Caveat:
This is not"How To Make The Best Deck Ever" or"How To Make A Deck That Will Win the PT" or even"How To Make A Deck That Will Win Friday Night Magic" - it's just"How to Make a Deck." This is a process article that follows my line of thought from idea to initial list. There are a bunch of steps that you can copy when you have your own deck ideas, including some tips on how to decide between two different cards, and what to look out for in terms of how you would advance your strategy over the course of a developing game. The deck that results may not be very good. I mean, it could be absolutely insane, and I think it might be reasonably good, but since it's not even going to be finished until four hundred or so lines from now (depending on your window size and screen resolution), who knows?

I'll let you in on a little secret: most decks and most deck ideas are terrible. Even the very best deck designers make mostly horrible, even unplayable, decks. When he was working with the masterfully innovative Pat Chapin in '99, Chris Pikula figured it out: Of course you have the best ideas when you have all the ideas. Most of us are lucky to stumble on just one or two for every format.

So anyway, there are lots of different ways to go about building a deck. There are active decks and reactive decks. Active decks are in a sense easier to build because you don't have to take as many factors and cards into consideration; if your plan comes online, you win... It doesn't matter what your opponent's twenty-turn game plan is. Reactive decks, especially those that plan to win over a very long game, have to consider all kinds of cards, some of them jank (but jank that would nevertheless beat them if left unaddressed). There are natural decks and there are predator decks. There are decks that are based on earlier designs and brand spanking new ideas. Increasingly, there are decks (or at least strategies) that seem custom built by R&D, with synergies encouraged by block mechanics such as Rebels, multicolored alliances, Threshold, Madness, tribes, or Affinity.

Today we are going to build a reactive Standard deck based on a new card that I find interesting. That card is...

Vedalken Shackles.

Vedalken Shackles is a ridiculously overpowered spell. Given the creatures being played in the relevant Constructed formats, I'm not sure if it is quite on par with Treachery because of its low aggregate cost or simply much better than Control Magic. Vedalken Shackles seems to me either a desperate ray of hope shined on their beloved Islands by my old friends Randy Buehler and Brian Schneider, or permissible only because of the decided paucity of Islands in Standard and MD5. The only thing that is really holding back Vedalken Shackles is the splash damage inflicted by all these Oxidizes artificially shoved into everything from Goblins to White Control by the prevalence of Arcbound Ravager.

The principle design constraint in playing a Vedalken Shackles deck is that you have to play a lot of Islands. Not Seat of the Synod, not Coastal Tower, not Polluted Delta; you need to have good old-fashioned basic Islands in your deck to make sure the card is playing smoothly. I would guess that fourteen Islands is the absolute minimum number we can play... And in this deck, we are going to overload in cantrips to make sure that our Islands are flowing and that we hit our critical spells.

Blue is traditionally good at a lot of things - but right now, it isn't even adequate at most of them. Permission is something we will get to in more detail later, but blue can still draw cards better than any other color. Since we are playing so many islands, there isn't anything really holding us back from playing with a cards like Future Sight... So mise.

The main hole left - especially given the current state of Standard - is in board control. Long gone are the days of Repulse and Treachery, Nevinyrral's Disk and Powder Keg... So even though we have Vedalken Shackles, it's not likely that we will be able to get by on just Aether Spellbomb. If we want to play any kind of a viable control deck, we will have to dip into another color. I am going to arbitrarily select red. There are a ton of reasons why red would make for the best synergy. Even though white is probably technically the better match for blue in a Standard control deck, all the good white control cards - Akroma's Vengeance, Pulse of the Fields, Wrath of God, et cetera - cost WW and we are going to play with fourteen Islands minimum. Black has fine control cards, but they are inherently limited... Some can kill Arcbound Ravager but not Frogmite, some can kill Goblin Warchief but not Disciple of the Vault, and so forth. I like red because its removal is inherently more versatile: Going long, it can also go to the head!

The subtle difference in selecting red over any other color of removal is that red currently has several different cheap options, all of which double as library manipulation. Had we had the capability to play white, we would still be bound by mana... But red lets us play with these great flexible cards that are custom-built to play the game in two different modes. In the early game, they can either eliminate threats or cycle into Future Sight. Later in the game, when Future Sight is already down, they let us play"Necro"... Overloading our opponents with tons of cheap but efficient cards while Future Sight generates more card advantage the more cards we burn.

So at this point, it is probably appropriate to list all the cards we might want to play as a natural deck:

4 Aether Spellbomb
4 Chrome Mox
4 Guardian Idol
4 Pyrite Spellbomb
4 Spire Golem
4 Vedalken Shackles
4 Condescend
4 Future Sight
4 Mana Leak
4 Serum Visions
4 Stifle
4 Thirst for Knowledge
4 Magma Jet
4 Pyroclasm
4 Slice and Dice
4 Spark Spray
14 Island
XX Mirrodin's Core
XX Mountain

That is obviously too many cards. I think that I want to play Chrome Mox because it is really good with Thirst for Knowledge, surprisingly synergistic with Guardian Idol, and absolutely insane with Future Sight. With four copies of Chrome Mox, assuming we play with all kinds of library manipulation, twenty lands is probably enough.

So let's say we move onto:

14 Island
2 Mirrodin's Core
4 Mountain

We have 40 spells to play around with. Some of them are spoken for...

4 Chrome Mox (as above)
4 Guardian Idol (mana that doubles as a way to win)
4 Vedalken Shackles (baseline)

4 Future Sight (ditto)

So we basically have 24 cards left to defend ourselves while we set up Future Sight and/or dig to the Vedalken Shackles that we use to dominate the board.

AETHER SPELLBOMB isn't going to make the cut. It's very synergistic with what we want to do, and is pretty high on the curve in terms of blue board control, but it just seems too Wicker Man Theory in the power department for me.

This is actually a good time to mention relative card power. As a younger deck designer, I actually worked on lists with a copy of Inquest next to me. Inquest was even then notoriously awful at correctly assigning card power (think of Elkin Bottle and Necropotence as notable examples that caused pro dissatisfaction), but they had a reasonable set of valuations for staples. While it is important to stay focused on the ideas that fuel your decks and strategies, the biggest mistake that a good (or potentially good) designer can make is to radically underpower his deck because he is so busy working on his strategy. For example, I am notoriously bad about putting too much library manipulation and mana acceleration into my complicated decks, and have to go back to the drawing board on even the best lists in order to put in enough cards that, as Brian Kibler says,"actually do something."

PYRITE SPELLBOMB, on the other hand, is going to the short list. It kills Goblin Warchief and Disciple of the Vault. It cycles into Future Sight and basically draws an extra card for one mana or three extra cards for two once Future Sight is in play. It even smokes Silver Knight. What's not to love?

SPIRE GOLEM is a really good spell, and I think it will make the cut in some deck... But maybe not this one. I think Guardian Idol took Spire Golem's spot in this deck, but it is entirely possible that the resulting deck will not have enough early-game defense and that Spire Golem will be necessary to block. It isn't going to make the first cut, but might make the cut down the line.

CONDESCEND looks a lot better than it is for exactly the same reason that Power Sink was played only rarely in competitive decks. For a permission spell to be really good, it has to be cheaper than the card it is countering. But that said, Condescend is still a playable counter, and has the more important distinction of doubling as library manipulation.

MANA LEAK is the best permission spell available for defending in the early game. That said, it is atrocious in the long game... And this is a long game deck. It is also quite bad in conjunction with Future Sight.

SERUM VISIONS is fabulous. When you are cheating on lands and building your deck around a couple of key spells, this is the kind of card that you really want.

STIFLE is really good because it can keep your opponent from beating you. Say you go long against a White Control deck, but suddenly he plays a twenty-point Decree of Justice... You just lost. It will also counter Mindslaver and Myr Incubator, which happen to be two of the most important win conditions in the upcoming format. It can manascrew players on turn 1, but I wouldn't suggest trying that if the opponent is breaking a Flooded Strand or Windswept Heath. If you blow your Stifle on turn 1, and you are pretty sure the game is going to go thirty turns, your short term mana advantage isn't going to mean very much when you eventually get overrun by a Decree of Justice. You would be surprised how often some absolute destroy you play ends up yielding a win for your opponent. Danger of Cool Things and all that.

THIRST FOR KNOWLEDGE is straight in (see what I told you about that library manipulation thing?).

MAGMA JET is another automatic. It digs for Future Sight and makes Future Sight better once it is already online.

PYROCLASM will make the cut... But probably not in the main. I don't think it is flexible enough, but I might be wrong.

SLICE AND DICE seems quite good against many of the cards that will be problematic for a U/R control deck, but is also very expensive. In this kind of a deck, you can ultimately do just as much damage with a flurry of one-for-ones in the mid- to late game as with a lone sweeper.

SPARK SPRAY is not going to make the cut. I really like this card, and it does everything that Pyrite Spellbomb does... But not as well for the most part.

That leaves us:

4 Chrome Mox
4 Guardian Idol
4 Pyrite Spellbomb
4 Vedalken Shackles

4 Condescend
4 Future Sight
4 Mana Leak
4 Serum Visions
4 Stifle
4 Thirst for Knowledge

4 Magma Jet

14 Island
2 Mirrodin's Core
4 Mountain

That is just a little too fat a deck.

I'm going to tentatively cut Mana Leak because of its weakness long game, and because Stifle can essentially counter cards like Siege-Gang Commander anyway. I don't know that a pronounced lack of permission is actually that big a deal, because this deck really just wants to one-for-one into its midgame, and anyway, we can sideboard Mana Leak like the Affinity decks or add it to the main later, if testing says that we need a turn 2 response to some threat that can't be easily burned.

4 Chrome Mox
4 Guardian Idol
4 Pyrite Spellbomb
4 Vedalken Shackles

4 Condescend
3 Future Sight
4 Serum Visions
4 Stifle
4 Thirst for Knowledge

1 Beacon of Destruction
4 Magma Jet

14 Island
2 Mirrodin's Core
4 Mountain

Sideboard:
4 Annul
4 Mana Leak
4 Dwarven Blastminer
3 Pyroclasm

The last tweak I'm making is to cut one of the key cards (Future Sight) for a lone Beacon of Destruction. This deck is designed to go long. That means that it has to flow properly in the midgame... and a redundant Future Sight can be annoying if not fatal. Beacon of Destruction, on the other hand, seems fabulous. It can kill almost any creature that gets played typically, and plays Fat Moti pretty well all by its lonesome. It is also a great card in conjunction with Future Sight as our deck spreads all over the board.

So is this deck good? It seems reasonable to me given some of the test draws I've taken. I can already identify a hole in the Troll Ascetic department, so maybe Mana Leak is going to have to return to the main. That said, it shouldn't be much of a problem to counter a Troll or just Shackles up some good buddy who is big enough to hold off a 3/2.

The next step is obviously the testing stage, where we decide if a U/R Vedalken Shackles deck is worth our time and effort when compared with Affinity, White Control, Goblins, and X and Nail. That step is beyond the scope of this article (but maybe not one down the line). Testing is the step that teaches you what you missed in the first build stage, and is the time between your idea and the trial of a tournament where you can fix your mistakes before tournament time.

The biggest thing to remember when you are building a deck based on a new card is that your opponents also got some neat new cards, and you might not be prepared to face some of them... I know that I've been caught that way in almost every block since Aether Flash and Abeyance in Weatherlight.

In sum, here are the steps we took to build this deck:

1. We came up with an idea.
2. We identified the design constraints implied by our idea.
3. We chose a strategy potentially complimentary to our idea.
4. We listed all the cards we might play, given the idea we wanted to follow.
5. We identified the"must haves" and got a general idea of how much additional room we had to customize our design.
6. We pared down our list based on evaluating one card against a similar card, the synergy between the remaining cards and our core"must haves" and overall card power.
7. We laid out what we had in anticipation of a sixty-card cut.
8. We looked at our sixty-card list and made final tweaks.
9. We are going to test the deck against the expected opponents.
10. We are going to decide, based on the testing data, whether it is worthwhile to try to fix the deck's problems or just abandon the deck based on its lack of consistency or performance.

That's it!

--michaelj

Post Script
Just so you know, this is actually a deck idea I had this week, and this article actually describes the process of how I went about the task of fiddling with the cards on Apprentice in order to get to the sixty-card list at the end. Is it perfect? My data isn't anywhere near complete. But I had Mana Leak at one point and I have Stifle now. I added Beacon of Destruction a couple of days in because it seemed like a good fit, and I was getting so many Future Sights so quickly that I really did stall on Future Sight often enough to want to cut one.

The development process, rather than the design process, is in my experience the most laborious and rewarding part of making a good new deck - and any tournament-winning builds I've ever been a part of have been the result of good metagame analysis, strategic awareness, and identification of choke points that can come only from a correct understanding of how games in a particular environment develop. If this deck isn't perfect at stage one... It really isn't supposed to be.