This week, I'd like to talk a bit about multiplayer politics.

Oh no, politics! Why on earth would you choose to talk about a subject like that? It'll get you in so much trouble at work…

I'm not at work.

But it isn't a good topic for parties either! People get so upset and angry; it's just not something you should talk about in polite company!

I have nothing to say about political parties or otherwise. This is about how you interact with the other players in a multiplayer game.

Oh. In that case, carry on, I guess…

Thanks.

Now, for those who only know politics as that evil word that refers to the goings-on in one's capital, behind closed doors with a bunch of guys in suits arguing about how to spend large sums of money, today is about how to interact with your gamer buddies that you sit down to play EDH/Commander with on a regular basis. Whether you choose to wear suits when you do it or not is up to you.

The Theory Behind Multiplayer Politics

First, let's talk a little basic game theory. For those who aren't interested in the theory beyond this, just skip down to the next section. If you like Mark Rosewater's articles on design theory, you'll probably like this too. As I may have mentioned before, I have a psychology degree that focused strongly on motivation and cognition, and understanding why we do the things we do in a game is right up that alley. I wrote my undergraduate thesis on this particular topic, though looking back at it, I spent a lot of time not talking about the right stuff, so today you get the abridged and improved version!

The basic idea we need to look at today is the Prisoner's Dilemma. In short, the Prisoner's Dilemma explains why we don't cooperate with our fellow gamers, even when it's in our best interest to do so.

From Wikipedia:

Two suspects are arrested by the police. The police have insufficient evidence for a conviction, and, having separated the prisoners, visit each of them to offer the same deal. If one testifies for the prosecution against the other (defects) and the other remains silent (cooperates), the defector goes free and the silent accomplice receives the full 10-year sentence. If both remain silent, both prisoners are sentenced to only six months in jail for a minor charge. If each betrays the other, each receives a five-year sentence. Each prisoner must choose to betray the other or to remain silent. Each one is assured that the other would not know about the betrayal before the end of the investigation. How should the prisoners act?

And because everything is easier with a table:

The problem here becomes clear with a little review. If you choose to cooperate with your fellow suspect, you both end up doing time, but much less than if you both choose to defect and sell out to the police. However, if you choose to defect and your opponent chooses to cooperate, you do zero time. Consequently, the theoretical best payoff for you to do is defect and rat out your fellow prisoner. This is referred to as the Nash equilibrium of the system (so named because of the man who discovered it, John Nash) because it's the expected outcome assuming both participants are acting rationally.

However, if both of you do this, you both get screwed and end up doing 5 years instead of zero time. The actual best payoff for the group is for both of you to cooperate, do your 6 months, and get back to your normal lives, hopefully with spending less time in jail in the future. This is known as a Pareto optimum. As you increase the number of players, you actually push the Nash equilibrium farther and farther toward defection for all players because you have to rely on more and more players to take your best interests into mind when participating.

To make it more accurately model our actual multiplayer Magic gaming scenarios we'll treat each game action as a Prisoner's Dilemma scenario, and allow all players to know what the previous person picked in all of the previous scenarios. This is known as the iterated Prisoner's Dilemma. You know what each player has chosen to do previously, and can adjust your choice of cooperate or defect accordingly. Unfortunately, this doesn't actually push people toward cooperating more. The Nash equilibrium still pushes people towards defecting, but the Pareto optimum stays the same – all players should elect to cooperate to ensure maximum benefit for the group.

All this theory makes things look pretty grim, doesn't it? It turns out that the best way to do well in the long run is actually to follow a strategy known as tit-for-tat, which is, in short, doing whatever your opponent did last. In some cases, a tit-for-tat strategy with forgiveness is best, where you don't always immediately retaliate and get stuck in an endless cycle of retaliating. Regardless, it's actually in your best interest in the long run with an undetermined number of iterations to cooperate first and only retaliate if your fellow players prove they can't be trusted.

So, knowing this…

Hey Rob…

Yes?

That's all great, but this doesn't really mean anything to me. What does this have to do with Magic, let alone multiplayer?

Fine, we'll just skip to the good part now, then.

Oh good. I love the good part.

Sigh.

How to Apply Game Theory to Multiplayer

After reviewing all that we know now cooperating has the best payout over time, assuming everyone is willing to follow the same strategy. What does that mean in terms of what I should do when playing multiplayer Magic?

Specifically for Commander, you should familiarize yourself with the "social contract concept". The social contract is a gentlemen's agreement to be a good sport and try and make the game fun for everyone. Sounds a lot like all that game theory stuff I described doesn't it? While you may have fun playing your awesome-turn-4-combo-kill deck, it's highly likely no one else at the table enjoyed the game. I could, of course, be wrong.

There are definitely very competitive players out there who enjoy the Commander format, and if you're in a playgroup full of those players then by all means go for it. However, remember that the majority of people who have chosen to play Commander are not there for that. There are many formats in Magic, and every officially sanctioned format other than Commander is aimed at Spike, because Spike is trying to prove something and the best way to for him to do that is to win. In Commander, the format is not about who finished first, but having the most fun getting there.

My Experiences with the Social Contract

Learning Commander

When I first started playing Elder Dragon Highlander, now known as Commander, it hadn't yet caught on as a hugely popular format, at least in the northeast. I played around with a few deck lists, and around the time Shadowmoor came out I decided I really wanted to give the format a try. I started with Wort the Raidmother and basically extended my red-green mana ramp Standard deck, that I'd been playing in local FNMs for over a year, into a 100 card singleton monstrosity of token generation, big silly spells, and copying them.

I had a good time playing the deck against my competitive friends because they'd built decks of similar caliber to play with, and we were admittedly fairly cutthroat. I built a Jhoira of the Ghitu deck, because I loved the suspend mechanic, and eventually turned it into Sharuum the Hegemon when even my Spike-laden group got sick of playing against it (Even Spikes get tired of turn 3 Darksteel Colossus every other game).

This went on for a bit until my local store closed and I went into a bit of a rut in my Magic career. It's happened often over my 17ish years of playing, so I wasn't particularly concerned because I figured I'd find someone to play more EDH with soon. As it turns out, I made a few friends at work that played casually and absolutely loved the EDH format. I couldn't have gotten luckier, nor could I have come closer to blowing it.

Learning how to have fun playing EDH

My new group was, as some people would put it, "Timmy-tastic". Gone were the decks with 10+ counterspells and board sweepers and combo finishes. I was suddenly faced with decks of extremely variable power level that in some cases weren't even close to the tuned, tournament-tested monstrosities I was used to facing off against.

In short, I felt like I'd brought a bazooka to a knife fight.

I quickly learned that if I kept playing decks like the ones I'd played with my other group, two things would happen:

1. I would make every single game into an Archenemy match without the benefit of extra life or schemes to make up for fighting 3-4 opponents at once. While that's fun sometimes for me because I do like challenges, doing it every game appeals to me about as much as the idea of running into a brick wall repeatedly in hopes of bouncing into orbit.
2. I would stop getting invited to play with this group of guys, because it wasn't very fun for them either.

While I hate the idea of playing bad decks, I came up with a solution: choose restrictions for the deck construction, then be sure to stick with it. For example, my Sharuum deck quickly gained a "must be over half artifacts and no ways to kill the table in a single turn". I made helpful suggestions in as a non-insulting and non-combative way as possible for deck ideas and improvements to the members of the group who showed some interest.

To their credit, with a little practice the guys improved quite quickly in both deck building skills and play skill. But I was still almost always universally considered The Threat, and all too often ended up playing against everyone from turn 1 until I died, or managed to fight through everything. It was getting there, but it still wasn't really fun.

Previously, I'd built a Rafiq of the Many deck that won way, way, too fast, even without counting commander damage. It wasn't really fun for anyone to play against, and no matter what I did for restricting the deck building process it was still over the top if I even made the slightest attempt at building around Rafiq. It was at that point I had an epiphany. It was time to build Group Hug.

Group Hugging for Fun and Profit

For those of you who may not be familiar with the archetype, a Group Hug deck is a deck that has basically no win conditions, or if it does it doesn't play them to win. It gives everyone extra mana, cards, life, and creatures in an attempt to make sure everyone's having a blast playing, generally speaking. I've piloted my Group Hug deck several times, and I think it might actually be more fun for me to try to come in second by giving everyone the most ridiculous board position and resources I can manage than it is for me to actually try to win.

Everyone has a good time and the games are always memorable. I even found a way to win one game (for which I will probably never be forgiven by the one player I actually decked), and the guys in general seem to think the best way to end a good game with Group Hug is to scoop to me for grins. I just laugh and shuffle up for another game.

The moral of the story is by building and playing Group Hug, I've shifted the focus away from myself for a few games, while still getting to participate. Everyone understands what the other players are capable of with different decks, not just that "Rob must die ASAP". It's made things much more fun for everyone, and all it took was me taking a step back and focusing on the fun, rather than the win.

If you find yourself in a similar situation, stop and think about what you're trying to accomplish.  Commander is a format aimed at being a great way for a bunch of social gamers to sit down and have a good time playing a game they love.

If not everyone's having a good time, I can honestly say you're probably doing it wrong.