence of Sensei’s Divining Top). A masterful politician will weave a narrative out of these contextual clues, one in which his opponents’ interactions loom and threaten, while his own deeds are simply the fun-loving durdlings of a playfully purple hippo.

    • Graveborn Muse says:

      @Matt – sorry about that!
      @Estel – Fair point. To be honest, any attempt to define something like this is going to involve drawing somewhat arbitrary lines somewhere – I think your term “context” may be separated from both known and hidden information by such a line, although I’d like to hear you define it more clearly.
      Constructing a narrative out of a given context/game state, is as you say the work of a master politician, although I would hesitate to define it as the whole thing. However, that could be interpreted as speculating about hidden information…we don’t know what the Future Awakening player will do if it’s a new deck, but constructing a narrative about what cards might be coming up is certainly related to the manipulation of hidden information, wouldn’t you say?

      • Estel says:

        Context as I mean it is the meaning or importance of something when viewed holistically through its surrounding circumstances or associations. That ‘something’ can be a card, a play, a series of plays, a player’s personality, what that player has been saying or not saying, etc. Offering context can mean presenting a falsely neutral fact (You *know* he has X, right?) or pitching in your own analysis (He’ll beat us all in the long game, look at all that recursion!). Context can be communicated nonverbally through game actions (Revenge-type play in order to discourage future attacks).

        “I bet you he has Sensei’s Divining Top in there and will combo us all.”

        In my Future Sight-Helm example, the politicking player (furthermore P1) doesn’t have the authority to make a claim about the Helm player’s hand, library, or intentions. He actually reveals no hidden information to the pawn (P2), and in fact hasn’t the appropriate hidden information to reveal. I don’t think that the sort of information conveyed in P1’s speculation could be fairly called either hidden or known. P1’s comment is not about communicating information, really. It’s what the language philosopher J.L. Austin would call a “performative utterance.” It’s neither true nor false, but acts as a kind of invitation to speculate for P2.

      • Graveborn Muse says:

        That’s very interesting. You’ve certainly succeeded in problematizing the concept of ‘hidden information,’ but I would offer three points.
        Firstly, “context” doesn’t relate to the less sophisticated (and more annoying?) types of politics, like a player who has an Icy Manipulator out and asks “Who are you going to attack?” at the start of your combat step. You either have to commit not to attack them or take the chance that they’ll neutralize your biggest creature even if you were planning to attack someone else. I zeroed in on the manipulation of hidden information as the core of High Politics because it includes this type of thing as well as the more subtle machinations that you talk about.
        Secondly, I used the word “manipulation” rather than “revelation” because it isn’t always about showing cards or saying things that are 100% true. In the Icy Manipulator example, I have to give the Icy player hidden information about who I plan to attack. If I say “Don’t worry, I’m going to attack Daisuke” then they will let me move on to the Declare Attackers step, at which point I could attack Icy anyway, albeit at the cost of never being trusted again. In your example, the SDT is still hidden from P1, but suggesting the existence of the combo to P2 will further P1’s chances of winning. It isn’t accurate to say that P1 doesn’t have the authority to make that claim; P1 can suggest that one of the hidden cards in Future Sight’s deck is an SDT as a political gambit. How successful they are in manipulating P2 will depend on a range of other factors, but it still involves the manipulation of information that is hidden from both P1 and P2.
        Thirdly, I’m happy to acknowledge that the line between hidden and known information is a little bit blurry – so is the concept of utility in economics or power in international relations. However, I think I can clarify the examples you gave. If P2 really knows that P3 has a certain card, then saying “You *know* he has X, right?” fits my definition of Low Politics, while the future is unknown, so saying, “He’ll beat us all in the long game, look at all that recursion!” is High Politics. It’s primarily when the play is visible (“If he untaps with that Balthor he just cast, he can get back all of his creatures and kill us both – look in his graveyard if you don’t believe me!”) that it would become known information. For example, I might hope that P2 kills the recursion player so that I can hit P2 with Bojuka Bog, and then cast a one-sided Living Death – that would be manipulating the fact that the Bog and LD in my hand are hidden from P2. Finally, avenging an attack to deter future attacks seems to be primarily strategic, not political, unless it’s accompanied by a specific promise.
        I look forward to hearing your thoughts.

  4. Pingback: Graveborn Musings – Dynamic Threat Assessment | Muse Vessel

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s