Superior Numbers - Land Counts
Editor's Note: We've re-filtered the way the below data was acquired. A lengthier explanation and updated evaluation on average card type distribution can be found here.
This land was made for you and me
The average deck on EDHREC is running 29 lands and four mana rocks. Is that number too high or too low?
Welcome to Superior Numbers, where I try to do numerical analysis on cards and deck building trends using slightly more math and slightly less snark than in my usual column.
The first few times around in this series we took a look at individual cards, but this time around we're going to go a few levels higher and take a look at the average amount of lands being ran in EDH decks in our database.
This stats dive was prompted by the last column I wrote, where I examined Vedalken Orrery. Over the course of goldfishing a hundred or so hands with random decks online, I noticed that I was unable to hit four colorless mana by turn four way more often than I would have thought, and by and large this seemed to be due to a dearth of both lands and mana rocks in the decks I was testing. How to find out for sure though?
Don Miner to the rescue!
EDHREC mastermind and unrepentant stax player Donald Miner did a little custom stat parsing for me, and pulled the following information from our database:
- The average deck on EDHREC is running just over 29 lands.
- Decks added in the last year average 31 lands.
- The average deck on EDHREC has 4.15 mana rocks.
A couple of notes about those stats:
- Some decks choose to not run any mana rocks, but that doesn't mean they don't have other ramp options. Looking through my own list of decks, for example, my enchantress deck has zero artifacts, and thus zero mana rocks, instead electing to run enchantment-based ramp like and . Decks like this will drag the overall stats down while still being able to ramp effectively.
- Landfall or other themed decks may also run less rocks. Again, looking at my own decks, my Landfall list is running a single rock - - because the deck is designed around generating Landfall triggers. The ramp there is skewed heavily towards green's land ramp spells like and . Green is also statistically the most popular color in Magic, and its native land ramp probably also somewhat depresses the mana rock count while still leaving a deck with enough non-rock ramp to function.
- Mana dorks like were also not figured into the equation, giving us yet another ramp archetype that allows a deck to function while dragging our above statistics lower and lower.
- Small draw spells and filtering like , , and also make it easier to find the lands and rocks in a deck with a lower land count.
Additionally, cEDH decks tend to run leaner land counts thanks to the decks having generally lower curves and shorter game lengths, and there are plenty of cEDH decks showing up in the database. I asked Cameron of the Laboratory Maniacs YouTube channel if he could tell me succinctly what the rough guidelines for a cEDH deck were, and like the overachiever that he is, he wrote me a detailed, multi-page breakdown of the cEDH theorycrafting behind lands and mana sources. It's almost a separate article unto itself that he should write for us, but for brevity's sake I'm just going to reference the most interesting part: this breakdown grid of the lands and mana rocks ran on four popular cEDH decks:
The Breakfast Hulk deck has the lowest land + mana source count, sitting at 33, which is comparable to the 29 lands and 4 rocks we saw above. Jeleva and Kess Storm are a bit higher, and Food Chain Tazri tops out at 28 lands and 43 total sources. Look at the average CMC of those decks, though: 1.7, 1.92, 2.04 and 1.68. That is radically lower than the 3+ CMC average of decks in our database.
It's free real estate!
Number of lands = 31 + amount of colors in commander's color identity + commander CMC, with 0-CMC mana rocks counting as lands.
We're seeing numbers nowhere near that formula in decks nowhere near the average CMC of Cameron's cEDH lists. 29 lands and 4 rocks feels really, really low, even if you do have land ramp and enchantment-based mana production and mana dorks.
It feels too low, but it it truly too low? Let's goldfish out 100 random decks and see what we find.
Some ground rules: I'll mulligan away any hand with only 1 usable mana-producing land (, , , and other non-mana-producing lands wouldn't count as lands). I will only keep a hand with 2 usable lands if I'm able to cast a mana source or a way to fetch another land. or would count, as would . Additionally, if there were any castable draw or filtering spells like available, I would cast them to hit a land if needed.
- 26% of the decks tested missed their T3 land drop and were unable to play any mana source within the first three turns.
- 21% of the decks tested missed their T3 land drop but were able to play a mana source, i.e. two lands and a or or fetching a land with something like .
More than 1/4 of all decks tested just flat out missed their T3 land drop. That is terrible, and it will make you lose games before they've even begun.
So how many lands and rocks should you run? Certainly more than 29 and 4. Nate's formula is a good place to start for most decks. If nothing else, just up your land and rock count to the point where you can consistently make mana equal to the turn number. Regardless of how you get there, the current average numbers are far too anemic for anything but the most tuned of cEDH lists. There's a reason preconstructed Commander decks come with an average of over 37 lands!
Thanks for reading! I'd like to leave off with a new decklist of mine for a high-CMC commander, where land counts and mana sources are especially important. Hopefully I've hit on a good number of mana sources for the deck to play properly, but just like the experiment above, playtesting will be the best way to tell!
If you have any suggestions for other topics to cover in a future Superior Numbers leave a note in the comments below!