(Metagame Archive) Design Vs. Philosophy

By Dave Smith

When I finally convinced Danny to let me abscond with his column (a complicated process that involved plying him with sashimi and oat bars, coupled with the solemn promise to pour weekly libations of cranberry juice to the effigy of his dog, Jojo, that is enshrined on his desk), he imposed two conditions. First, I was not to discuss the infamous light socket incident, since the publicity might negatively influence potential jurors, and besides, he was pretty sure the saltpeter was finally taking hold. Second, I was not to discuss philosophy. “Whatever you do, don’t get all abstract and wacky. Discuss how you made a few Marvel Knights cards, drop in an R&D blooper, make fun of Humphreys. Stick to the formula and you should be fine.”

It is amazing how many situations in life you can navigate effortlessly with just a nod and a smile.


Now, Danny’s worry was not entirely unwarranted. For reasons known only to God and Jeff Donais, R&D is simply teeming with philosophy degrees—Justin Gary, Brian Kibler, and Ben Rubin to name a few. I, showing even less sense and restraint, have a graduate degree. Now I imagine that some of you might be poised to flee in horror before the prospect of a discussion about how the X-Statix loner decks shed light on ethical solipsism, or the relevance of off-curve strategies to Plato’s discussion of the One and the Many. Do not be alarmed, Gentle Reader. The only philosophy we are concerned with here is design philosophy, or perhaps more accurately, design method. I hope to offer you a window into not just how, but to some extent why we work the way we do, and to show you how such reasons and methods impacted Marvel Knights.

First Method: Empiricism, or Design by Development

Many years ago, the first game that Brian Hacker and I ever worked on was an NBA trading card game that was eventually bought by Wizards of the Coast (although they only used one of our mechanics in the game that was released). Our initial version of the game proved to be . . . how shall I say it . . . quite instructive. Day after day, we would get together and conduct elaborate abstract discussions about the game’s mechanics, the precise modeling of certain players, the structure of the rules, and timing. About two months of this went by before we finally took the bold step of generating roughly 30 playtest cards on a bunch of index cards. As we were walking home to try it out, excited and more than a little apprehensive, Brian made a quip about the astounding potential for black comedy if the game turned out to be horrible.

As you might have guessed by now, Brian’s words were completely prophetic. The game was so incredibly baroque and overcomplicated that it was actually quite random. Since game calculation was pretty much in the strict province of angels or supercomputers, you just made whatever play you felt like. After some grinding calculation, the result of that choice was revealed! Getting Karl Malone to post up in the paint, and then turn and fire a fade-away jumper, involved a maze of card interactions that took nearly ten minutes. (Sadly, he missed.) The experience was by turns excruciating and excruciatingly funny. (We ended up throwing the whole thing out—see Second Method below.) It did, however, give us our first insight into proper game design. The only proper way to create something new is by direct experimentation—testing as soon and as often as possible.

Years later, we brought this lesson to bear in the creation of Marvel Knights. As Brian noted in his article, we were interested in developing a number of mechanics that broke up the Vs. combat phase, especially the initiative. Along with concealed (then called enshrouded) and hunter, we also considered a mechanic that had a working name like slow or sluggish. Such characters would essentially only have a chance to act after all non-sluggish characters (both yours and your opponent’s) had an opportunity to attack.

Once we had a rough idea of which mechanics we wanted to try, we generated a bunch of placeholder cards to test them. There were about 40 cards in all, I think. We then proceeded to play a number of games with just these cards to see how the mechanics felt.

We pretty much scrapped sluggish in the first few hours. Not only did it generate hideous rules problems, but the play feel was atrocious. In order to justify a sluggish card’s inability to act in the normal combat sequence, we had to compensate the characters with much better stats or abilities. Because of this, the game tended to revolve entirely around stunning them.

Hunter and enshrouded, however, seemed interesting enough to try out on some real characters in actual Sealed Pack. We generated an initial file of commons using simple plot twists, and then ran a bunch of drafts to see how it played out.

Perhaps my favorite thing about UDE’s design process is the thorough integration of design and development. One of the most shocking things I discovered when talking to game designers who work for other companies was the fairly strict isolation of the design and development teams. One company went so far as to institutionalize this territoriality by having the designers include a document specifying their design intentions with each card. Who knows . . . perhaps the vision of those designers sweeps high and low across the plains from the heights of their ivory towers, taking in all before them. As for us, we’re stuck with empirical trial.

Second Method: Mercilessness, or the Importance of Not Falling in Love

Just as important as testing every step of the way is the importance of not falling in love with one’s creations. As I mentioned earlier, Brian and I threw out our first game, a labor of over two months, in its entirety. We also threw out the second and third versions. One of the biggest dangers when one is creating something is becoming too attached to the pieces and being unwilling to sacrifice them, even if experience reveals that they are not the best means of attaining your goals. Another way of putting the same point is that if you are going to test, you need to be able to follow through on the results of your testing. It is psychologically very easy to fall into either the Scylla of a sunk investment or the Charybdis of favoritism. As Nietzsche would say, a creator must be hard.


This method played out in several ways in the generation of Marvel Knights. As others have noted, we threw out one of our major mechanics (hunter) fairly deep into the design process. Hunter involved characters being able to stun exhausted characters directly. Hunter was an exceptionally flavorful mechanic for the dark themes that pervade Marvel Knights, in that it allowed us both to model vigilantes like Punisher and assassins like Electra in a very compelling manner. It also fulfilled one of our major goals for the set, which was to break up the initiative. It even resulted in some very neat tactical exchanges on occasion. However, testing revealed that it was too absolute, and as Brian noted in his article, too “swingy.” After a couple of attempts to rework it, we decided to remove it in toto. Not only did this entail a major rewriting of the card pool, it had an even greater impact on the Crime Lords team, which up until that point had been hunter specialists. I think that a few of the playtesters were a bit shocked with the ease and rapidity with which Brian scrapped what had, until then, been one of the cornerstones of the set. But if you had the right method and the right mindset about what we were trying to achieve, it was actually a pretty painless choice, especially since we had never fallen in love with it.

Personally, the main way this method affects my process is card generation. My general strategy was simply to generate a slew of cards—as many as three to five per character or plot twist. Some of these were junk, of course, and some of them I was pretty excited about when I made them. Some, I’m sure, fell into both classes. Many of them suggested entire families of cards that could have similar mechanics. Two good examples of this are the Underworld discard pile manipulation and the Crime Lords defender switching. I would then ship these cards to Brian, who would subsequently select some, rewrite and combine others, and most frequently just discard my suggestions. I am certain he went through much the same process with his own card designs. Even more would get weeded out in development, and then the whole process would start over again on a smaller scale.

Were there innocents on the casualty list? Cards that should have been allowed to plea for clemency, for a second chance?

Almost certainly.

But so what? Whatever the cards may try to tell you, there are always more where they came from.

Another word for the Second Method, in case you’re interested, is faith


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

%d bloggers like this: