The cunning ruse: Deception in game design

Though I was largely disappointed with Watch Dogs, the thing I like to salvage and look back on fondly is the multiplayer, an interesting interpretation of Dark Souls invasion system, focusing more on deceiving the player you’re invading than slaying them. In particular, I enjoyed the tailing assignments the most, tasking you with blending into the background of their game, becoming just another face in the crowd. If you did it right, they never even knew you were there.

The thrill was partly voyeuristic, of seeing someone else go about what was ostensibly a single-player experience, something that you rarely get to bear witness to unless through streaming, but even there it’s a communal thing, airs affected and commentary provided. Instead here you’re just watching what someone does when they’re there for no one but themselves. It was fascinating.

But most of that thrill came from the actual skill of it, obeying traffic laws and wandering down the pavement at walking pace, trying not to move too erratically or act too much like a player would. It was acting, deceiving someone who didn’t know they were being deceived.

What made it so enjoyable was how refreshing it was. Here I was, interacting with another player, and neither of us were eager to pull out our weapons and introduce one another through the medium of lead. It’s so rare in games like this that player interaction is anything other than violent, that when it isn’t something as banal as driving down the street while tailing someone going about their business becomes engrossing.

“It’s kind of like crash landing on a Polynesian island and you realize that there’s edible fruit hanging from every tree.” Chris Hecker has been experimenting with these ordinary interactions for years during the development of his game SpyParty, which pits one player as a spy at an NPC-attended cocktail party, planting bugs and retrieving microfilms, while another plays as a sniper with a single bullet, watching the party from afar and trying to figure out which guest isn’t just there for the martinis.

“It’s kind of amazing to me because my experience with designing SpyParty is that there’s just game design lying all over the place. When you just put normal people in a normal room things just lend themselves to mechanics. Even putting drinks in people’s hands is a mechanic. It’s the idea that if you’re designing for normal human to human interaction the design risks are much lower. Deception is one of those mechanics. It’s just so unexplored when it comes to making game design. Making a game is hard no matter what, but that said, for the most part there’s a very fertile field. As long as you have a discerning sense of what’s working and what’s not, there’s a lot of new and original things you can do.”

Talking to Hecker, it quickly becomes apparent that my simple enjoyment of actively deceiving another player isn’t nearly as simple when it comes from a design standpoint. It’s also something he’s having to break down and understand on a computational level as he attempts to create a singleplayer component for SpyParty.

“To earn the word ‘deception’ it’s almost as if your opponent has to have a formulated plan, which is just so rarely true when dealing with a computer. It’s like if you bluffed in poker and everyone always folded. That’s what’s happening when you deal with a computer most of the time, you know what’s going to happen.’

“There’s a level of cognition, or at least perceived cognition, for deception to take place. It’s the idea of winning even though you’re losing, and that’s a pretty magical thing. It’s not the same as coming from behind; it’s winning when you have no right to. All these kinds of things require someone at the other end who can appreciate that they were deceived.”

Which is a roundabout way of explaining why deception isn’t a more explored avenue of development. It’s not so much that it isn’t interesting, or worthy, but more that it all but necessitates a multiplayer game.

“There really aren’t that many games designed specifically for multiplayer, and that’s why you don’t see it. It’s just starting to happen now that the internet is an integral part of games. It’s only in the last decade that we’ve been able to assume that people can play over the internet.”

Deception as a game mechanic isn’t really that unexplored, though, so long as you don’t limit yourself to video games (though let's make sure to give a nod to the Spy class in Team Fortress 2). Tricking other players and making them believe something that isn’t true is an integral part of a huge swathe of board games, as well as a natural tactic in anything live action. Games like Werewolf, Resistance, & Mafia are predicated on the idea that some players will out-and-out lie to one another about who they are and what their intentions might be. The premise of Werewolf, where a few players out of the group are murderous and are slowly picking off the ‘innocents’, has even been rendered in the popular Garry’s Mod, ‘Trouble In Terrorist Town’.

“When you play a board game with your friends, you know they’re going to follow the rules.” Alex Austin makes Sub Rosa as Cryptic Sea, a game about shady dealing and betrayal, where players are just as likely to pay you as pull out an AK-47 and gun you down. Players form teams (companies) and try to increase the value of their company through successfully doing deals with others. That said, unsuccessfully doing deals isn't always bad, so long as you get what you came for. Dirty deals, backstabbing and deception are all major parts. “You could easily cheat in Werewolf, but you know they’re not going to do that. But with an online game any sort of exploit people are going to use. That makes it much more difficult to establish a style of play.”

It basically means that you have to limit a player’s options for interaction. In SpyParty this isn’t so difficult; there’s no way for the spy to interact with the sniper directly at all, let alone attack them. But with Sub Rosa it’s a first person game with plenty of automatic weaponry. It wouldn’t be that hard to imagine it devolving into gun fight after gun fight. Austin has achieved this firstly through having a relatively small player base for the moment, but also incentivizing "good" play; being trigger happy will cost you progress, whereas doing a good deal will advance you significantly. So the question then becomes how to weave deception into the mix.

“I think the tricky thing is the balance between allowing deception but not making it so easy that people always do it, and not so hard that people never do it. How do you make it so that there’s balance, so you try to deceive someone and there’s a reward, but it’s a risk?”

Which brings us back to poker, and Hecker’s obsession with it from a game design perspective. “Pokeris an immensely skill-based game, but it’s based on probabilities and projected value, and knowing when to push and when to be conservative.” He explains. “Any rookie can beat a pro for one hand, but over the course of a full game the pro is always going to win. It’s incredibly important in poker that you never have to reveal your hand after you win; the question always has to be there of whether you were bluffing or not.”

What poker, SpyParty and Sub Rosa all have in common is that while deception is a big part of their game design, it isn’t all that’s there. It doesn’t work unless it’s optional, otherwise there’s never a question of whether you’re being deceived or not. Instead you ask yourself the question, and how strongly you are confident in the answer dictates your behavior: Is this person lying to me? If they are, what can I do about it? Before long you’re second- and triple-guessing, and the truth, when it comes to light, comes as a satisfying slap across the face. Of course! How could you be so blind?

“From a design perspective deception is incredibly interesting,” Hecker says, “but it’s also incredibly niche, so you don’t want to base your entire game around it. Only a certain subset of people are even interested in deception as a mechanic, so only recently has the audience become large enough to support a game about deception. SpyParty is a weird game, and the idea of SpyParty 10 years ago just wouldn’t work.”