On Role-players

Every single card in every single deck must justify its slot. When deciding "how many of this card should I play?" the most common response should be zero. Let's take a further look at some basic deck construction theory regarding card selection.

There are, in general, three overarching roles cards can play within your deck:

Completely uncuttable, generally good/useful, metagamed



Uncuttable cards are fairly self-explanatory. Your deck will cease to function as you have intended if you cut this card from your deck. Usually these are cards that your deck is built around. An example of this is Polymorph. Most players have a problem that they feel like too many cards fall into this category.

Iona does not fall into this category. Emrakul does not fall into this category. Neither of these cards is necessary in and of themselves. You can Polymorph into either of them, and that means that you don't need to be running one or the other per say. You can also just as easily Polymorph into Terastodon, or some other big giant fatty. Just because these are the two most efficient options doesn't mean they are the only options. Since there are other cards that fulfill their function (Polymorph target), they are not in the "uncuttable" category. They are in the "nigh untouchable" category due to their efficiency, but "nigh untouchable" is not "uncuttable."

Baneslayer does not fall into this category. It is a win condition. There are many other win conditions available in UW. It just happens to be, far and away, the most efficient one. That doesn't mean you can't cut it for something like Sphinx of Jwar Isle or even Sphinx of Magosi. You can, it's probably wrong, but you can and your deck will still work.

Does this mean you should cut Baneslayer Angel from your UW deck for no good reason? Of course not. It falls very solidly on the high end of the "generally good/useful" category. I just made the point because I feel that people frequently get attached to cards in their decks, and don't realize when those cards stop serving their function as efficiently as possible.

Most cards fall into the "generally useful/good" category. These are highly efficient cards that are good. Some of them are very borderline uncuttable (like Baneslayer Angel), but you have to realize that these cards are not technically uncuttable. There is an important distinction there. They are simply the most efficient way of fulfilling the role you need them to fulfill. Metagame changes may dictate that it becomes more efficient to run something else that fulfills this role, even if such changes are unlikely. Most cards you select for a deck will fall into this category.

Metagamed cards are cards that are good because of certain metagame factors, like being good against the current incarnation of the "best deck." These are generally the cards that change the most within a list, based on the metagame you expect to see.

Let's take a look at some roles that cards play within decks. This list is by no means exhaustive, but it's a good start:

Flagship - These are cards you build your deck around. Polymorph is a good example. These cards are always uncuttable.

Win Conditions/Finishers - This is a category for control decks and combo decks, largely. These decks generally need some way of actually closing out a game that they have achieved their desired game state in. Usually these cards are also not cuttable, but sometimes you have cards that can serve similar functions, albeit more inefficiently. An example is Brain Freeze vs. Tendrils of Agony. Assuming you can cast both, Tendrils is, by far, the better choice, but that doesn't mean you should forget that you can cast Brain Freeze. In fact, Brain Freeze may be better than Tendrils in certain situations/metagames. You have to remember that Tendrils is not actually untouchable, although it is pretty darn close.

Creature Removal - This is usually good for most decks to have, especially in Standard or Extended. Not to say that you have to have it, but it is probably a good idea. Options here are so varied though that you need to analyze what is best for your deck. The most powerful removal spell is not always the best one. Remember that Path to Exile does have a downside, and it is relevant.

Beaters - These creatures are designed to smash face really hard. Usually big bodies are good.

Combo pieces - self explanatory. Sometimes these are flagships (Vampire Hexmage/Dark Depths), other times they can be control components (Punishing Fire/Grove of the Burnwillows).

Sweepers - self-explanatory. Also usually good to have if you are a control deck.

Card draw, mana ramp - also self explanatory.

Hand Disruption - Thoughtsieze and friends.

Graveyard Disruption - Relic of Progenitus and friends. Fairly self-explanatory.

Board control - These are cards that affect the board, but don't actually kill anything. They affect the board by nullifying the effectiveness of certain cards on the table. A good example of a card like this is Ensnaring Bridge.

Mana disruption - This includes Rishidan Port, Stone Rain, and Manabarbs (among others).

Tutors - Once again, fairly self explanatory. Everything from Trinket Mage to Gamble to Demonic Tutor.

General utility - these are cards that serve a variety of middling functions. Usually they are not the best answer to a specific card or group of cards, but they make up for it by being able to answer a much wider variety of cards than the "best answer" cards. A good example of a card like this is Pithing Needle (or Trinket Mage). Counterspells also fall into this category now, although they have not always been here and can overlap into other territories.

Permanent removal (non-creature) - Usually artifact/enchantment removal, although sometimes land destruction/planeswalker removal comes attached. A good example of a card like this is Maelstrom Pulse.

This covers most of the basics.

On a theoretical level deckbuilding is deciding which of these roles is necessary and how much space to devote to each role. If you can get added value by using a card to cover multiple roles you gain redundancy, which makes your deck better. This is exactly why burn is so good in aggressive decks (and why all the best aggro decks are almost invariably red), because it acts as both a finisher and a removal. If you can get a card to fulfill multiple roles, all the better. The way you figure out which roles you need and the balance thereof is by 1) playtesting and 2) keeping up with the metagame (whether you are preparing for your local playgroup or a PTQ there is still a metagame that is shifting; you need to know about it).

Still, a card only needs to fulfill one role to technically justify itself.

Let's move on to a practical example for Standard right now - Spreading Seas (for UW control variants). When this card was first "discovered" it served to disrupt Jund's manabase. Being able to deal with manlands and other utility lands has been added to the function of mana disruption. But, mana disruption is no longer a necessary function and Spreading Seas is not necessary to deal with manlands.

So, let's look at the two functions of Spreading Seas.

Mana Disruption
Jund has become a smaller portion of the metagame, whereas UW and Mythic have become more popular. Thus this card is less likely to have an effect on the opponent's mana. Also, many Jund lists are now running some form of manafixing (Trace of Abundance/Rampant Growth/Borderland Ranger) or just running more fetchlands, all of which diminish the effect of this card. Before there was a good chance Spreading Seas would stop them from casting a large portion of their spells for 3-4 turns. Now it usually just slows them down by one turn, and sometimes not even that, which just isn't worth it, especially with a deck as slow as UW.

Dealing With Manlands
This is also not a slot you really need to devote cards to in UW now. Between your lands (Tectonic Edge and Colonnade), Elspeth, and more Path to Exile (which you should be running to help combat the rise of Mythic), you have sufficient cards to deal with manlands.

You can argue that Spreading Seas serves two functions, and, in addition to that, cantrips, but it is extraneous in both of those functions. Therefore, in effect, it is a card that has an extraneous function that cycles. The poster child of cards like this is Street Wraith. If cards like this were good, every black deck would run 4 Street Wraith. The fact of the matter is, your cards have to serve a necessary function to justify themselves. Spreading Seas does not do that anymore, so you should cut it from your deck (if you are playing a UW variant).