For me the idea of an edh deck generally starts when I see a legendary creature that meshes well with a theme that I want to explore or re-live from something else in my magic career. For instance, I built Mimeoplasm in 2011 and the deck has been with me ever since. I am a fan of Dredge as a mechanic, specifically due to it being the first Extended deck I ever played. Mimeo gave me an opportunity to build a Dredge edh deck, which essentially didn’t exist at the time. What’s unique about him was that he was a dredge pay off card that could live outside the deck, giving you powerful options as time went on. Being an entirely graveyard deck presented a number of deck building challenges that were fun to solve. The deck doesn’t play Sol Ring or any of a number of staples from the format because they don’t do anything from the graveyard.
As the deck evolved it took on a number of sub-themes. Its got a Big Dudes theme since Mimeo is often an enormous creature, and it feeds on huge creatures well. It took on a lands theme since Life From the Loam is so integral to a dredge deck. It has a minor reanimation theme, and its got a couple of engines that are good on their own, but infrequently assemble to allow the deck to have a combo finish. What I look for in decks I want to build is open ended abilities that let the deck evolve and take on new roles over time.
There is an upcoming new legend that I want to build, Syr Gwyn. For a long time, I’ve been looking for a legend to accommodate my love of Mardu aristocrats. I played the original Aristocrats where the archetype takes its name from during the original Innistrad block. What was fun about Aristocrats was that it was a blend of aggro and combo. You had sticky creatures that were strong attackers, and had the potential to just get your opponent dead out of nowhere if your draw came together. I want to explore getting your fodder dudes and sticky creatures to generate value through attacking with cards like Sword of Fire and Ice.
Regardless of how I settle on a commander, the first thing I do is dissect the card to understand its ins and outs. I’ll be using Gwyn as an example but the principles apply to any commander. The first piece to look at for a commander is the mana cost. Not only the colors, but also the cmc. Gwyn costs 6 mana. 6 mana is a lot. The commander rule lets you recast a commander, but at 6 mana its likely Gwyn will only get cast 1 or 2 times a game. If she cost 5 I’d say she could be cast 1-3 times a game depending on how long the game goes, and at 4 I’d say she could be cast 3 times a game. Mana cost is the single biggest factor in determining how likely it is your commander will be on the battlefield in a game. The second thing to note about the cost is the color. Gwyn is Mardu which means that your mana acceleration will have to be from artifacts. This makes the 6 mana cost all the harder.
The next important bit to absorb is the text. Gwyn has quite a bit of text. The first line is her keyword abilities. She has Vigilance and Menace. Vigilance means she can be aggressive without leaving yourself open to attack. Menace means its harder to chump her, and might be impossible depending on what she’s equipped with. The next ability is the meat of the card to me, “Whenever an equipped creature you control attacks, you draw 1 card and lose 1 life.” This ability lets you gas back up after dumping your hand. It wants you to have lots of creatures and lots of equipment. Since you don’t get rewarded for stacking equipment, its better to have 3 creatures each holding something than to have 1 creature holding 3 things. This is what made me think of aristocrats.
Aristocrats is aggressive. It wants to be attacking. It also has a lot of dinky creatures that don’t do much without their combo partners. Karmic Guide is a great card. It doesn’t do too much after it has come into play. Lingering Souls gives a ton of bodies, but they don't have much meaning in EDH aside from generating fodder. Equipment is a way to mitigate against that since it lets these creatures get a lot of value just by being in play. To go back to Gwyn, she caps off this style by digging you a few extra cards, letting you find missing combo pieces after you dump your initial solvo. The last line is also pretty good, “Equipment you control have Equip Knight 0.” This indicates a tribal theme, by letting you get your equipment onto Knights for free. While it certainly is better with other knights, she herself is a Knight, so the ability immediately has a target already.
Lastly, she is a 5/5 creature, which is pretty large though it isn’t huge for her cost or for what people can sling at 6 mana.
From this dissection, we can start to see the shape of the deck. We care about the aristocrats package, we care about equipment, and we care about attacking. We want to include as many knights as make sense, but we don’t want to stretch for them. The deck will probably be weak to board wipes, and we’ll need a bunch of mana acceleration to be able to play Gwyn when the time is right.
That's all well and good, but how do I find these cards?
Well, now that we know what we're looking for, there are a number of tools to help find cards.
The primary tool used in Commander is EDHREC. Its a decklist aggregation site that compiles all of the lists of a given commander and gives you what the 'average' deck for them looks like.
It's useful to see the most common things people are playing with the commander so you don't overlook something obvious. However, its falls flat when you're doing something more niche.
When you want to really look, you need to go to gatherer. Gatherer is a database of every magic card ever printer. It also has a strong search function. I bet you weren't expecting data science?
For Gwyn, I've identified a number of things that I want to compile. I want to see all the Knights in my colors to determine which match the concept. So I go to gatherer, I select RWB, exclude unselected colors. I check the box for "types" and "text" and leave name unchecked, and hit search. Then I look through the list of cards and pick out the ones I want to potentially include in the deck. Then I repeat this process with but I replace Knight with Equipment. This is a broad search so that we get an idea. But, there isn't a clean keyword to search for when compiling the list of Aristocrats.
For a Theme that isn't a keyword you can search for, EDHREC is your best resource.
Not only do they have the ability to sort lists by theme, it clues you in on patterns.
While the commander you're using might have limited lists, or pull towards the average, you can look at other commanders that are closer to your theme. In this instance, Alesha or Edgar Markov are the premier Mardu aristocrats generals. So, I go into their lists and compile cards that I think would work in the shell that I'm building.
Unfortunately, this bit only comes with time. Since I've built so many decks, my brain has shortcuts about what belongs with what. For instance, this deck is R/W and has an equipment subtheme. To me, that instantly means Sunforger. And Sunforger comes with its own sub-set of cards that should included. Things like Anguished Unmaking or Wear//Tear are part of the Sunforger package. Sunforger is part of the equipment package. The Knights that are being included are ones that line up with the Aristocrats package.
A deck is a series of relationships.
When putting a deck together, its not a matter of whether the card is good, but its weighted evaluation with the other components in the deck. Corpse Knight probably wouldn't make the cut in a different take on Aristocrats, but being a Knight raises his evaluation here. For an example of exclusion, I don't run Necropotence in my Kess deck. Necro has a clause where discarded cards are exiled, which means the core engine of the deck, Dream Halls, isn't filling the graveyard. Here, Stonehewer Giant is normally a shoe in for equipment decks. I don't think it makes the cut in this deck. We aren't all in on equipment, and it needs a turn to come online, and we're hopefully curving into our commander at 6. Since his ability costs mana its likely that we'll have to choose between activating him or casting Gwyn.
Now that you have a list of cards you want in the deck, you can start looking at metrics.
Common wisdom is that your deck should be 40% mana. It varies based on the average cost in your deck. A mono-red aggro deck can probably run 1/3 mana and be fine. A control deck with 6 mana finishers probably wants to tend toward more mana, maybe 1/2 of the deck. The calculation also changes based on how much cheap draw and acceleration you have. A deck with 4 Ponder 4 Brainstorm can probably run 4 or so less land since they have efficient card selection.
Given that Gwyn costs 6 and is a card draw engine, its probably correct to make the deck 40 lands and 5-10 pieces of artifact mana. We can easily dump our hand to get to 6 mana, then refuel with Gwyn. That gives us 50-55 slots for cards. The next part is kind of mathy. So, we want to know how many Equipment we need in the deck for there to be one available when Gwyn is played. We can assume that we've drawn 13 cards by the time we play Gwyn.
This article explains what's going on here: https://www.channelfireball.com/articles/an-introduction-to-the-hypergeometric-distribution-for-magic-players/
Then, I plug the number into this calculator: https://stattrek.com/online-calculator/hypergeometric.aspx
In the example above, if we have 10 ways to have an equipment on the table by the time we play Gwyn, there's a 75% chance that there will be an equipment for her to pick up. If that number goes up to 11 it goes to 80%, but the gains taper off after that and at 15 its 90%. The question is how important is this to have on time? I'd say about 75% is fine. Especially since the larger the sample size, the probability increases. At 14-15 cards seen we go up to 80-82 percent. I think the number I'm most comfortable with is 11 equipment/equipment tutors. That puts us at 85% by the time we've seen 15 cards. As the sample size shrinks the number of successes needs to rise to maintain the probability.
If I want to guarantee a mana accelerator in the opening hand, the number goes from 10 to over 30. 10 is a 50% shot, 30 is a 90% shot. This is where hedging against variance becomes a skill both in deckbuilding and gameplay. It is unreasonable in most decks to run 40 land and 30 mana accelerants. The question becomes what probability is reasonable to accept?
Generally I take 50/50 for opening hands to be acceptable.
So, now that we know some of the probabilities involved, we can set thresholds we don't want to go under.
I don't want to have less than 50% to have the following in my opening hand:
An accelerant
An equipment
A sacrifice outlet
A sacrifice payoff
3 land
That means we'll need to include at least:
37 land
10 mana artifacts
10 equipment/equipment tutors
10 sacrifice outlets
10 sacrifice pay offs
I am willing to count tutors as stand ins for the cards themselves. So 9 equipment and Stoneforge Mystic would be counted the same as 10 equipment for this math.
Now that we have a pile of cards, we need to winnow it down. This is where card evaluation comes to the fore. Where the previous efforts are proactive statements, "I want to have enough equipment to make my little guys threats.""I want to have enough acceleration to play Gwyn earlier that turn 6." The next bit is all about questions. "Does this support my game plan of accruing value through attacking until I assemble a combo?" "How will I handle problem permanents?" "How do I rebuild when someone wipes a board?"
The best deckbuilders are the ones that consistently ask the right questions.
I’ll go through this process with one of the subsections of cards for brevity's sake.
These are the knights that I think positively reinforce the themes of attacking and value.:
Cavalier of flame
Cavalier of night
Corpse knight
Elenda, the Dusk Rose
Haakon, Stormgald Scourge
Hero of Bladehold
Knight of the last breath
Knight-Captain of Eos
Midnight Reaper
Olivia, Mobilized for War
Puresteel Paladin
We know mana might be an issue so the first question is, do any of these cards cost what I would consider to be too much mana? I would say that Knight of the Last Breath is immediately too much at 7 mana. It made the list because of Afterlife and being a sac outlet, but 7 is too much to justify for this deck. The cavaliers and Knight-Captain of Eos are both under the cost of our general but might be too much at 5, especially since the cavaliers are triple of their color. I think I’m going to cut the Cavalier of Flame due to cost, but keep the Cavalier of Night since its both a sacrifice outlet and recursion. The Knight-Captain goes away since there are more efficient sources of food, and he only sacrifices soldiers.
Are there any that don’t do enough for the deck’s game plan?
Olivia, Mobilized for War and Hero of Bladehold don’t seem to reinforce out theme as well as the other options. The Haste generated by Olivia has value, but I don’t think we’re doing anything with madness and we want all of our cards. Hero is such an enormous threat that we’re keeping it, it also produces food every turn, which is much better than Knight of the Last Breath.
Are there any with special concerns?
Haakon requires you to put it in the graveyard first. If you can get him there, then he lines up with the deck quite well, turning every knight into a creature that you can sacrifice repeatedly for value. How will we get him in the graveyard? There are a number of options here. A discard outlet, Entomb, etc. I think he’s worth the trouble but he could end up getting cut if it doesn’t work out in practice.
So, the new list is:
Cavalier of night
Corpse knight
Elenda, the Dusk Rose
Haakon, Stormgald Scourge
Hero of Bladehold
Midnight Reaper
Puresteel Paladin
Applying this method to my compiled list of cards, this is my rough draft:
https://tappedout.net/mtg-decks/gwyn-aristocrats/?cb=1568408773 It's a flexible aggro-combo list that might have trouble consistently generating enough mana to do all the things it wants to do. The only way to prove the concept is to build it and smash it against other decks. I can tell that there's some optimizing to be done already but its a decent starting point.
So...that's how I build decks and what I value.
I like to make decks that are on theme and explore a variety of synergistic cards.
I look at the proactive statements of what I want the deck to be and compile the cards I think match the mission statement. I use reductive reasoning to ensure that I'm asking questions about how the deck will handle problems and spot issues it might have in operation. Then, I put the deck together and tune it over time until I stop having fun with it or decide it isn't working and break it up.
