EDH Manabase Guide

Here’s a question: How many sources of green do you need in your 99 cards to make sure (100%) that you have at least one of them in your opening hand of 7? The answer: 94.

Yes, that is somewhat unhelpful, but it also shows you that you always play with odds when designing decks, including the manabases. Designhing you manabase is a complicated problem that is often simplified into rules of thumb like “you need x lands and y sources of ramp in your deck”. These are largely based on experience or just copies of how other players approach this. At least nowadays x seems to be around 37 and y seems to be at least 10, but many players still mise on lands and want to go with less. I get that having spells in your decks is fun, but I also personally believe that being able to cast those spells is more fun. Therefore I play a lot of lands and a lot of ramp. Many recent Standard decks have been playing upwards of 28 lands and ramp on top of that. Your curve in EDH is going to be higher (in general), so why should you be playing less lands relatively? I do often play close to 50 lands, but I usually settle for less, but just because I also keep my curve very low.

My personal x is 40 and y is 10, but I also tend to move these numbers around. Usually upwards with the lands and sometimes downwards with the ramp, because of thematical constrictions, but these numbers are also largely based on the style of deck I like to play. I basically play midrange decks that aim to garner value over time. I also like to keep my curve relatively low as I do want to be able to things throughout the game or at least have the option to do something.

Here’s the first thing you should do: Familiarize yourself with hypergeometric calculators. You can find one online here. In this particular case (of an EDH deck), the first input (population size) is 99 (the size of the deck), the second input (number of successes in population) is the number of whatever you are looking for in your deck, sample size is whatever the number of cards you are seeing (7 for the opening hand) and number of successes in in sample should be how many you want at minimum. For example, supposing I would like to see three lands in my opening hand of 7 with 40 lands in the deck, these input would be (in order) 99, 40, 7 and 3, which would produce the result of 0.593711515 or just under 60%. If you always take a mulligan for hands that have less than three lands in them, your probability of getting that three plus land hand is way over 80.

If you design your deck properly, you’ll using this calculator many, many times… at least before you begin to have a better grasp on the results. I do still often return to it despite having used this particular site for at least five or six years, probably longer, and using charts for similar purposes before that. You’ll want to not only check your deck for lands, but also the ramp, different colors of mana, sweepers (if those are important) and other key parts of the deck.

So, based on my personal x and y, more than half of my deck is mana in one form or another. How do I reconcile this and the low manacurve with the idea that I want to be able to play spells throughout the game? I just figure out ways to use the mana. Card draw, manasinks, and spells with optional additional costs are all helpful in this. I also do like to wait and see for the correct moment to use my spells, in which case I want a lot of mana available to me.

But here’s the starting point to figuring out x: I want to be able to play my commander on curve without drawing ramp, within reason. If I’m playing an 8-drop as my commander, I would need to play almost 60 lands to have an 80% probability of doing so. If I play 40 lands, I have over 72% chance of playing my 4-drop commander on curve, but this already drops to 58% for a 5-drop commander. These numbers don’t take mulligans into account, which do make all of this much more complicated, but let’s just say that we are comfortable with that 58& probability.

Suppose our 5-drop is, I don’t know, Brokkos, whose casting cost is 2BUG or 2{U/B}GG if mutated. That is a lot more complicated than just having five lands. Depending on how you plan to use it, you now have to also have two or three different colors of mana by that point. So, take out the calculator again and put in the numbers. I usually want aim to between 16 and 20 of a certain color in my deck, depending on when I want to see it and whether I have double costs in cards. If I have triple costs in cards (say, Krond the Dawn-Clad), you have to go even further.

This does get more complicated as well, because of course it does. Suppose I want to cast Brokkos, Apex of Forever (not mutate) and I have in play four Forests and a Watery Grave. I can’t do it. I still need a blue or a black source. There are bunch of these cases which are pretty much impossible to account for, so what you should do is just go overboard with each color in the manabase.

What about the ramp? Here’s a rule of thumb I use: As stated previously, I default to 10, but this can change depending on the circumstances. I don’t count things that cost four or more towards this total. I do play Solemn Simulacrum fairly often, but it’s not a land or a source of mana when I design my deck. The reason is that I need to know that I can play spell practically always in order for me to count it towards my mana. This means that I will rather play two mana ramp cards or cards that find me lands than three mana ones. I might be missing out on some value, but my mana is there for me to be able to play the spells I have, so I want to make sure I can play the on the way as well. Also, you don’t often use your mana on turn two anyhow, so it’s a good opportunity to ramp.

Let’s complicate this a little more: Suppose you are playing green and most of your ramp is therefore green as well, but you are still playing a three color deck. Can you just count ramp towards your land count? No. An easy rule is to just have some default measure you use. Like two mana sorceries, such as Rampant Growth, could be .75 lands, while Elvish Mystic should be somewhat less, as it is quite fragile, so maybe 0.25 lands. Artifact sources could be somewhere between these two depending on your meta.

If you want to go deep with this, take out the hypergeometric calculator again and check whether you can play your Farseek reliably enough on turn two. Then just take that probability and count your spell as that many lands. You can do the same thing with Cultivate as well. The thing to remember with all of this is that if you have three lands in your opening hand, that actually makes it less likely that you have the spell in your hand, because now three of the seven spots have already been taken. So, simulate this by looking at various hands on some software which lets you goldfish, for example.

Now, just because I haven’t made all of this complicated enough, let’s add in some card draw. Baleful Strix is actually one of my most played cards in EDH, because while you don’t generally want to draw Nature's Lore on turn 14, the Strix might still have a meaningful presence on the battlefield and it will draw you a card. Altogether it’s an extremely efficient card. If I have 40 lands in my deck, that one card draw is going to be a land a little over 40% of the time, so I can count it towards my mana just a little bit.

I also often play cards like Read the Bones and Secrets of the Golden City, because they can smooth out my curve. They can help me draw the lands I need early, but they are still good later in the game.

Let’s have a deeper look at my Brokkos deck. I don’t usually include lands in the decks I publish, because I just expect everyone to use whatever duals they have, but here is how this specific deck ended up for me. (Note: This is an example of how to approach this. I’m not claiming this is optimal, as I know I cut a Triome and a Castle just because I needed them elsewhere and I just don’t have the original duals anymore, or some of the fetches.)

I started out with 10 basics, 6 Forest, 2 Islands and 2 Swmps, so that I’ll have enough targets for my ramp. I’m emphasizing green, because most of the rest of these cards require green, so that’s the color I’m most likely to need. Also, historically, there are more allied color duals, so in this particular combination of colors, green sort of gets pushed out by the other two, so I’m trying to fix that imbalance as well.

Starting with the lands that can be produce or be converted to any color I need:

You shouldn’t really count all of these as sources for each color, because once you’ve committed to a certain color or combination, that’s it in some cases, and you can’t always know what you’ll be needing in the future. Still, it’s easier this way.

After that, the duals (I tried order these UB, UG, BG for easier calculations, but the new plugin I’m using decided to alphabetize them):

That’s 10 basics + 5 tricolors (sort of) + 23 duals (11 UB, 6 UG, 6 BG) = 38 lands, which gives me 23 G, 24 U and 24 B. I don’t really like that, so if I had more good UG and BG duals to add, I would definitely cut some UB duals for them. However, the numbers are pretty good. On the other hand, later on when we look at the ramp, some of those cards specify Forests, so in that sense I’ll often have more green sources on the battlefield.

I left out two lands at this point:

These are there for the utility. You shouldn’t have too many of these, but some are fine as long as you have all the colors you need.

Cards that ramp me:

The problem here is that I just don’t have enough 2-drop creatures t play the mutate creatures early reliably enough. Still, most of my ramping is going to be through creatures, because the deck requires a lot of bodies to mutate on. (One human managed to sneak into the deck, but I guess we’ll let that slide.)

Other cards that find me lands (always or sometimes):

Again, I emphasize creatures here, even though I did decide to add one sorcery here, but just because it’s so useful later in the game as well.

Cards that help with the mana, but which I’m not including in these calculations, because they are not helpful in the early turns:

If you’ll count all of these together, you’ll get 68, which means that over two-thirds of the deck contribute to my mana in some way. This might seem like a lot, but at the same time, you need a lot of mana and you can design your deck in such a way that these cards do something else as well. Some of these are great value engines or even win conditions on their own. Think of how you can maximize your benefits from your ramp cards.

TL;DR: Play at least 40 lands, unless you really know what you are doing, and play a lot of ways to ensure you can draw those lands or otherwise get them on the battlefield. Also, Richard from MtGGoldfish had a real nugget of gold, you should always follow (paraphrasing): What’s the point of playing your two mana ramp on turn two, if you can’t follow it with a land? Therefore, why not just play a land instead of the two mana ramp in this situation? Or play both, the two mana ramp and more lands.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.