facebook

CoolStuffInc.com

Star Wars: Unlimited Spark of the Rebellion available now!
  
Account Meeple
Location IconLocations
Account MeepleLogin
0 Item
Advanced Search
   Sign In
Create Account
Main » Articles » Card Selection in Deck Building

Card Selection in Deck Building

Chingsung Chang
Posted on March 22, 2011
 Reddit

The most important thing to remember during deck construction is the fact that every card in your deck must serve a strategic purpose. I've mentioned this a few times already, but let's take a look at it in more detail. The big question is as follows: How do you go about deciding those cards and their purpose? There is a simple three-step process that forms the basis of deck design.

STEP 1: Determine your deck's fundamental strategy and necessary cards.

STEP 2: Categorize your role-players.

STEP 3: Select your role-players from the available card pool.

Step 1: Determining Your Fundamental Strategy

Stage theory is a good tool for this process. Because a game of Magic, in particular a deck's game plan, can be divided into three stages, your deck needs to have a plan for all three, even if that plan is "ignore the opponent" or "pray it doesn't happen." You develop this plan by determining how your deck plays out ideally. Usually, this strategy is relatively simple. As examples, let's take a look at two Legacy decks, B/U/G Landstill and Zoo (reference deck lists are in the Appendix).

Zoo:

STAGE 1: Play some creatures to attack with; attack with those creatures.

STAGE 2: Use burn spells to finish off an opponent, either by removing pesky blockers or by sending them straight to the dome.

STAGE 3: Pray it doesn't happen, or that you top-deck the last necessary burn.

B/U/G Landstill:

STAGE 1 and STAGE 2: Leverage your defensive spells and card-advantage engines. The goal is to attack your opponent's board position and to generate incremental card advantage.

STAGE 3: Leverage your card advantage into a winning position by deploying a trump.

A model such as this is going to serve as the basis for your card selection. Each card you select has to fulfill a specific role in advancing your game plan or fundamental strategy.

Step 2: Categorizing Your Role-Players

This step is the most involved of the three, and the one you should spend the most time on. The first main categorization is offense vs. defense (or offensive vs. defensive). The dividing marker for this type of classification is whether a card slot is designed to hinder your opponent's progress or advance your own progress. Cards that hinder your opponent's progress are defensive, whereas cards that advance your own are offensive. It is important to note that the same card may have offensive and defensive applications, but generally one is more important than the other, and that is where those types of cards should be classified.

Example two-way player: Mana Leak

Offensive application: Stopping strong defensive spells like Day of Judgment or critical removal spells.

Defensive application: Stopping important threats like Primeval Titan, Sword of Feast and Famine, and Jace, the Mind Sculptor.

The most important thing to remember when doing this is that "offensive" cards advance your primary game plan. Thus, in a deck like Ad Nauseam Tendrils or Pyromancer Ascension, Preordain, Ponder, Brainstorm, and Echoing Truth are "offensive" cards. The fact that they can find defensive cards is gravy.

For decks that are at both extremes (fast combo, aggro, and heavy control), having as many nonland slots devoted to either offensive or defensive cards is ideal. This is one of Jace, the Mind Sculptor's biggest strengths. While he is primarily an offensive card in most control decks, he still has very relevant defensive applications, and thus limits highly the number of slots most control decks need to dedicate to completely offensive cards.

From here we move on to the second division: typology. Here is where the finer breakdown occurs. You, as a deck designer, need to determine what kinds of offensive and defensive cards you need. Removal for various types of permanents, creatures to attack with, burn spells for reach, mana acceleration, and specific combo pieces like Aluren or Pyromancer Ascension all come into play. For making this type of categorization, math is your best friend.

For your information (these numbers aren't 100 percent exact, but are accurate enough and easier to remember than entire charts):

  • Probability of seeing a 4×: 40% opening hand; +3–4% for the first 8 cards you draw, tapers after.

  • Probability of seeing a 6×: 54% opening hand; +4% for every card you draw, tapers after.

  • Probability of seeing a 8×: 65% opening hand; reaches 85% by 5 cards drawn; 90% after 7.

  • Probability of seeing a 10×: 74% opening hand; reaches 90% by 4 cards drawn; slow improvement thereafter.

  • Probability of seeing a 12×: 81% opening hand; reaches 90% by 2 cards drawn; slow improvement thereafter.

We'll start with offensive cards. There are two types of offensive cards: cards that generate questions and cards that remove answers. Goblin Guide in Zoo is an example of the first category, whereas Echoing Truth from Storm combo is an example of the second. It is important to note that many of the cards that generate questions are creatures, which is especially important for decks that plan on winning quickly through the combat phase.

Defensive cards, on the other hand, are classified by the type of spell they deal with. Creature removal is obviously the most important type of defensive spell, but removal for artifacts, enchantments, Planeswalkers, and even land can be critical as well. This is why counterspells traditionally have been such an important part of defensive control decks. They are nonrestrictive answers to a wide variety of threats. The relative importance (i.e. frequency and speed at which you want to draw them) of each type of defensive category will dictate the number of each type of spell you are playing.

We'll return to our two example decks (Zoo and B/U/G Landstill). Since Zoo is an offensive deck, ideally 100 percent of the nonland slots are going to be devoted to offensive cards or threats. These cards also have to be cheap, because Zoo has to come out of the gates extremely quickly. Thus, 1 to 3 mana is where we will be looking.

We'll begin by slotting Part 1 of our offensive questions—creatures. This is going to be our largest slot. We want all of these cards to be cheap, efficient threats, and we want to draw them. Twenty slots is the accepted number (and a good one) for this application, with twelve of them being one-drops (see math above for why). Part 2 is our burn suite, of which most, if not all, should go both to the face and to opposing creatures. This enables the maximum flexibility for use of our burn suite.

Understanding this requires a discussion of potential "answers" to our strategy. Most "answers" will come in one of two forms—creatures to block our guys and removal to kill our guys. Life gain to pull them out of burn range is a tertiary concern, but one that generally arrives attached to a body (Kitchen Finks, Baneslayer Angel, potentially stuff like Loxodon Hierarch as well). This just makes those creatures particularly effective. Even so, they are still creatures. By having burn that goes both places, we are maximizing our ability to place our burn efficiently. Ideally, we will have as many "Type 2" offensive cards as we have "Type 1" offensive cards above. This means another twenty slots both for face-burn and creature removal.

You'll note that I've identified sixty slots for the deck, and we haven't gotten to lands yet. This is because cards can cover multiple slots. We'll talk about this in card selection. Let's go through the logic for our second deck, B/U/G Landstill.

For B/U/G Landstill, defensive cards are far more important, and that is what we will focus on first. The first priority is creature removal, of which we want about sixteen to twenty pieces (more would be good, but we'll start there). The second is enchantment/artifact answers, of which we will want about ten to twelve. We'll probably want about the same number of answers to Planeswalkers as well. The third category is going to be "situational answers," which will be three to four slots. As a final slot, some ability to smooth your draws would be great. Eight slots here is probably reasonable.

Having taken care of our basic defensive spell allocation, we look to threat allocation. Our main threat is going to be overwhelming card advantage, so we should dedicate as many slots as possible. Ideally, this would be about twenty to ensure consistent threat generation, even if it isn't fast. Now, once again, you'll notice the heavy slotting overrides the number of actual nonland slots you have. Once again, look at card selection to see how this is resolved.

Step 3: Card Selection

So how do you decide what individual card fulfills each individual role? The answer is simple: efficiency. Once you know exactly what every card is supposed to accomplish, you look for the most efficient card at doing the job. It doesn't matter how weird the card is; if it does the job, it does the job. Let's go back to our two example deck lists and see how they fulfill all the slotting requirements.

But what does efficient mean? There are two kinds of efficient: cards that do everything and cards that do one thing really well. Cards that do everything are cards that fulfill every (or almost every) role in your deck. This type of card is like a Swiss Army Knife, which is great because you can use it to fill in holes in your draw. Cards that do one thing well are generally extremely mana-efficient. They provide a whole lot of "bang for your buck," and thus are very good at accomplishing the specific task given to them.

Zoo:

Twenty creatures (12 one-drops): There are exactly twenty creatures on the nose, and a whopping sixteen of them cost 1 mana. All of these threats are extremely efficient damage engines. Grim Lavamancer is the least efficient, but his selection has other applications.

Twenty face-burn: Once again, our Zoo deck exceeds expectations here. four Lightning Bolts, four Rift Bolts, four Fireblasts, two Price of Progress, and four Chain Lightnings make for eighteen burn spells. However, Grim Lavamancer also applies here, bringing the count up to twenty-two total ways to burn face.

Twenty creature removal: four Lightning Bolts, four Rift Bolts, four Chain Lightnings, and four Grim Lavamancers are the main things here. While Fireblast technically also kills creatures, using it for that application is not the primary idea; however, we can't discount the application, so it'll count for half. Doing so, Zoo comes to eighteen cards in this area, which is short of our ideal slotting—but not far, so it's acceptable.

The most important aspect of this card-selection process was the placement of Grim Lavamancer to cover all three slots. The selection of Qasali Pridemage and Sylvan Library are both "answer removal" offensive cards. Pridemage can kill problematic enchantments/artifacts, and Sylvan Library "removes" offensive cards by overwhelming them with threats. You should notice the heavy work done by the burn suite. With that, we round out our quick look at Zoo. Let's turn to the other deck.

B/U/G Landstill:

This was our slotting:

  • Sixteen to twenty creature removal

  • Ten to twelve enchantment/artifact removal

  • Ten to twelve Planeswalker removal

  • Three to four flex answers

  • Twenty offensive cards (most of which will be card advantage)

Twenty creature removal: three Innocent Bloods, two Smothers, one Ghastly Demise, two Cunning Wishes, four Force of Wills, three Counterspells, two Spell Snares, and four Pernicious Deeds are all creature removal. Jace, the Mind Sculptor also counts here, but not for full value, as we are doing things with Standstill. Since his primary function is elsewhere, we'll count him as one in this category. This brings us to twenty-two total creature-removal cards of varying speeds. The faster ones kill a large number of creatures for very low cost, whereas the slower dedicated removal kills multiple creatures.

Ten to twelve enchantment/artifact removal: four Pernicious Deeds, four Force of Wills, three Counterspells, two Cunning Wishes, one Spell Pierce, and two Spell Snares. We are once again over fifteen cards that handle the problem, but this is likely good since these cards will often be used for other purposes.

Ten to twelve Planeswalker removal: four Mishra's Factory, two Creeping Tar Pits, four Force of Wills, three Counterspells, one Spell Pierce, and two Cunning Wishes. We are once again at fifteen slots, which is good for the same reason as above.

Three to four flex situational answers: three Cunning Wishes (Extirpate, Tsabo's Decree, Stifle, Mindbreak Trap, and Ravenous Trap). We're on target here.

Eight draw smoothing: four Brainstorms and three Jace, the Mind Sculptors. We got to seven here, which isn't terrible. I'll take it.

Twenty offensive cards (mostly card advantage): three Jace, the Mind Sculptors, four Standstills, four Pernicious Deeds, two Life from the Loam, one Cunning Wish, four Wastelands, four Mishra's Factory, two Creeping Tar Pits. We have twenty-four slots, of which thirteen generate hard card advantage. The one Cunning Wish is because there is a Forbid in the sideboard, which is one of the primary ways to lock a game down (Forbid + Loam + lands). Life from the Loam + Wasteland, Jace's ultimate, and Standstill + man lands are the other vectors of attack.

Note that the key components to filling the slots in this deck are Cunning Wish and the counter suite. The counter suite contains some of the most efficient individual counters in Legacy (Force, Spell Pierce, Spell Snare, Counterspell; the only one missing is Daze), and it uses them in a variety of different means. Cunning Wish alongside the slots dedicated to it act as redundancy.

Look at the incredible focus present in the two final deck lists. This type of focus is extremely important when constructing most decks. You want every single card in the deck to further your own primary strategy. Zoo has thirty-nine dedicated, offensively oriented cards, whereas B/U/G Landstill has a bunch of two-way players while maintaining thirteen rather dedicated threats—three Jace, the Mind Sculptors, four Standstills, four Mishra's Factory, and two Creeping Tar Pits.

I hope this exploration of card-selection methodology has been useful to you.

Chingsung Chang

Conelead most everywhere and on MTGO

Khan32k5 at gmail dot com


Appendix:

"Zoo"

"B/U/G Landstill"

TAGS articles, theory

Recent Articles

View Archive
Sell your cards and minis 25% credit bonus