Deck Taxonomy

[MGB]

If you want to classify decks, it seems that the most common starting point is by win condition. There are lots of ways to classify decks, but most people gravitate toward this one.

I like to view decks win conditions as a binary category: Accrued vs. Non-Accrued. By this I mean: does the deck most want to fulfill its win condition(s) over the course of multiple turns or in one turn? Which is the *primary* mode of victory (i.e. which happens most frequently)? And which of the four (Life total, Poison Threshold, Cards in Deck, Explicit "You Win" effect) objectives is this deck seeking to satisfy to win the game?

For example, I would classify Storm combo as a Non-Accrued Life Depletion deck. In most cases (98%+) it wins by depleting opponent's life total with a Tendrils of Agony in one turn. Or it will generate a ton of Goblin tokens and win in one attack phase (maybe 85% for this win condition, and maybe 90% in two turns or less). It *sometimes* wins over the course of multiple Tendrils or multiple attack phases, but by far the most common pattern of winning with this deck involves a single burst of life depletion.

Any "aggro" deck is an Accrued Life Depletion deck. You win, typically, over the course of multiple turns through combat damage and/or burn.

Any "mill" deck (non-competitive but for the sake of argument) is an Accrued Deck Depletion deck. It treats decks in the library in the same way that an aggro deck treats opponent's life total. A combo deck like Imperial Painter, however, is a Non-Accrued Deck Depletion deck. Instead of incremental milling, it is done in one shot. Occasionally that type of deck will win through the attack phase, but the primary goal is a single activation of Grindstone.

Control decks like Miracles are trickier to classify because they can win through Jace ultimate, or through Snapcaster / Batterskull / Clique attacks. The more heavily controllish builds I like to classify as primarily Non-Accrued Deck Depletion / Life Depletion because the main goal is to either build toward Jace ultimate and win in one turn, or to build toward a massive Entreat the Angels and win in one attack. The important thing to note here is that the natural mode of the deck is to win in this manner. A control deck doesn't *want* to win through a steady attrition of opponent's life total. A control deck wants to know that the game is won through control of resources / board state and that actual winning is a formality. In this respect, the desire of the deck is a Non-Accrued win condition even if in practice it has to attack for 3-4 turns to formally seal the victory.

Of course, there must be other categories by which we can classify decks, such as whether or not they are active or reactive in regard to secondary resources like Card Presence and Tempo Presence... but the first category by which I would classify them is through their win condition, and in this manner.

[Smmenen]

I view Magic as a game of resource management where the resources are split into the following groups: Primary and Secondary, and there are three types of advantage to be gained over the course of the game: Card Presence Advantage, Tempo Presence Advantage, and Play Presence Advantage.

Primary resources are those involved in the satisfaction of the objectives of the game. Life, Poison threshold, and Cards in library. The objectives of the game are to win by either a.) depleting life b.) going beyond poison threshold c.) depleting library or d.) explicit win condition. An opponent can concede before any of these goals are met but every game is won either explicitly or implicitly through one of these objectives.

Secondary resources are those more heavily involved in the actual strategy of the game itself.
l

The way that I deal with those distinction is simple. In my Gush book I distinguish between four different types of cards that constitute The Plan:

1) Ultimate strategic objectives (Finishers)
2) Interim or intermediate strategies objectives
3) tactics
4) mana resources

ultimate strategic objectives are the cards that satisfy the conditions established by Rule 104 for winning the game. Intermediate strategic objectives are the ones that most decks pursue, to the extent that they aren't actually the win conditions themselves. Tactics are the resources you use to support your strategic objectives and disrupt the opponent's pursuit of their own.

Categories 1 would correlate to your "primary" objectives, and 2-4 woudl be "secondary" in your rubric.

If you want to classify decks, it seems that the most common starting point is by win condition.

I disagree with this. Many decks can share win conditions, but that doesn't tell you about their strategy. For example, Control and combo decks in Vintage may use Tendrils of Agony. But the combo deck may use Tendrils aggressively, whereas the Grow deck uses Tendrils as a secondary win condition. That really doesn't tell you anything about how central it is to their game plan.

active or reactive i

I don't believe active and reactive is a tenable distinction, when closely inspected.

[MGB]

I disagree with this. Many decks can share win conditions, but that doesn't tell you about their strategy. For example, Control and combo decks in Vintage may use Tendrils of Agony. But the combo deck may use Tendrils aggressively, whereas the Grow deck uses Tendrils as a secondary win condition. That really doesn't tell you anything about how central it is to their game plan.

I agree that a more fine-grained definition is necessary beyond simply categorizing win condition, but I believe it to be a good starting point. Eventually, I want to develop a system that categorizes a deck with regard to its approach to both Primary and Secondary resources (i.e. Accrued Life / Active Tempo Presence / Reactive+Active Card Presence).

Also, I disagree with you regarding your statement that Win Condition classification "doesn't tell you about their strategy". You are saying it yourself that the combo deck uses Tendrils aggressively, while the Grow deck uses Tendrils as a secondary win condition. That tells you quite a bit about their strategies. The Combo deck is first and foremost a non-Accrued Life Depletion deck. In my system, I categorize by what the deck wishes to do in an ideal winning scenario. Every deck can have a secondary win condition, but what differentiates most decks is the primary goal. And the primary goal informs the entirety of the rest of the deck. All other card choices are filtered through the lens of the primary objective. If Tendrils was merely a secondary goal of the Combo deck, more cards would be devoted to support of satisfaction of a more Accrued Life Depletion game, and thus the deck would take on a different character maybe featuring more Dark Confidants and other creatures, and would thus be shifted in its taxonomic status.

I don't believe active and reactive is a tenable distinction, when closely inspected.

How else can we make a distinction between a card like Thalia, Guardian of Thraben... and a card like Nightscape Familiar? Both affect tempo in a similar manner, but one does it while interacting with the opponent's pool of tempo, and the other does it while interacting with its controllers'. Is there no way at all to make a meaningful distinction between these two cards? Can you suggest something better to label these distinctions when discussing cards?

[TsumiBand]

Exactly. A problem which is further compounded by the fact that decks do more than one thing .

Consider Aggro-Control. An Aggro-Control deck can play a tempo game against Control, but it can play a pure or nearly pure control role for a long time against Combo.

In the combo matchup, it is the Control deck. In the control matchup, it is The Beatdown, to put it in "Who's The Beatdown?" Terms.

I pointed this out in my first post in this thread, but people truly don't understand "Who's The Beatdown?" on several levels.

First, they think it means that it defines a deck's strategic orientation in general. This is not true. Who's the Beatdown is a *purely * contextual question: it asks: IN THIS MATCHUP, what is my optimal role?

Second, and this is even more subtle, "Who's the Beatdown?" was framed in a matchup of SIMILAR strategies, i.e. two aggro decks or two control decks.

This second point is especially important because it clarifies that "who's the beatdown?" does NOT help you classify a deck or an archetype. Remember, it doesn't tell you who is a control deck among two control decks. It simply tells you which deck should play the control role in that matchup.

Yeah, that. This is probably why I favor guys and burn or aggro-control and why I've long loved Legacy for permitting (whether falsely or correctly permitting is up for debate I suppose) aggo-control strategies to flourish. While people were still arguing that the only good creature decks in Vintage were Fish and UR Fish, Legacy had this whole range of decks from Goblins to Madness to Zoo to Angel Stompy that had all these moving parts that could end the game or control the game in different ways. It's not *easy* to play strict control with something on a par with, like, RG Kird Ape beats, but I love it for that - if push comes to shove the deck does not have to be 'the beatdown'.

Do people really get perplexed by the concept of WtB? I thought it was clear as day; you can ask it on a per-match, per-game, per-turn basis. And you should do, in particular when playing Guys and Burn Spells; you probably don't have robust draw at your command, so each spell matters a ton - if you're not racing when you should be, you're fucked. And if you're trying to play first-to-zero when you should be playing the attrition game, you're fucked. The tools are simple, but the decisions are more complex than that. I think that's why I have a difficult time accepting tempo as anything but a residue of design and potential for interaction, because as a Guys and Burn deck will quickly point out to its pilot, not being able to play stack games sucks - no wonder RUG is a thing right? Burn + Counters + Guys = ways of generating tempo in every major zone in the game.

[HammerAndSickled]

Classifying decks by their method of winning seems asinine. Functionally, in a game where your opponent is playing combo, it really doesn't matter whether you're dying to Tendrils or Brain Freeze or some Omniscience bullshit or whatever: the fact is the method of interacting with all of those decks is 90% the same (counterspells, discard, maybe some Stax effects, mana disruption) and it does not matter in most cases how you actually die. Similarly, losing to control is the same no matter how they actually win. A Jace ultimate or Snapcaster attacks are fundamentally identical, because both mean that you were locked out of the game and any win condition would have been identical. Once you have countertop assembled and the board is clear you could usually win with a Rainbow Efreet. Entreat the Angels flips the script a bit, though, because sometimes they can win from an otherwise losing board state by just sticking an Entreat. This is where your distinction between accrued and non-accrued actually matters: there are probably decks that can win a fair game with Miracles but cannot beat a sizable end step 5 point Entreat, because there is a functional difference between grinding and winning in one overwhelming sweep. But if the entreat had been some hypothetical mill spell or poison spell or even a burn effect, it's not functionally different in most games HOW they actually kill you.

My main deck is Lands, and I play the prison build of Lands. My win conditions are Jace ultimate, Creeping Tar Pit, Dark Depths combo, and recurring an artifact with Academy Ruins so I can't lose to decking. But really, how I win every game is with prison tactics and mana denial. Most of my games end when I establish Ghost Quarter lock or Chasm lock or even just a few wastes and ports. The actual win method rarely matters.

[MGB]

Classifying decks by their method of winning seems asinine. Functionally, in a game where your opponent is playing combo, it really doesn't matter whether you're dying to Tendrils or Brain Freeze or some Omniscience bullshit or whatever: the fact is the method of interacting with all of those decks is 90% the same (counterspells, discard, maybe some Stax effects, mana disruption) and it does not matter in most cases how you actually die.

That's exactly why classifying decks by "Accrued" or "Non-Accrued" win condition *is* illuminating - all of the combo decks you mentioned are trying to win in one turn, and the majority of their deck construction is designed to support that kind of win. By saying they are "non-Accrued" we instantly know that the design will likely feature a centralized plan to amass overwhelming play presence on a single critical turn, while when we say "accrued" we know instantly that the deck is likely de-centralized and full of redundant individual parts.

By bringing up Tendrils / Brain Freeze / Omniscience you are just reinforcing how clear and simple this classification can be.

Similarly, losing to control is the same no matter how they actually win. A Jace ultimate or Snapcaster attacks are fundamentally identical, because both mean that you were locked out of the game and any win condition would have been identical. Once you have countertop assembled and the board is clear you could usually win with a Rainbow Efreet. Entreat the Angels flips the script a bit, though, because sometimes they can win from an otherwise losing board state by just sticking an Entreat. This is where your distinction between accrued and non-accrued actually matters: there are probably decks that can win a fair game with Miracles but cannot beat a sizable end step 5 point Entreat, because there is a functional difference between grinding and winning in one overwhelming sweep. But if the entreat had been some hypothetical mill spell or poison spell or even a burn effect, it's not functionally different in most games HOW they actually kill you.

My main deck is Lands, and I play the prison build of Lands. My win conditions are Jace ultimate, Creeping Tar Pit, Dark Depths combo, and recurring an artifact with Academy Ruins so I can't lose to decking. But really, how I win every game is with prison tactics and mana denial. Most of my games end when I establish Ghost Quarter lock or Chasm lock or even just a few wastes and ports. The actual win method rarely matters.

That's why when we classify control we also use the term "Non-Accrued"... because the incremental gains made by control decks are not the kind that a deck chipping away at life total is making. Most incrememental gains the control deck is making is in the realm of Card and Tempo presence. The win condition, like in combo, is ideally a combo kill, but can sometimes be a manland attacking. But what does the control *want* to do? It wants to establish control of the board and making the actual process of winning a foregone conclusion. Thus, the ideal state for the control deck is to have a Non-Accrued win condition.

In a deck like Miracles, ideally they want the same thing. Just because some games are won with multiple attacks of Snapcaster or Entreat tokens doesn't change the fundamental goal of the deck - to make the actual process of formalizing the win a formality.

[HammerAndSickled]

You misunderstand: I actually like the concept of accrued/non-accrued advantage. My point with Entreat was to illustrate that there is a real difference in play between the two styles. I just think that classifying it as life depletion/deck depletion is utterly meaningless. It doesn't change how the deck plays at all. You're saying nothing meaningful about a deck by saying how it actually closes the game.

[TsumiBand]

You misunderstand: I actually like the concept of accrued/non-accrued advantage. My point with Entreat was to illustrate that there is a real difference in play between the two styles. I just think that classifying it as life depletion/deck depletion is utterly meaningless. It doesn't change how the deck plays at all. You're saying nothing meaningful about a deck by saying how it actually closes the game.

I think it's got potential as a way to describe a rudimentary counter-strategy, but that's definition by inference I guess.

Like, as a surface level observation - life loss in increments can be addressed by incremental life gain. Exalted Angel vs. a 7/7, both swinging for the opponent's life totals, is a win for the Angel player, not the 7/7. It's dead simple math:
Player A@20 life: attack for 7 (opp 13)
B@13: a4 (opp 16, self 17)
A@16: a7 (opp 10)
B@10: a4 (opp 12, self 14)
A@12: a7 (opp 7)
B@14: a4 (opp 8, self 11)
A@8: a7 (opp 4)
B@4: a4 (opp 4, self 8)
A@4: a7 (opp 1)
B@1: a4 (opp 0). Win.

It's interesting, actually, because it isn't always enough to say 'aggro loses to X', because in the above situation, Exalted Angel wins out over time. Decks like Goblins or Elves! might take the 'burst' route and nix that line of play, but their ability to do so shouldn't be the deciding factor in whether or not the deck *is* aggro or combo. Any given aggro deck can swing for a player's entire starting life total if left unanswered. It isn't their primary means to an end, but it exists within the realm of occurrence.

Also, not trying to horn in on anyone's nomenclature but IMHO Incremental is more intuitive than Accrued, and an antonym would be better than the 'non-' prefix, but I'm not a technical writer or anything. :)

sent from phone, don't be a dick