How do you start building your decks?

[Aetherium Slinky]

Do you pick a commander and build around that? Do you pick an interesting card and build around that? Do you like an archetype and build around that? These are all subtly different and I understand we all probably do all of these but do you notice a theme in your deck building?

Recently I have noticed that I tend to build around a single card. This obviously means a lot of tutors and draw but I feel like it makes for an interesting brewing session. Sometimes I add other win conditions along the way so that my play experience isn't always the same. My Kami of the Crescent Moon deck is a perfect example of what I do: built around Iron Maiden, can mill/deck other people, can deck myself with Jace, Wielder of Mysteries out and sometimes win off of other people's things with a couple of clone effects I have. Love the deck, no game is the same as another one even if the end result tends to be one of the three aforementioned options.

[Morganator 2.0]

I nearly always start with the commander. I'll see a commander and think "I want to build this". As far as I know I've only had one exception to this, where I instead wanted to win with Near-Death Experience.

[Slyvester12]

Same as Morganator. I always pick a commander I want to build, except for my myr tribal combo deck. There isn't really a good commander for myr, so I went with Thrasios and Silas for colors, recursion, and a win outlet in the CZ.

[Rust71898]

Personally; I never build around a specific commander or card. The primary reason is that it makes a deck "brittle". If you're in a play group that isn't playing 4 player solitaire, there tends to be wraths and removal flying around. Being forced into looking for 1-3 cards, then being forced to recur those cards after they get destroyed can get tiresome. This can get further compounded when you add protection for these cards and you have to find those as well.

I find decks that are constructed so robustly that you can use the commander as a threat or trump card to be much more fun. This form of construction also helps to keep the deck feeling fresh after your first 100 games or so; as you are not looking at the same card over and over again. It also helps you to become a better player, as your deck is flexible enough to offer multiple paths to play around obstacles that your opponents will throw at you.

To directly answer the question you posed: When I first started playing commander I would build around a combo or commander, but now I pick an archetype and build a deck around that. Before I really understood the nuance of commander games/ construction; it was much easier to pick a card I liked and brew around that. Now that I have been playing for commander for ~5 years, I don't really enjoy the narrow minded thought process of: does this card support the card(s) I'm building around?

[robort]

Pick a commander and go from there. Looking mainly for synergies with my commander for this will be the engine of the deck. Without an engine it just won't run.

[Potato Chop]

First, one of two things will happen. Either I see a commander that I want to build around, or have a cool idea for a theme and find a commander to build the theme off of. Either way, I end up with a commander to helm the deck and a specific theme / goal / wincon that I want to build around.

From there I use EDHREC and Deckstats for a bunch of the suggestions, but that's only so I can find cards that I may have never heard of before or missed while initially brainstorming. If I want to find cards that do a specific thing or meet specific criteria, however, I'll use Gatherer and search keywords / characteristics.

All the cards that I like I'll madly add to the list, then cut all but the best of them when I add lands, ramp, and other utlity cards.

Then, I'll playtest the deck on websites like Archidekt or Moxfield and make all adjustments that need to be made to achieve a polished, balanced deck. After that, I'll have a complete deck that I can buy the singles for with confidence.

[Shafi5557]

I like tribal decks and dual colored decks. First I decide on a tribe. Then I go to scryfall.com and search t:legendary t:(tribe) o:(tribe) and look through all the commander candidates that are 2 colored. Once I have a suitable commander for the tribal deck. I go to edhrec.com and look at decks people are playing to gauge what kind of cards are good for the deck. Then I netdeck a basic tribal deck with the commander from edhrec. Then I refine the deck with scryfall searches to find better cards for the deck.

I use these rules for the deck (slightly different rules for landfall decks).

37 <= Total Lands <= 40
19 <= Total Basic Lands <= 40
0 <= Total Tap Land <= 5
0 <= Total Colorless Land <= 5
1 <= Total Destruction Land <= 3  (I hate Gaea's Cradle and Cabal Coffers)

10<= Total Ramps <= 15
5 <= Total Draw Cards <= 10
1 <= Total Tutors <= 2              (I think tutors that can search any card should be banned in commander - some decks become auto win with tutors)

29 <= Total Creatures <= 32
20 <= Total Tribal Creatures <= 32

2 <= Total Artifact/Enchantment Removal <= 4
2 <= Total Boardwipe/Creature Removal <= 4
0 <= Total Counters (for blue decks) <= 4

100 >= Total Cards that fit the theme of the deck >= 20

I try to keep the mana curve average at around 3.5. 3.4 is best but I sometimes get to around 3.75.
You want the mana curve to peak at 3 and the bars to be descending on both side of 3. 0<1<2<3>4>5>6>7>8>9

For mana source distribution (deciding how many lands of each color). I look at X (Major Color's Mana Distribution) and Y (Minor Color's Mana Distribution)

Then I plug those numbers into the formula. (For more then 2 color decks, the formula is slightly different)

if Y/(X+Y) > 16.666...%

y (number of minor color lands) = (#Total Lands - #Colorless Lands - #Multicolored Lands)*(1/2)*(1/2 + (Y/(X+Y)))
x (number of major color lands) = (#Total Lands - #Colorless Lands - #Multicolored Lands) - y

if Y/(X+Y) < 16.666...%

y (number of minor color lands) = (#Total Lands - #Colorless Lands - #Multicolored Lands)*(Y/(X+Y)*2
x (number of major color lands) = (#Total Lands - #Colorless Lands - #Multicolored Lands) - y


I use mtgassist.com to find cards that are similar to each other. For example if you search Propaganda in mtgassist you will find Ghostly Prison as the first card (which is a white version of Propaganda) and Elephant Grass as the 7th card (which is a green version of Propaganda).


Edit: Of course all rules have exceptions. These are just the basic for getting my deck started.
Edit2: I have trouble building green decks because green has so much options for everything.

[WWolfe]

For the most part, it usually starts with the commander be it a commander that speaks to a tribe I like (Varina zombies or Sidisi zombies), a theme that I like (Skullbriar and =+1/+1 counters), or a strategy I want to try (GAA and permission). I have at times found myself building around specific cards (Shadowborn Apostles in Karador or Archangel of Thune in Rune-Tail).

My latest project is making a "steal your stuff" deck where it's just built around the concept of stealing your stuff one way or another. I haven't decided on a commander yet as other things keep getting in the way of me taking the time to do it, but the commander will likely just be there to support the colors.

So I guess my answer is it leans more towards starting with the commander but I have built decks for other reasons.

[Mynus]

I start with a commander and build from there. I personally like very commander-centric decks. I want the deck to be built based on what the commander can do or supports. My strategy is always centered upon the commanders theme. This means I even avoid putting in some "good" cards because they don't fit with what I want to do. I think the easiest example is a card like Torment of Hailfire, a really good card that can win games on its own, but I don't want to win that way if I am playing my Gyruda, Doom of Depths clone/mill deck.

I understand this makes a deck less "competitive", but I am not and do not want to be approaching cEDH levels. I like variance and creativity, and focusing on a commander for a deck helps make that deck unique. I enjoy winning, and I do want to make sure my decks can win games, but my priority is enjoyment, if winning means less creativity/uniqueness for more proven methods then that gets boring real quick. Josh Lee Kwai said the following a couple times in recent Command Zone videos and I love his sentiment, "I don't want to build my deck to win, I want to play my deck to win." In context he was talking about knowing what other people at the table were playing and designing a deck to beat theirs, and he would rather build a deck he enjoys and play it to the best of his ability. I think this is a good way to looks at things.