Breaking Through – From the Ground Up


Deckbuilding is sometimes referred to as an art while other times being called a science. Personally, I feel like putting it into either extreme does the process a disservice as it invalidates some important aspect of the act, and would rather like to think of it as a fusion of the two extremes. While there are plenty of moments in deck building where you just “feel” the right card choice or numbers on that choice, there are also plenty of times where you need to grind out some numbers and reach for your math book to figure out the perfect mana base. Today I would like to talk about a method of deck building that uses both forms of building.

A few weeks ago I alluded to the numbers game in building, where you funnel your card numbers from a large, generic base, to more refined numbers, even if those numbers remain at 4. Let’s discuss this process in a little greater detail as it is a prime and easy approach to deck building when the builder has enough time to work the process.

Typically, when I start with a deck using this process, I try to have as many 4-ofs as possible. There will be some exceptions right off the bat, such as legends or some planeswalkers, but in general, most of the cards in your deck should be rocking a full set. If you ever see a list online that looks primarily like this, there is a good chance that the builder has not refined the list a ton, and is presenting a more raw form of the deck. It does not automatically mean that the deck is crap and you should move on though. Every deck starts from somewhere. Let’s look at an example.

Example U/W Control

Let’s say we are working on a typical U/W list for Standard but do not have enough lists around to draw some sample numbers from. This is usually the case at the beginning of a season when there is basically no supporting data for your deck or metagame. We have narrowed down the list of cards we are willing to try to a dozen or so and need to start fitting them together. To start off, something like this is a good example.


We first must set our mark at the rough amount of lands we wish to be running. In this case, I set that number at 26. As you can see, our numbers are all 4-ofs down to the Martial Coups, which I cut to 2 to allow for 26 lands. Ultimately, you are going to have to decide what the least important spell or most redundant spell in your deck is and start with it at fewer than a full 4 of. In this case, with 4 Day of Judgment already, Martial Coup was supplementary and was probably not needed as a 4-of.

The biggest reason to start off this way is that you cannot be entirely sure as to what cards will be best or needed in full strength for your list. You may have a very good suspicion, but it is best to validate that with testing first. In order to get an appropriate feel for each card you need each one to have a significant impact on initial games. The only way to almost guarantee this is to start with each card as a 4-of. The mana base is obviously a bit different though.

Here you are going to want to look at a few different factors. These include the timing of the mana, meaning when you need to hit certain colors, the amount of mana symbols, and the density of those symbols on individual cards. If for example you had 32 white cards all with a single white mana requirement and 4 black cards each with BBB in their costs, it is not correct to just look at 32 White symbols and 12 black symbols, as that does not paint the whole picture. Using the above deck we see the following:

1 2 3 4 5 6 7 X
White W-4 W-4 WW-4 WW-4 WW-2
Blue U-4 UU-4 UU-4

From this we can see that we need either a blue or white mana on turn 2 depending on which spell we want to cast and either two blue or two white on turn 4. This essentially means we need both blue and white mana on those turns, since we cannot control which spell we draw for those turns. Therefore because of this, and the fact that blue provides draw spells (meaning that while it shows up less often, finding it will lead to finding white), we need a rough 50/50 split on mana. That leaves us with the following.


With 26 lands in our deck, we definitely want to make use of “excess” mana, meaning the manlands and edges are a must. We don’t want too many lands that enter the battlefield tapped and currently are sitting at 8 that do so. Again, remember to stay basic at this point in building, the fancy numbers will come later. There may come a time when you are able to skip a step or two, but for now just take it easy and slow.

Of course, once the initial build is finished, the next step is testing. This is not intended to be an article on how to test properly, so I will simply assume you are getting in enough games to adequately evaluate the deck and all of its cards. With such high visibility for each set of cards, if you play enough games, each one should tend to showcase itself in various situations. Be sure to test different matchups as well of course, or else your results may be skewed. You ultimately are not looking to see if you win or lose during this time, but want to learn how each card functions in a matchup. This is a good place to take notes as well so that when you go back to the drawing board, you can know what to change.

Since this is a mock build, I will just throw down the following “notes” that we acquired through playtesting.

-If possible, more shuffle effects for Jace, the Mind Sculptor.
-Always seemed to have extra Wraths in my hand. Martial Coup was fine in this regard, Day of Judgment wasn’t.
Path to Exile was good at times, but usually a Wrath was all I needed. Could cut some of these.
Baneslayer Angel was good as usual, but the opponent tended to have removal. Would like to try other win cons.
-Probably don’t need 4 Mind Spring and 4 Jace, could look to cut a 1-of here.

With those rough notes, let’s go back to the drawing board and begin to fine-tune some numbers. The first thing we are going to do is shave a Day of Judgment. 6 Wrath effects is a lot and Day is just worse than Martial Coup in the late game.

-1 Day of Judgment

Next, we can go look at our Path to Exile count as well. Since we are shaving a Wrath effect, we don’t want to go crazy and shave too many Paths, otherwise if we begin to lose to creatures, we cannot attribute that to one of the two cuts. Instead, we are just going to cut a single Path for now and find out if that is still solid in our next round of testing.

-1 Path to Exile

Looking at the issue of too many Jace, the Mind Sculptor and Mind Spring, I am more inclined to cut a Mind Spring as Jace has other uses and is quicker to die without other Planeswalker support, which brings up the idea of trying other walkers out. For now though, out comes 1 Mind Spring.

-1 Mind Spring

Regarding the shift to more shuffle effects, we can likely get away with cutting a single Tectonic Edge, Sejiri Refuge, Island and Plains to add in 2 of each fetchland. If we utilize Arid Mesa and Scalding Tarn, there is even a chance that some opponents think we are UWr planeswalkers, which can have its perks.

-1 Sejiri Refuge
-1 Tectonic Edge
-1 Island
-1 Plains
+2 Scalding Tarn
+2 Arid Mesa

We are also going to go ahead and try out Sphinx of Jwar Isle over Baneslayer Angel just to get a different feel for each win condition. Unlike Baneslayer though, you typically only need 3 Sphinx of Jwar Isle due to it not dying nearly as often.

– 4 Baneslayer Angel
+3 Sphinx of Jwar Isle

This leaves us with 4 open slots now and we should use this space to experiment a little bit. I want to try out a pair of [card]Gideon Jura[/card] in here as he can protect Jace pretty well. If he ends up working out, we can even drop 1 Jace for the 3rd Gideon. To help make up for the loss of a Path and Day of Judgment, I want to mess with [card]Oblivion Ring[/card]. It is able to attack opposing Planeswalkers which makes it very valuable in the maindeck.

+2 Gideon Jura
+2 Oblivion Ring

This leaves us with the following list that we are going to take back to the table and perform a similar testing session to the one we did before.


Once again though, before we take this to the testing table, we need to verify that our mana ratios are still right, here is the updated chart for the new list.

1 2 3 4 5 6 7 X
White W-3 W-4 W-2 WW-3 WW-2 WW-2
Blue U-4 UU-4 UU-3 UU-3

It looks like white has become more present, but in reality, it has lost some mana symbols. 4 white symbols were lost from the 5 drop slot, 2 from the 4-drop slot, and 1 from the 1-drop slot, while only 2 were gained at the 3-drop slot. Therefore our rough 50-50 mana base should be almost perfect now. Back to the testing tables!

Note that we still don’t have any 1-ofs in our deck, as we still want to feel out how our spells work. 1-ofs are awesome at the later stages of a deck, but right now they just add too much variance to our testing when we are looking for stability. The 1-ofs will be the topic of our stage 3 building though, so we will get back to that shortly.

Again, here are some mock notes from our second phase of testing.

-Gideon performed well, protected Jace, can probably finish out with that swap
-Oblivion Ring may be better as a 3 of over Path as Tectonic Edge and seas already help with man lands.
-Plenty of mana in the deck leads me to want to cut a land or Chalice. I always have enough.

Your notes this time around should be more specific and preferably shorter, as you have already gone in and rewired some things. Let’s address this new batch of issues really quickly.

The swap for a Jace and Gideon is pretty simple and should help the deck close faster.

-1 Jace, the Mind Sculptor
+1 Gideon Jura

The swap between Path and Oblivion Ring is also pretty simple and should help your own walkers from getting legend ruled out so much.

-1 Path to Exile
+1 Oblivion Ring

We want to cut a mana source, but cutting a land is pretty risky. I am fine with cutting a Chalice as your turn 2 is going to almost always be dominated by Spreading Seas or Wall of Omens, meaning you really don’t need a Chalice until turn 4. In its place, we are going to try our first 1-of.

-1 Everflowing Chalice
+1 Elspeth, Knight Errant

Once you get to this point in the building process, trying out 1-ofs is actually extremely helpful. You should know by now what cards do and how they do them, meaning you will be able to tell if you miss that card’s impact on the game. For example, while Path to Exile is obviously good, we aren’t learning much from playing it as it is tried and true. Instead, maybe we want to try out 1 Oust in place of a Path. In the long run, Path is likely to get the nod, but we have the time and capability of trying out a new card so we may as well. Of course your card choices still need to be grounded in solid theory, but branch out and explore some new options.

Elspeth is a proven card but has not seen much play in control decks. We are trying her out just to see if her abilities provide anything extra to the deck. If she does not work out, there was not much lost and you still felt the impact of 59 other cards. At this point, simply switch her out for some other 1 of until you find the right one.

Imagine if we had started with all 1-ofs though. We would not be gathering any information on how good individual cards were in common situations. This is because we would have no common situations. It is easy to learn to hold onto your Path when you are planning to cast a Day of Judgment in two turns. If however, we have 1 Day of Judgment in our entire deck, this set of circumstances comes up so infrequently that we must now play our Path differently and therefore learn nothing about the final deck.

This type of deckbuilding will not be as useful for established decks, as some idea of their numbers is already known so you may as well use the existing data. At the beginning of a new format, or for rogue decks, this method works really well as it establishes your deck before you can go in and begin tweaking anything. It is important to understand the block of granite as a whole before you begin chiseling away. It is awesome if you can see the statue that will emerge, but take your time in getting there.

Conley Woods


Scroll to Top