Honor, Profit, and Net Damage

Everybody loves the bluffing aspect of Android: Netrunner. If you never "junebug'ed" anyone (or got "junebug'ed" yourself), you're missing a huge fun part of what playing Android: Netrunner really is. So, go ahead and throw yourself into an ambush. I'll wait.

[...]

Ok, now that everyone knows how usings ambushes (often miscalled as "traps" - remember: trap is a piece of ICE subtype, the correct assets subtype is ambush) is important in this game, it is time to understand how these cards are intended to play. I often see people buying terribles bluffs, and even more often I see people including ambushes in their decks that no one that knows the game should be falling to.

I remember that back when I joined the CardGameDB.com team, I had a huge debate with another user who claimed that Runners should always facecheck unrezzed remote servers, no matter what. I tried to explain that diving into a two advancement ambush when you're playing against a "Tag'n Bag" deck wasn't a smart move, but due his misconceptions about game theory, he thought that this was an actual good move.

The Math Behind a Bluff

Today, I want to introduce you the concept of "mixed strategies", but first things first. When we talk about game theory, it is important to understand what is a game for mathematicians. Every struggle between two or more players is a "game" - in our case, the match is not a game, but every click is a separated game that's played over and over and that will eventually stop being played (when someone wins the Android: Netrunner match). The idea of mixed strategies is that you'll "mix" your different options from game to game in order to end up with the best result, on average.

I'll give you an example. Let's say I'm playing soccer, and my friend is taking penalty kicks on me. He has two options, kick left or kick right; I have two options, defend left or defend right (yes, I'm simplifying the problem for the sake of the example). So, what can he do to maximize his score? It is easy to see that, if he chooses one side and kick there everytime, I'll block all kicks - I'll always block the same side too. He has to "mix" his strategies up, kicks sometimes to one side, and sometimes to the other side.

(It can be math'ly proved that the best strategy is 50/50 for both players, but the results itself is not as important as your understanding of the concept, so I won't show you the calculations. If you're really curious about it, google "Nash equilibrium in mixed strategies".)

The Bluff Behind a Piece of ICE

Just like kicking always to the same side is one of the worst strategy my friend can follow, your traps will not work while ignoring your remotes is the best strategy for the Runner. That's why nobody uses these things in a fast advance deck - because a fast advance deck would never "install-advance-advance" an agenda. What you want are those cards that wait for the best moment to be rezzed: like Ronin, Project Ares, and my new BFF, Vulcan Coverup.

It simply wasn't worth to run on a remote server against Jinteki back in the Core Set days because anything that's has two advancement tokens and was not scored is obviously an ambush. Now, it can be a Ronin. Now, ignoring remotes is not the best strategy against Jinteki anymore, because not trashing a Ronin might lead the Runner to a flatline. And that's why it is important to advance your Project Junebug 4 times as fast as possible: because its job is not to pretend to be an agenda, but to pretend to be a Ronin.

When Geoff Hollis said that you're goal as a Jinteki player was to rush to the match point, what he was really saying is that, if you want your ambushes to work, you must force the Runner to take a mixed strategy by manipulating the game and making a mixed strategy (sometimes ignoring remotes and sometimes running remotes) the optimal strategy for the Runner.

"Is that a Cerebral Overwritter or is it a Vulcan Coverup waiting for a Scorched Earth to kill me?"

"Is that an Aggressive Secretary or a Project Atlas waiting for me to install my last program to wipe my rig?"

The Emptyness Behind the Threat

Let's be fair: the old member who argued with me wasn't completely wrong. Sometimes, bad bluffers will try to bluff with things that are not good for themselves - we call that a "non-credible threat". This is a little weird concept to explain if you don't have all the math background required, so I'll break it to make it as simple as possible:

1. People are rational and they'll always make the move that's optimal for them;
2. You threat someone that, unless they do something you want, I will follow the X strategy;
3. Following the X strategy is worse for you than following the Y strategy;
4. You're rational, so you'll always follows Y over X;
5. People know you are rational so you won't follow the X strategy, you'll always follow Y strategy;
6. People won't believe in your threat.

If you have to make a non-optimal moves in order to bluff, you're doing it wrong. Making non-optimal moves, well... is not optimal for you. What you should be doing is to design your strategy in some way that baiting the Runner into a trap happens to be your optimal move.

There are several cases of non-credible threats in Android: Netrunner. A good example would be to use a 2 advancement Project Junebug to threat a Runner with 5 cards in his or her hands when you don't have enough credits to play the Neural EMP that would be necessary to flatline the Runner. The Runner will see that it would be a waste of an ambush, and that it is not the optimal move for you, and will run it.

You might be thinking "hey, but if the Runner gets into the trap that way, it is good for me!". Yes, it is better to you that the Runner gets hit by a trap rather than steals an agenda. But the issue is that it still is not an optimal move, even if it hits, and that's what it makes it a bad move. It is an empty threat not because it is not a threat at all, but because a rational person would not do it (once again, because it is not an optimal move).

Thanks, Von Neumann!

I hope all of this makes sense to you. These game theory concepts might not be very intuitive if you're not familiar with this sort of math, so feel free to ask anything you might not have understood. If you want me to talk about anything in particular, you're welcome to ask it too.

Good luck, good bluffs, and good waiting on the Double Time spoiler.

João “Hraklea” Almeida is a brazilian amateur card game player, the responsible for the Android: Netrunner league in Porto Alegre - RS -, in partnership with Lojas Jambô, and the writer of Root Cause, a bi-weekly series of articles about playing Anarch.