Oh dear, oh dear. Apparently I failed to get this into my editor's hands a few weeks ago when I thought that I had sent it. I apologize to anyone who expects to see an article by me showing up regularly. I really did have this written in time.

In a couple of recent (or not-so-recent by the time you read this) articles, Josh "One Man Crowd" Bennett has been contemplating the value of archetypes and theory to the development of Magic decks. He has decided that working with higher level concepts, such as "control deck" or "aggressive deck," is pretty much useless if one is trying to build a deck. He would much rather consider specific decks and build to those.

As Josh said in his second article on this topic, I disagree. I believe that there is a place for high-level, or at least higher-level, theory in deckbuilding.

The original article, that prompted my corresponding with Josh, had to do with figuring out just what is a deck archetype. At this point I think that Josh and I agree that the word "archetype" has been used in the Magic community to refer to a variety of levels of deck specificity from the purely abstract (e.g., control decks) to the particular (e.g., the B/U control deck that Chorzon used for States).

Now the question is whether you, or I, or anyone else can build a deck to play using the more theoretical/abstract deck types - or do you need to focus on decks that actually exist and are being played? Josh has clearly dismissed the abstract levels of analysis in favor of the particular and has challenged me to provide a defense of the use of the more abstract.

Let's start out by looking at the preparations for the recent State Championship tournaments. These tournaments were being held only days after Invasion became Type II legal and Urza's block rotated out. This is a classic case where you lack any specific decks to beat and have only theory to guide you.

In my efforts to build a deck for States, I started out at a fairly high level of abstraction. After figuring out what my favorite cards in the new environment were going to be, I put together a deck of 38-40 of them, along with some land to give me a 60-card deck. At this point I haven't done any serious deckbuilding...I've just built a prototype deck out of some cards that I like.

The next step is to start figuring out if the deck will actually work.

Theory one: Mana curve.
Determine if your deck has one. If you play long enough, you will eventually run into some newbie who built his tournament deck out of nothing but four-plus casting-cost monsters. Heck, you may have tried it yourself when you were starting out. But these decks never work because the player is dead before they have enough mana out to cast their monsters. In order to work, a deck has to have a mix of spells at different casting costs. If I don't have something resembling a mana curve, I start rebuilding before my first shuffle. Some folks even build their decks allocating a relatively fixed number of cards to casting cost slots in order to assure a viable mana curve.

Once I have something resembling a mana curve, I can move on.

Theory two: How many things is my deck trying to do?
One, or maybe two things, is optimal. Many of us have tried building a deck full of devious little tricks; these unfocused decks never work. I may not recognize this as a problem as I build the prototype, but I am going to keep it in mind as I goldfish the deck.

Theory three (this is the big one): How does my deck deal with other decks?
How does my deck function if I face an aggressive creature deck? How does it do in response to control strategies? At this point, I do have to bring the theory down a notch. For example, control decks exist in several broad categories... mana denial, discard, and the classic counterspell/Draw-Go, to name the obvious ones.

I have no idea as I write this what will be showing up at States. So as I build and refine my deck I am looking for options that will strengthen me against each of the control strategies I can think of. I also think about how I will handle a creature rush or a Stompy deck.

If my deck obviously collapses under a creature rush, or depends on expensive spells that can be disrupted by mana denial, I know that the deck needs to be reworked. I don't have to play the deck at all... I just need to consider how my deck plays against those higher level theoretical archetypes.

If there are no obvious flaws in my deck, I can go on to the goldfish testing stage of the deck's development. I start and play to the end of five draws (six turns). I then decide where the game would be if my opponent has done nothing and ask: Is the deck developing fast enough to start delivering damage to my opponent by turn 5 or 6 if my opponent does nothing? (Perhaps I should have killed my opponent by turn 5 or 6 if they have done nothing. Am I doing that?)

What if my opponent has a creature rush going? How do I fare then? What about counterspells... how much of what has been developed so far would be disrupted by counterspells? Could I have gotten by on less mana if my opponent were playing a mana denial strategy?

Then I play another five turns and re-evaluate. If I was still safe from a weenie rush on turn 5, would I still be alive on turn 11 - or would I have been overwhelmed? Would my opponent still be alive? How badly would I be disrupted by mana denial or counterspells?

After four or five goldfish testings, any large flaws in the deck should be obvious and I can go about fixing them.

If you are familiar enough with the cards available to you, even if you do not know which cards other players are using, this will bring you to the point of having a solid playable deck. If you have done your theorizing well, you will start out with a tier 2 or 2- deck.

You can take this deck to Friday Night Magic and not make a fool of yourself, you should be able to win some games and even an occasional match.

If you want a tier 1 deck, you need to go to the next step and get less theoretical. You need to consider how your deck plays against specific decktypes that are in the current environment. You need to play your deck against these decks to see how it performs. You need to do exactly the kind of deck development that Josh Bennett is claiming you need.

So I claim that the high level theory will lead to a deck that is playable, but not necessarily tier 1. To get a tier 1 deck, you have to get less abstract about the potential threats you face.

Now that I've explained how deck development might happen for us mere mortals, I want to address the argument that I expect to hear from some of the good deck designers out there that "I don't do high-level theorizing when I build my decks."

My simple response is: yes you do.

For the best deckbuilders, the theory is often so automatic for you that you no longer even notice that you do it anymore.

A lot of learned skills follow a pattern of requiring intense effort when they are first learned to becoming second nature as the skill is developed. This is what has happens to skilled deck builders. They look at a pile of cards and start picking out cards to fill their 1cc, 2cc, 3cc, and so on slots. They look at a card and can decide if it is overcosted for what it does.

There is no way that you can decide if a card is overcosted if you don't have a good dose of meta-theory floating around in your brain. Just look at Llanowar Elite if you don't believe me. Tell me, quickly, is it a good card for Constructed?








Time to think.







No, it isn't.

Why? **

If you can answer this without much effort, you have a good deal of high-level theory floating around in your head. Theory that you no longer have to actively think about to use. It is just sort of there. You might even believe that you are functioning without using such high level theory. But you aren't... you depend on that theory, even if you can't tell that you are using it. You certainly aren't going through multiple decks thinking about how this card does if your opponent is playing a Rebel deck or a Rising Waters deck. You are using your meta-theory to make your decision.

So Josh, it isn't that you, and others like you, don't need the high-level theory. It is that the high-level theory is so automatic for you that you use it without even noticing that you are doing it. This is typical of highly skilled people.

For the low-skilled folks, the good news is that with practice the skills can be developed.

Later,

Michael Granaas

Michael owns shares in Hasbro, Inc., parent of Wizards of the Coast

* - I edited in a couple of places to get the time references correct.

** - In case you are still wondering, nine mana for a 6/6 trampler is just too much when you know that you are likely never going to get to nine mana to cast him without serious mana acceleration. When you know that for the same nine mana you could cast a Kavu Titan as a 5/5 trampler (five mana), a Chimeric Idol (three mana), and this same elf as a 1/1. That's a total of nine potential damage on the board for the same cost, and three threats that your opponent has to remove instead of one. I'm sure that the skilled deckbuilders could add a lot to this example, but I am still at the stage where I have to do a lot of brute force testing to decide if a card is good or not.