How many lands is correct (on average)?

[Jemolk]

Voted 37-38. My deckbuilding process involves the starting assumption of 38 lands, and I occasionally go down to 37. I almost never drop to 36 (I have one deck that runs that few, IIRC). Mana screw is miserable, and only really avoidable with a good pile of lands in my experience. I avoid flooding out as best as possible with a good bunch of utility lands in the mix. Then, some rocks and a decent chunk of card draw gets me the rest of the way to having adequate mana. I don't really ever build a deck from 99 cards, even. I go with 61 nonlands plus commander, and then fill in the 38 lands. Easier to avoid the temptation of cutting too many lands that way.

[Treamayne]

Hmmm...I use the aforementioned Turntimber Symbiosis as whatever I need at the time. I generally won't hesitate to play it as a land, especially in the early game. But later on, the spell side is pretty good. I count them as both when I'm analyzing such things.

Sounds like it works for you. I tend to only count things as "both" when I can always get both (e.g. Qasali Pridemage counts as a creature and removal); but for MDFC I count the preferred/most-likely version. For XLN style (e.g. Arguel's Blood Fast // Temple of Aclazotz) I'll only count both sides if transforming is likely (as per previous example, the chances of going below 5 without dying are so slim, I never count the Temple as a land in my land count - if it flips then "Bonus!"),

[hyalopterouslemur]

Objectively correct answer: it depends on your curve, what turn it becomes acceptable to miss land drops, how much ramp you run, and your ability to draw cards.

I answered 35-36, as I've found that to be the sweet spot for me, but I primarily play control with strong draw engines. I'm thinking of bumping up Shadowheart to 36, but cuts are so hard right now.

Besides curve, the other relevant question: Do you have any mana sinks or x spells? Those things generally can become a wincon in their own right with enough mana.

[TheGildedGoose]

Objectively correct answer: it depends on your curve, what turn it becomes acceptable to miss land drops, how much ramp you run, and your ability to draw cards.

I answered 35-36, as I've found that to be the sweet spot for me, but I primarily play control with strong draw engines. I'm thinking of bumping up Shadowheart to 36, but cuts are so hard right now.

Besides curve, the other relevant question: Do you have any mana sinks or x spells? Those things generally can become a wincon in their own right with enough mana.

Very true. As a pilot of a big mana deck for years with a mana sink in the CZ and several X spells, I can attest to this being an important consideration.

[Legend]

36-39 because whenever I've tried to get by with fewer than 36 lands, I inevitably get mana screwed.

[Mookie]

My default is 38. I often end up going to 37 in final cuts (because finding slots is hard), but I always feel bad about it. Some of my decks have drifted lower over time (due to having lots of card draw / filtering / low curve), but it's actually pretty rare for me to go higher - even if I have a deck particularly high on ramp / land-fetching, that just means I increase the percentage of basics instead of adding more lands. I suppose that could change if I were running lots of effects that gave me extra land drops, but if you're mostly doing Cultivate and friends, I don't think you need that many more total lands than normal.

[BaronCappuccino]

I voted 37-38. If my commander draws, tutors or otherwise filters, I'll up that quite a bit to ~43, because the most important cards to hit are early game lands, and the deck naturally takes care of the rest. My current deck is huge on treasures, and the average CMC is ~3, so I'm experimenting with something lower with 34 lands, but it's an outlier in decks I've made, again, mostly because it's so big on treasure production.

[capitacommunist]

For me it really depends - I have decks that are as low as 30-32 (Tymna, Adeline) which have very low curves and/or lots of draw power and ramp. But I also decks with 40-42 lands with almost no artifact ramp.
In a typical EDH deck I would not play more than circa 46 mana sources - with addition of card draw spells you otherwise lose too many slots without any action or reaction, and your card draw spells also become weaker as a result. I voted 35-36.
On a side note, I always play Tithe in all my white decks replacing a land slot, and would consider Land Tax as a land slot too.

[Yatsufusa]

I run the 36-37 spectrum myself. Back in the really old days you would start around 40-42 than start deducting for ramp, but nowadays ramp is so ingrained in the format it basically accounts for its own section and you don't start reducing land counts until the ramp gets excessive (and even then you'll consider trimming the ramp instead). The actual land-trimming happens after you playtest the deck and start tuning the mana-curve of the deck (usually towards the leaner end) and can start shaving a couple of lands if your curve permits for it.

37-38 sounds about right, I only tilt down a single land because all my decks runs Mana Crypt and Sol Ring, which while part of the ramp package numerically, are the two premier rocks strong enough for me to shift the entire spectrum down by 1 count. And even then I've had years of fine-tuning so cards like MDFCs can easily replace a land without endangering the curve of the decks, which brings the practical count to about 36 (and in decks without MDFCs, the curve still lets me overlook "non-land lands" Maze of Ith and Glacial Chasm instead).

[PrimevalCommander]

If we don't count lands that don't tap for mana, then 36 is my golden number. If I'm adding a land that doesn't tap for mana (Maze of Ith), I'll add it on top of that number. Yes, I still play Maze

Mono Red artifacts with about 15 rocks plays 35 lands. Was 34, but I kept getting hands with 1-2 lands and no rocks, go figure. Paradox Engine ban just turned into a Mountain.
Mono Blue draw-go plays 34 lands with 10 rocks and 12 draw spells. I don't typically have trouble finding lands. 1/2 the deck is either mana or draw.
Mono Green lands plays 38, which is low for a lands deck, but I like jamming in too many other cards. That is about as high as I could go and play what I wanted.

Everything else 36-37 and I'm usually very happy with the performance at that average. I have started asking others about their land count, and 33-34 is something I hear all too often, and they have a habit of stalling out mid game.

edit: I voted 37-38 because 36-37 wasn't an option, and 35 is too low.

[hyalopterouslemur]

Objectively correct answer: it depends on your curve, what turn it becomes acceptable to miss land drops, how much ramp you run, and your ability to draw cards.

I answered 35-36, as I've found that to be the sweet spot for me, but I primarily play control with strong draw engines. I'm thinking of bumping up Shadowheart to 36, but cuts are so hard right now.

Besides curve, the other relevant question: Do you have any mana sinks or x spells? Those things generally can become a wincon in their own right with enough mana.

Very true. As a pilot of a big mana deck for years with a mana sink in the CZ and several X spells, I can attest to this being an important consideration.

And besides all that, even just a lighter deck, I always say most decks don't have enough lands (by which I mean mana-producing lands), removal (especially the instant variety) or card draw. Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

[PrimevalCommander]

Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

I think this is a bit extreme. Even in budget 3 color mana bases I'm only playing ~12 basics. In 3 color green decks I'm only playing 10 basics with lots of dual-typed lands that are Forests. I devote most of the rest to color fixing, especially in non-green. Unless I'm actively playing with someone I know has a hard-on for hosing non-basics, the benefit far outweighs the risk. I can count on 1 hand the number of times I've been punished for a high non-basic count, even when I played a ton of commander. Usually if someone hates on lands, they hate on all of them, but not always.

story time: I recently got hit by a turn 6 Destructive Force, but was able to draw out of it with the help of a regenerated Iymrith, Desert Doom and a sufficient land count to make my land drops for the next several turns. Ended up winning that game after the D-Force caster ate Dragon Fire (lightning?) 4 turns straight and I curved out better than the rest of the table.

[TheGildedGoose]

And besides all that, even just a lighter deck, I always say most decks don't have enough lands (by which I mean mana-producing lands), removal (especially the instant variety) or card draw. Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

There's another consideration for running a density of basics: check lands. If, like me, you don't have fetches or OG duals, check lands are top tier fixers and need enough basic land types to function properly.

[NZB2323]

I voted 37-38, but most of my decks run less than that.

Do we count cards like Boseiju, Who Endures as a land card? What about Malakir Rebirth // Malakir Mire? Can we count Search for Azcanta // Azcanta, the Sunken Ruin or Legion's Landing // Adanto, the First Fort as a land card?

[RxPhantom]

And besides all that, even just a lighter deck, I always say most decks don't have enough lands (by which I mean mana-producing lands), removal (especially the instant variety) or card draw. Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

There's another consideration for running a density of basics: check lands. If, like me, you don't have fetches or OG duals, check lands are top tier fixers and need enough basic land types to function properly.

Check lands are in every single 2-3 color deck I'll ever build. I have fetches, too and that just makes them better. They're really good, and they've been reprinted like crazy, especially the allied ones.

I voted 37-38, but most of my decks run less than that.

Do we count cards like Boseiju, Who Endures as a land card? What about Malakir Rebirth // Malakir Mire? Can we count Search for Azcanta // Azcanta, the Sunken Ruin or Legion's Landing // Adanto, the First Fort as a land card?

I count the channel lands as lands. Same for ZNR lands that can come in untapped. I'm still not sure about the tapped ones. And I guess the Ixalan ones depend on how reliably you can transform them, but I'm gonna say no. I would count them as a spell, I think.

[3drinks]

I know when I played Nicol Bolas|chr, I played 42 + maze & tabernacle. But that's a control deck that aimed for the super late game as well as played a dopey 8mv commander. That's by far been my high end.

My lowest being ~34 in Alesha because I can function just fine on 3-5 lands all game as I play a game of value-over-time™ with moderate threats than six+ drop haymakers.

[NZB2323]

And besides all that, even just a lighter deck, I always say most decks don't have enough lands (by which I mean mana-producing lands), removal (especially the instant variety) or card draw. Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

There's another consideration for running a density of basics: check lands. If, like me, you don't have fetches or OG duals, check lands are top tier fixers and need enough basic land types to function properly.

Basics are also good for cards like Choked Estuary, Sunken Hollow, Tainted Isle, Castle Vantress, Mystic Sanctuary, Land Tax, Deep Gnome Terramancer, Loyal Warhound, and Knight of the White Orchid.

[Jemolk]

Even outside nonbasic hosers, I find that the upside of being able to run Thawing Glaciers and Terrain Generator more than justifies running a solid number of basics. Even in 5-color I generally run a minimum of 10 basics. Color screw has happened occasionally as a result, but it's far rarer than the payoff being quite beneficial in my experience. Of course, this is in the context of my specific decks, which are incredibly slow, grindy, and card-advantage-focused, so your mileage may vary wildly.

[PrimevalCommander]

I find Terrain Generator to be exceptionally underwhelming recently. Even in a deck that draws a lot of cards, and holds mana up for interaction, I still have a very hard time finding the mana to activate, and the land ETB tapped of all things. Thawing Glaciers isn't much better due to how painfully slow it is. Even when my meta was turn 15 battlecruiser bashing 8 drop creatures against each other, Thawing Glaciers was noticeably slow.

I still have Terrain Generator in mono blue because I have used it on occasion, but it will be kicked to the curb if I every need the utility slot for something else. I'm not playing it anywhere else than heavy permission decks that keep mana open on opponents turns. Otherwise there are better uses for my mana.

Basics are also good for cards like Choked Estuary, Sunken Hollow, Tainted Isle, Castle Vantress, Mystic Sanctuary, Land Tax, Deep Gnome Terramancer, Loyal Warhound, and Knight of the White Orchid.

Many of these can be triggered or tutored by dual typed lands. And there are an increasing supply of affordable dual typed lands as well (finally). Though I can see several of these would appreciate a few more basics. In a 2 color deck I would play more basics, but at 3 color and up, I can have many of these active with the limited selection of cheap dual-type duals, and even Triomes have a full cycle. I guess it bears mention that currently I only play 3 color and mono-color, so my basic count is skewed by the number of colors in my active decks.

[hyalopterouslemur]

And besides all that, even just a lighter deck, I always say most decks don't have enough lands (by which I mean mana-producing lands), removal (especially the instant variety) or card draw. Within the land slots on your curve, you need to diversify your mana (with nonland permanents that produce mana, as well as mixing both basics and nonbasics if you aren't playing something like Hermit Druid combo; I recommend half of all lands, bare minimum, be basic, if for no other reason than to avoid Ruination, Blood Moon, Back to Basics, Primal Order, Price of Progress, and other lol you did that one to yourself land hosers) as well.

There's another consideration for running a density of basics: check lands. If, like me, you don't have fetches or OG duals, check lands are top tier fixers and need enough basic land types to function properly.

True. And there's also ramp spells. You can either ramp to basics (in green) or to basic land types (in green, or in any other color, but only to their basic land type). There are a few things like Burnished Hart, but, you know, 6 is a lot for Explosive Vegetation.

And always remember, "enters tapped" is usually the absolute worst downside for a dual land, except for "when this is tapped, put a counter on it, as long as it has a counter it doesn't untap during your untap step, remove a counter at the beginning of your upkeep" or whatever the hell is going on with storage lands from Fallen Empires/Mercadian Masques. If it's an easy hoop to jump through (e.g., checklands, Lorwyn tribal lands in a tribal deck) or is simply a modal downside (shocklands), that's one thing. And I am willing to accept an "enters tapped" for something like Bojuka Bog or the temples where the upside is just that good.

Manabase crafting. It's so complicated.

[Cyberium]

I do 35. Less, if I use more rocks and double face lands.

[Maluko]

I nearly always run 40 lands in my decks for three reasons:

• I appreciate the consistency between decks;
• I run few or no mana rocks;
• I don't want to worry about whether my land count is appropriate for the curve or not, I just want to build decks.

That being said, I've run 37-38 lands in the past in some decks that had an obvious lower curve. But this is rare nowadays, because I realized the decks that I play with more skill and more enjoyment are midrange-combo decks that win with an epic bang rather than by chopping off my opponents' life total piece by piece.

[Jemolk]

I find Terrain Generator to be exceptionally underwhelming recently. Even in a deck that draws a lot of cards, and holds mana up for interaction, I still have a very hard time finding the mana to activate, and the land ETB tapped of all things. Thawing Glaciers isn't much better due to how painfully slow it is. Even when my meta was turn 15 battlecruiser bashing 8 drop creatures against each other, Thawing Glaciers was noticeably slow.

Sure, they're slow, but being near-guaranteed to never miss a land drop for Thawing Glaciers, and repeatable ramp from Terrain Generator have both been exceptionally valuable. I definitely lean more reactive than most, but you hardly have to be mono-blue counterspell tribal to get fantastic value out of them. In fact, I expect it to be better in a more removal-based control shell than heavy permission, since you have more flexibility for timing your answers with removal than counterspells.

I'm also kinda baffled by the suggestion that there are always better uses for your mana than repeatable 3-mana ramp. You can accelerate quite significantly off of even two activations of a Terrain Generator. What are you even doing with three mana that ramp isn't a solid option? Particularly when the deckbuilding opportunity cost is so low.

[materpillar]

I usually start out my deck with 37-38. Cut a few lands over the coarse of a few years for more fun cards until I'm down to 33. Then, I get mana screwed 2 games in a row and check my land count. I realize the count is too low and add a handful of lands some I'm around 37-38 again. Other times, I have lower curve decks that I keep cutting lands until I'm down to 33 and it works fine. Hello Lazav, the Multifarious.

Also, flip cards have greatly increased my overall land count and moderation.

[BlackbirdPlaysMTG]

I don't really have a set-in-stone number for the amount of lands I run. The number of lands in my decks is dependant on curve, gameplan and the amount of ramp.

My decks:
- Sythis, Harvest's Hand runs 33 lands
- Ezuri, Claw of Progress runs 35 lands
- Nissa, Vastwood Seer runs 35 lands
- Titania, Protector of Argoth runs 38 lands
- Yeva, Nature's Herald runs 37 lands