Honestly this video is pretty dumb. Its depends on what your deck is. Running my karn legacy reforged i could probably cut out a bunch of lands as he ramps my mana quite a bit. Depending on your mana curve. What kinds of things your deck needs to do. The list goes on and on. Weird you say its the perfect amount for you in the video. Then go on to say its what everyone should run in your title. Go back to begging on the streets. Im sorry your parents never gave you enough attention.
@@gauwal another issue is that most mana rocks cost a certain amount of mana themselves, and if you're missing land drops because you've cut too many lands, you might not be able to cast the rocks either. This obviously scales inversely with the cost of said rock (even with 28 lands you can still cast Sol Ring on turn 1 about 100% of the time) but the effect is exponential; in a deck with fewer than 43 lands, a Sol Ring is worth about as much as a colorless land, an on-color signet is far worse than a dual, and high cost rocks like Relic of Legends are nowhere near worth a land drop. Lily mentioned other issues like rocks costing your mana for the turn. I agree with her assessment that rocks and ramp spells are for just that, ramping to expensive commanders or combos or Storm turns or whatever ahead of curve, not just because they have a tempo cost for setup but because you also need lands to even pay that cost.
I run 34 lands and like 18 rocks, getting to the 3/4ish mana I need is no biggie. But my deck is just an Esper tutor shell for thoracle. Side note: In "casual" commander is it as important to hit a perfect mana curve? Like really, if you miss a few drops who cares, its not like you're gonna get killed or anything anytime soon. If you play a bit sub optimally people will pity you. And its SUPER curve dependent. Commander isn't as linear as other formats. You have high odds of getting wiped, or setback anyways, meaning playing on curve will lead to be just as slow as if you missed one or two drops. If your game is going to turn 8, 10, or even like 20 turns, missing a couple lands won't change anything. I kinda thought the point of "casual" commander was to play weird decks that take a while. From my small understanding on kitchen commander, doesn't the person who pops off first usually get beat down? And what are you trying to curve into? Do you HAVE to have a 5th land on turn 5? Can you not spend a turn dropping a sol ring/arcane signet, or casting a Kadoma's Reach? I may just not understand enough about casual gameplay. I can only think of a few cards I'd even use that are above 3 mana (FOW and the other free spells don't count as their actual cost in this case). What are you rushing to? An approach? A bola's citadel? Craterhoof? Tooth and Nail? I'd assume you'd want a boardstate before casting those so you're already not playing them the turn you get the mana too. Usually when one of my friends explains their kitchen decks to me its like 30 synergistic cards and you can win with like 20 of them on the field and you'll get infinite turns or dozens of extra combats or 100m tokens and skullclamp them to draw your deck until you hit this pet card of yours that you can then cast and it makes the heavens sing and the winning lottery ticket falls onto your lap. You're already playing off curve, you're already playing janky sub optimal decks. you're trying to see who can assemble a Rube Goldburg machine without getting interrupted first. Your Atraxa superfriends will probably function if it misses a drop or two and your sage of hours will probably get to its 5 counters in your ezuri deck. You already chose to walk the path of the inefficient, just live it up with your 31 lands and free muligans and whatever.
I really like this video and think it provides a lot of useful information about alternative ways to look at mana bases. I think one of the most interesting things magic has done in a long time is make lands way more dynamic and useful in various ways. I actually enjoy the push and pull of non-basic/basic/utility lands when building a mana base for my decks and often try to use my lands to help shore up weaknesses in my mono color or two color decks. I feel like a always start with 38 basics and then slowly start to remove basics as I add mdfc’s, utility lands and color fixing lands until I get down to like five to ten basics depending on how many colors I’m in.
My typical decks run with 36 to 38 lands. The video starts from a premise that the average curve of the deck or at least average card is above 3 cmv ... my decks tend to have an average which is below that. Couple this with an amount of cheap draw spells and 43 would just be too high, as you would flood your hand. The thing with 'the best amount of lands in a deck' is, that there is no one single answer as it depends on the deck. I have landfall decks which do run 44 or so lands, I also have my Jhoira artifact storm deck, which runs 28 lands only (although that deck is illegal since the banning for jeweled lotus and mana crypt... but I refuse to take it apart... I only play it on very rare occasions with close friends anyway). So no one is ever going to convince me 43 is the 'golden' number... plus we all know the answer to the most important question ever is 42 and not 43 :P
Well, there goes trying to have a “nonland mana cards to land count” ratio to have 50 cards that care about mana bases while having 49 (or 48 for partner commanders) open for your playables.
I want you to know that after I watched this video I sighed like a 16 y/o who was just told to clean their room and took out a 6 drop for a 37th land from a deck 😂
Clearly the answer is 96. With one copy of treasure hunt, one Ad nauseam, and one thassa’s oracle. Gyruda doom of depths as the commander to throw people off.
what if Treasure hunt + Ayula's Influence tho? (discard a land, make a 2/2 bear) if you also have a reliquary tower in there, you can just make bears on your opponent's turn when everyone's tapped out, and say "bear with me" as you spawn 90 2/2 bears before your untap step, then proceed to say "hor hor hor hor" as your opponents grumble and scoop.
The method from Karsten's Feb 2024 article that Rebell references is time-consuming, but more on point. Especially if your deck plays a ton of cantrips, eggs, mana rocks, rituals, and MDFCs, which Karsten weights differently from either normal lands or just any old spell. My Grixis storm deck is at 33 also
I've been playing since 2001, and we've always known the "40% rule." It was actually hard to get into the habit of playing fewer lands when I first started playing EDH around 2010, but 37 seems to be the sweet spot for most decks, with 38 being the most I have ever run. With the unique dynamic of a commander table, I've found in almost 15 years of playing this format that you don't realistically get screwed or flooded more often than you expect when running fewer lands than what is mathematically optimal. You DO, however, get to play with a few extra really cool cards, and that is priceless.
The problem with the 43 lands amount is that you also don't want 50% of your deck to be mana sources. Unless you have a special case, an EDH deck should typically have 55+% of the deck dedicated to strategy and the rest dedicated to mana base. If I use 43 lands, I wouldn't be able to put more than 2 mana rocks in the deck. The calculators used by those pro players works because they aren't casting signets or additional mana sources. They are going right into their strategy, which puts more dependancy on lands for mana. So for them, it is absolutely important to hit every land drop. This isn't the case in EDH.
@@Arctanis-vt3hl where are you getting the 45-55% ratio from? Essentially normal Commander decks are midrange ramp decks in every other formats. You need all the mana you can get.
I think increasing the "cards seen" part of the equation is underrated though? Drawing 1 extra card by turn 4 means you only need 40 lands instead of 43, and drawing 2 extra cards by turn 4 would mean 37 is enough. For example in a Sythis harvest hand deck you can easily draw na extra 2-3 cards by turn 4 just by casting your spells.
Sure, it's underrated for Sithis, but how many decks draw even remotely close to the same number of cards, especially that early? One card draw spell does not equate to an extra draw - you need to draw (and cast) that spell. So that means you need enough of a density of draw spells to reliably be able to cast them. According to the Frank Karsten article, it works out that you need ~4 draw spells costing 2 mana or less to replace a single land. So for Sithis, you can definitely cut a ton of lands because every 1-2 mana enchantment counts as a draw spell. But for your average deck, running 12 draw spells (which must cost less than 3 mana) only lets you cut 3 lands.
Hard land count recommendation make no sense. Land count should take into consideration the mana curve of your deck. The mana value of your commander and ofc the deck color/archetype.
99% of all decks are 5-6 Drop Legendary dragons or dinosaurs and the rest of the deck are 45 5-7 drop timmy creatures. Those decks want to run 42 lands and are what almost everyone is playing.
Something to consider is that since you always have your commander (and potentially a companion) available to cast, flooding out is not nearly as bad as getting mana screwed. I've been perfectly happy keeping many 4/5/6 land hands but have only 2 decks where I would even consider keeping a 1 or 2 land opener, both of which can win without ever getting above 4 mana. That doesn't make cutting gas for lands in my decks any less painful though.
You mention that ramp and card draw dont count toward the land base, but draw has a very different purpose. Forty three lands is calculated using a deck that only sees one card every turn, but the more low cost selection you add, the higher the sample size you are starting with. Faithless looting alone makes your opening hand + first draw a sample size of 10 at the cost of one card on the long term. Ponder adds up to 4 to your sample size, preordain adds 3. The math is alot more complicated than ponder = land but it will very quickly and easily increase your consistency while also being more useful in the late game than a landcycler that does very little in a real game of commander.
And it's not just blue cantrips and flooting, red has thrill of possibility +2 and all its clones, black has sign in blood +2 and nights whisper +2, white I can only think of land tax and green just has better ramp which can get you to harmonize +3 faster.
Plus the surveil and scry dual lands. I prefer to be able to draw, scry, surveil etc into actual gas but guaranteeing land drops is also a solid backup function xD
43 lands is assuming 1 draw per turn. If your commander guarantees you to draw an extra card every turn, such as on attack or end step then you'll effectively hit those lands earlier/easier and can probably afford to run less. Something people also forget is that ramp spells pulling lands out from the library reduce your remaining land count, making it harder to hit lands on the draw, meaning you should run extra lands.
@@TheTexasDice You shouldn't have kept a 2-lander though. 💀 A 3-land hand and nature's lore or other two-mana-ramp means you get to untap with four lands turn 3 and if your deck folds 90% of the time that happens, you gotta go back to the drawing board.
What's more important than the mythical deck-thinning is that in EDH everybody, including the starting player gets to draw 8 cards 100% of the time, instead of just 50% of the time (second player in 1v1) AND everyone gets a free mulligan. And in most casual pods, people just get to mulligan until they have a playable hand. This kind of disadvantages "honest" players who build their decks correctly. I tend to run 39+ lands in most of my decks (including a couple of mdfc, but I don't like to count them as a complete land because if I'm playing them as a land 90% of the time, I might as well put an actual land in the deck. -> mdfc should always be added on top of your minimum 37 lands in my mind) and I often have a playable hand in my initial draw or after the free mulligan. Meanwhile some players in my pod rocking the bare minimum of 37 lands or even a bit lower keep frustatedly mulliganing until they finally have a 3-lander or they keep a risky 2-lander after a couple of mulligans that might end up costing them the game. But if everybody gets free mulligans until they have a hand with 3+ lands, some players will just have a higher spell density because their greedy deck-building style doesn't get punished by the pod's leniency.
I've always used 40 lands and 10 ramp/mana rocks, for 9 years I've been trying to convince people that it's optimal and maybe one of the reasons I win so often
I think people undererstimate the fact, that your commander is your 9th card that is almost always available to cast. If your chosen commander is half-way decent, then having it in play will win you the game eventually (draws you cards, does the thing etc.) and if it's not in play (it got killed) then you likely want to be able to recast it despite the ever increasing commander-tax. Besides many lands letting you cast your spells more reliably (including "double spelling") and all that, being able to reliably recast your commander for a third time just makes even that 8th land a good "topdeck". I think more people lose to mana-screw than to mana-flood simply because it's way harder to be flooded, when you can always play your commander. So many games are lost, where a player has a full grip of cards in hand but only like 5 lands in play on turn 6.
Got a very similar concept. I'm usually using lands + ramp = 50-52 If I have a low curve and a deck that starts doing its thing early than ramp/ritual is more important than lands and if I have big drops (6+) I'm relying on hitting the land drops consistently and land count becomes more important. There's also the consideration of cards seen and draws. A deck that runs 10-15 low cmc card draws is more likely to see enough lands even with a lower land count.
Thank you so much for making this video. After watching Sam’s video I started playing more land in my decks but couldn’t seem to convince anyone that making a third land drop was better than 2 lands and a rock turn 3
99 lands, Commander is Garth One-Eye. And it's a fine selection of lands as well, such as Den of Bugbear, Fountainport, Khalni Garden, Kher Keep, Kjeldoran Outpost, Mirrex, Springjack Pasture, Urza's Factory, Vitu-Ghazi City tree, Barad-Dûr, and many other lands, such as manlands.
One thing i have found with lgs games is that many people follow a ‘gentlemen’s mulligan’ where you can mull basically any number of times to get a ‘playable’ hand. With the length of commander games and the multiplayer nature of the format I get the inclination, but it definitely reinforces the 37 land rule.
I seem to recall a video from The Professor where one pf the guys from the commander rules board said something along the lines of "the 'unofficial' commander mulligan rule is to allow enough mulligans until you have at least 3 lands in hand, but once you have that you must keep the hand", but also followed it up with "well, that can't be the official rule, cause then everyone will just cut down to 28 lands or less and then play super powerful shit they could only expect to run with a consistent starting hand"... which i guess is just what cEDH does anyways? Lol. The point is, if you are gonna allow that, then you really gotta trust your pod won't be willing to abuse it - or if they are, be ready to play a cEDH level game, lol.
@ between four players and up to two starting hands each, it greatly increases the number of games with sol ring in the opener for at least one player.
@@altejohlmao no, cEDH players mulligan down to four sometimes, but it often results in a loss. cEDH decks running almost no lands is sort of a result of the dockside meta where the goal was to cheat out or cast or clone dockside to accelerate your mana. Fast mana like the moxen also let you cut down on lands because you can get multiple mana sources out on turn 1 for a t1 rhystic study. Most lists even then still had 30 lands at least. It was only really Roghrak/Silas turbo builds that got some lists down to 24 lands. But it was normally around 27-29. Now, cEDH lists are looking around 32-34 lands a lot because the format is slower. The average mana value is also super duper low, reducing the need for lands. As long as I have several cards that can look for Esper Sentinel/Rhystic Study, I can start drawing 3-4 cards a turn.
I think that’s a terrible idea, and it’s the same as deciding to break any other rule in the game for the sake of it being more “fun”. If you are having to consistently mulligan your hand, there is something wrong with your deck and you should fix it. If you go by the gentlemen’s mulligan, you never have to address that there is something wrong with your deck because you can just keep fixing your hand.
Im curious why you didn't bring up bounce lands since they can allow a 3 land hand to generate 4 mana on turn 4 even if you didn't draw a land in those four turns. Thanks to Thunder Junction a two color deck can now run three copies of this effect (the perfect amount imo). They also have added utility with both MDFCs and cycling lands later in the game as they allow you to cast the spell side or cycle the land.
I didn’t want to complicate a pretty complex video already but that will be something I bring up in the next video for color fixing and duals, which the bounce lands fall into.
I love mdfc but I would only ever run the on color bounce lands. I care about fixing a lot so maybe you care about that less. The two colorless pips and general clunkiness of guildless commons and the otj equivalent make me really not want to play them.
Not sure if u did not mention them because they are bad or if you forgot about them, but you could run up to 5 bounce lands in a 2 color deck if you include the monocolored bounce lands. In my simic landfall deck I actually play all 5
@@devan9197 Yeah if you are playing three+ colors you dont want to run the colorless bounce lands since you can run minimum three of the guild ones, but most two color decks can afford to run them. If you're playing a two color deck and you have consistent fixing issues, the problem isn't the two colorless bounce lands.
@@shideon In two colors decks I have consistent mana fixing because I don't run those lands. Even if you can handle the fixing downgrade you also have realize that you're playing a tapland with tempo loss and you can't play it on turn one. Colorless taplands have to be really good to be playable and I don't think their worth it in two color decks. Mono colored sure you can throw em in there if u want.
Was a weird feeling for me. I clicked, remembering an argument two friends of mine had after one scooped for being land-starved. The other insisted he needed 36 lands. he looks at me for validation and asked "how many lands do you run?" I said "I think it's 46". So, for curiosity, I recounted them and it came to 43. Then I remembered it had been 46 a while back but I found it slightly land-heavy and I shed 3 of the least-preferable combination color lands to hit 43, and it had been at 43 for about 2 years. I fully expected this video to quote the classic 36 lands, and was hoping to learn something about why 36 is considered the standard, and then the author dropped the "43" Oh.... ok then.
The advice to use cyclers seems kinda weird. In EDH, I'm never going to cast Troll of Khazad-dum (or any of the others not named Lorien Revealed). You correctly point out that in EDH we want to build engines, and these cyclers don't really synergize with much of anything, or tap into the engine. So realistically, they read "pay 1 for a land". I always want my non-lands to either accelerate, answer threats, or advance my game plan. My gameplan is nearly never, "play over costed creatures with 1 keyword." Kyle Hill also made a joke during a CAH or EDHijinks that people should cut their cycling lands, because nobody every cycles them, and they are just tap lands. I agree with him on that. None of this controverts your claim that 43 is the magic number -- I'm just arguing against your proposed "solution" to cheat the number.
The idea is not that you cast Troll for 6 on curve as a normal creature. The idea is that you're in the late game, flooded out, head-to-head with another topdecking opponent, and you draw an evasive creature instead of a swamp. It's a reanimate target, a spell, a blocker, an attacker, it's a land you grab with Eladamri's Call, etc etc etc. It's just modal - nobody is saying the creature is "good" it's just that, in a lot of situations, a Swamp is worse.
@@colinfowler3021 Just as dead as any 90% of other land draws or six+ drops? Granted a six drop will likely be more useful later, assuming you aren't losing due to tempo loss already.
@@colinfowler3021paying 1 to fetch a fifth land is still cheaper then Farseek or Three Visits, which get tons of play. Sure you don't get the land in play, but in your case, you weren't making the land drop anyway, so there's no downside
@@colinfowler3021 you keep it in your hand for later, or you cycle it on opponent's end step so that you don't miss Land 5. By your logic is Valgavoth a bad card? What if you draw it on turn 4, or 5, or 6? You can't assess cards based on being able to extract maximum value out of them, exactly on-curve.
I usually run something between 32 and 38, but I think that's mainly due to the fact, that I usually run low curve decks and/or decks that are more likely to lose from mana flood than mana screw.
I would assume that it is because you don't draw if you go first in 1v1 formats, such as limited. this also happens 50% of the time so needs to consider the worse/smallest sample size.
the extra card might represent the ability to mulligan a low land hand in commander? having an extra seven means its not a downside, so it might lean heavier into the calculations than in a 60 card format
35 Lands - i feel like i'm constantly missing Landdrops, 37 Lands - i feel like i'm constantly flooded. 36 just feels right. Usually running 10-12 ramp spells. Of course there are exceptions (landfall, cedh, etc.)
@@falway5109 I'm running 40 lands as well in mine. Fetch effects + "land from the graveyard" effects _really_ get the snake going. Add in some "additional land per turn" effects and your hand is thicker than your library in no time. I also replaced almost all the draw effects with ramp spells, because Aesi.
I love these videos. I remember figuring out 43 was ideal with static paramators other than land count using the hypergeometric distribution calculator but it blew my mind realizing two of the other parameters could be altered by reducing my curve or increasing the number of cards i can see. I tend to run 42 sources of mana now a days with around 32 being lands 8 being rocks and 2 being some more arcane mana advantage effects. I almost always run a low curve with advantage engines and playing commander gives that free mull.
32 lands?? O.o That should mean that you regularly miss your land drops unless you have unlimited free mulligans, which is of course not what is generally assumed.
@lVideoWatcherl I'm not sculpting a hand with 4 lands and a rock. I want a hand with 2 lands, a rock and either an advantage engine, card selection or an additional land. The ability to see more cards allows me to draw into lands or an engine or to manipulate my graveyard for future value. The ability to consistently hit 3 mana sources is required to play the majority of my cards. I also Mulligan towards my gameplan so going to 5 can be acceptable as it allowed me to see 3 full grips of 7 cards. Seeing 21 cards although not guaranteed to be unique is still going to be around 1/5 of your deck seen before you play your first card. An example on how to think about this. When calculating for land drops consider the odds of hitting a rock, atleast 2 lands and an engine or selection card. If I play a looting effect turn 1, I look at 10 cards total with 6 cards in hand and 3 in the yard. If you fetch a surveil land that's 9 cards seen with 7 cards in hand and upto two in the graveyard. The ability to consistently see just a single extra card finds that next mana source at a rate more valuable than adding many additional mana sources. Also I can only play one land a turn so I tend to favor additional draw to find lands as it also finds engines, combos and interaction at the cost of actually expending your resources. If I ran landfall I'd favor more lands as they become spell effects or id favor more lands if I played cultivate effects.
@@Val-rak sorry but this just isn't good. 60 card, competitive formats with drastically more consistent decks have .33 lands as the floor. unless your decks are very low to the ground, there's no way this works out better than 36-40.
Every time Rebell makes a stats-like video, it reminds me that I'm not nearly as methodical in application of ideas. I'm just like," Well, I tried that, and it seemed to work okay." 😛
My only gripe with running extra utility lands is that your decks become less unique or explicitly synergistic. Like yeah, they'll be better, but less FLAVORED.
HA! You want to talk “flavored?” Try running 40 Lands while sticking through a theme that isn’t chock full of “staples.” (Now, granted, if you forgo Foils and card editions, my deck will cost WAY LESS than the $1000 price tag it currently works with…) but, keeping to such a land count, making certain that it stays on theme, AND it doesn’t go overboard with Staples is much more challenging - when “Skullclamp” isn’t in your 99 of a Marath, Will of the Wild deck. (However, getting the land base to work is still tricky, seeing that John Avon - Kamigawa Lands - *Retro (Foil)* - AREN’T A THING YET, WHAT IS WRONG WITH YOU WOTC-) (Ahem.) But creating a “World Map” theme is a super fun, curious journey to take. 37 Main. 3 Forest, 3 Plains, 2 Mountains, 5 Fetches, and quite a few duels and “utility(?)” lands? It’s actually kind of fun, when you think of your commander deck, and painstakingly build your land base with the theme.😅 Either way, (still waiting for that John Avon, Kamigawa Retro Foil Mount-) AHEM - I would love to know what you run in your deck. (I run “Wild Pair…” It’s a crime how the card is $2, yet runs the gamut of half your deck by just checking for creature stats.☝️😦 “I cast Primeval Titan (from my hand), it’s a 6/6, get Ojer-Tac - it’s a 6/6, get two lands, trigger Landfall - Twice, swing with Haste, get another two land(s), get landfall trigger - TWICE, again, cast (X Spell), then pass?” 🙂)
Somewhere after your commander’s mana value and average cost of spells in your deck. We hedge against that by placing more lands that do things so in the event you flood, you have sinks to take some action:
If you build your deck right, never! Mana sinks (mostly on lands or on permanents as a secondary effect) are very good in Commander. Cards like Warroom are a prime example for this.
Yeah, you're doing fine because of endless free mulligans. Completely cool if your playgroup is fine with it. You'll just run into trouble (non-games) more often, when you start playing with strangers in a store and they actually enforce the "up to one free mulligan"-rule.
@@artemi7 I don't have a play group I play at an LGS. I do the standard 1 free then its London Mulligan after that. We rarely have to use it, but I tell people they can mulligan and keep 5 if they need to. Maybe if you understand mana curves and actually balancing out your mana bases along with ample card draw and ramp, your deck might run more consistently regardless of how much land you run.
@@TheTexasDice The last time I missed a land drop was about 2 weeks ago in Grease Fang. I played a careening Minecart which was ramp in that deck's main strateg, I was able to get a few treasures to smooth out the game. Then hit more land drops and I ended up winning the game.
I just don’t know that Frank’s article really applies to commander in the same way that it applies to other formats. Your average commander deck has 10 to 15 pieces of ramp in it, which is something that is not true in other formats, commander decks also tend to have significantly more card advantage effects, since “dump your hand and beat face” isn’t really viable like it is in limited/60 card. Heck as soon as you get into older formats like legacy, which is a 60 card format these calculations go out the window. That being said I think 43 is it OK number I just think it’s a bit on the high side, somewhere between 37 and 40 is probably more “correct” in a format where every deck has Sol Ring plus 5-10 pieces of cheap ramp that doesn’t have any color requirements, and a decent amount of card draw effects, a lot of the time right in the command zone.
I'll compromise. I typically run 40-41. You say 43. Author Douglas Adams said the secret to the universe is "42," so that's the new number I'll go with.
@@RebellLily aha. Complete. Like the idea of using MDFCs & cyclers and do that often. Outside of that, I tend to use the Surveil lands as a way to "dig" ahead for a land drop if my hand is light.
Commie Commander here, Finally, someone out here praising the real gospel! I rarely drop below 40 anymore and I don't feel great if I'm not being given a really good reason to not LAND (hahhah ugh) between 42-45.
At casual tables, budget is a huge parameter for tapland percentage, especially because a lot of people want to play three or more colors regardless of budget. So once you grab the budget untapped lands, even I don't tend to find it enticing to spend 5+ dollars on a single untapped land in my 100ish dollar deck
If your pod isnt into proxying original duals and fetches for every Deck youll See those numbers. Most of my Group has 20+ commander Decks. Doing Goodyear unproxied landbases for those would cost us thousands.
Depends on the budget and colors, I'd say. I'm building a 75 bucks Ramos deck, so having 10 NON-taplands is more realistic, lmao. Meanwhile, a 75 bucks two color deck can probably run 0 taplands easily. Three colors in that bugdet is probably doable below 10, four colors... maaaybe? Haven't tried it yet.
It really isn't when you go for 40+ lands. Like, take your 3-color deck. You run your Triome, 3 Surveil Lands, 3-4 Bouncelands and probably one or two MDFCs that enter tapped (like Ondu Inversion or Bala Ged Recovery). That's 10 taplands right there.
Idk if it's just my algorithm, or it's land base winter but to copy a comment from a different EDH land video from a week ago which is heavily relevant to 3:36 in the video; The #1 commander on EDHRec right now is (still) atraxa, and it's average deck has 34 lands. The top 10 commanders are Atraxa 34, Ur Dragon (9 cmc) 35, Yuriko 29, Krenko 32, Lathril 32, Sauron 34, Kenrith 33, Edgar 34, Miirym 33, Kaalia 35. Not a single deck in the top 10 is playing 36 or more, and you have to go down to 17th to get to Velociramptor face commander Pantlaza with exactly 36 (the precon comes with 39). edit: Oh no, just heard 7:05
While I agree with your point, that 40 lands is probably more than most decks need, I'm not sure that EDHRec average decks are a good thing to point at as evidence. I suspect the average deck calculation *may* undersample how many lands are in decks.
@Dynme My belief is actually the opposite, sorry. The edhrec average is WOEFULLY too low. I didn't elaborate that, so I fully understand how it could look like that.
@@Temzilla2 Ah gotcha. I personally think most decks will function fine at 36 to 38, with 40 being "reasonable but maybe overkill" for non-landfall decks. But obviously there's a lot of variables that should factor into it.
@@Dynme Yeah I think that most of the time you structure your land count based on two major factors, when your draw engines come online, and what turn it's okay to start missing land drops with your goals for the deck. I.e. if you have the most powerful draw effects you can run in your deck, and it's okay to start missing land drops on turn 3, you can run 26 lands like some cEDH decks do, but if you aren't running a ton of low cmc powerful draw engines, and you need to be hitting land drops on turn 6+ you're gonna be looking at 40 or more lands.
Your content is great. I think I saw you while waiting in line at Mei Lai Wah on New Year’s Eve, but you were walking quickly so I didn’t want to bother you. Just wanted to say keep up the great work and thanks for keeping us all entertained.
*Me, playing 17 lands in my mono green commander deck, two of which don't tap for mana* "I'm subscribing, liking, I wanna hear Rebell talk about lands all year"
For what its worth, the lists I've tested with the most consistent land drops are 31 land turbo delirium money pile and 26 land dredge lands but those are some very specific deck churning archetypes that just care about four lands existing somewhere along the first 50 cards of the deck - sometimes not even that (untap source + fetch + w6 moment). And also just the mana bases are worth 300 bucks on each list which is why I don't actually own them in paper. But even as someone who likes having as few lands as possible in a deck, the number (43) tracks in a vacuum.
We can apply this to the previous "Cube Theory" and drop down to 7 x 8. That gives us 56 cards, 43 lands (or ways to cheat lands) and 1 commander. So either drop one face of the cube or reduce the number of cards on any face by 1.
Every tapland you consider playing should make you compare the land hit probability for one more turn with or without that land to make sure whether or not you should play it (as that land is coming into play this turn but will only be usable the next).
Honestly my gut feeling is if I took out the free mulligan rule me and my friends do for commander it would probably be a lot better to run more lands in my decks. Might bump up the number from 36 at the least.
Mulling to find ramp or important spells is soooooo much better than mulling to get enough lands in your opening hand though. Once you realise this, generous mull policy isn't even enough reason to play too few tbh.
"Hitting a land drop" is about efficiency of having playable mana per turn. As a base line we say one basic land a turn that gives you one more mana available each turn. Cultivate costs you three mana to cast plus a card draw (as if you top decked it). It gives (ramps) you an extra land on the battlefield tapped and one in the hand ; effectively a card draw as you still have the same number of cards in your hand at the end. Sword of the Animist costs two to cast and two to equip and on attack puts a tapped basic into play. So the first attack will let you ramp an extra land but used up four mana to get there. If it survives then the next attack nets an additional land with no extra mana cost. The point is using stuff like this to make land drops is not efficient as just drawing a land. So, the above examples are a poor way to hit a drop but a good way to ramp if you already are hitting your land drops. Card Draw is the biggest factor in land count. At 40 lands in 100 (I know 99, but easy math) you have 2 lands every 5 draws. You start with 3 in hand. At the end of 5 turns you should have held 5 lands given a perfectly spaced deck. This is where SD (standard deviation) would give you a more accurate percentage. Come turn 6's card draw there would be a 40% chance of getting a land and hitting your land drop. For simplicity lets say someone played a Howling Mine and on turn 6 you draw two cards you have an 80% (simplified math) chance of hitting that land per turn.
you're completely forgetting the value of scrying. being able to scry / draw a bunch of times consistently probably lets you get away with running less lands.
I am a HUGE proponent of playing closer to 42 lands. The thing most people dont talk about when it comes to playing more lands and complaining about mana flood is that in commander you always start the game with one guaranteed spell in hand (your commander). This means that I am much more willing to be slightly flooded. Especially in mid-casual games where the game usually goes a bit longer into the 8-12 turn range. And especially in a multiplayer game where often times its difficult to just card advantage out the table to win; you are usually winning by double spelling multiple times on critical turns. I find its very rare for me to lose a game with 0-1 cards in hand. Usually at the end of the game, I have a couple cards left in hand and I wish I had more mana to deploy them.
You are also encouraged to run mana-sinks as secondary effects on your permanents (or lands) in order to make use of excess mana. If you flood out, you can always use Warroom to draw a card or Dawn of Hope to make a 1/1.
Frank Karstens updated article actually accounts for cheap (2cmc) ramp and draw and comes down to 40/41 lands for a deck with average mana value of 3 minus 1 land per each 3-4 cheap draw/ramp. According to his formula most of my decks should run 36-37 lands but your individual results may vary. Karstens TL:DR: "A good formula for the number of lands in your 99-card deck, counting MDFCs partially in this fashion, is: 31.42 + 3.13 * average mana value of your spells - 0.28 * number of cheap card draw or mana ramp spells. This means that if your average mana value is three, which is fairly typical, then you should start with 40 or 41 lands and cut one land for every three or four cheap card draw or mana ramp spells in your deck"
Compulsion and Blood Sun are great budget cards for those color identities. In a mono, blue deck or a two tone blue deck compulsion essentially gives every card in your hand cycling, which, if it was phrased that way, I think that a lot more people would be playing it. Blood sun was seen as a throwaway rare in rivals of Ixalan but the fact that it gets you a card when it hits the battlefield as well as making it so that all of your lands, enter untapped can really level the playing field, not to mention the absolute absurdity that happens with bounce lands. As for the ratio, I did my own math a few years ago and I did come to a different number, however, I think that’s also due to the fact that I was looking at Man of value itself per land as supposed to lands per card. I will definitely however, point out that I’m rather certain that both the pros and myself we’re looking at 60 card formats. In my personal testing, I found following my formula meant that I needed way too many lands to play the cards that I wanted to play, which is probably why, instead of increasing the land count that vastly, I just lowered my average mana value, Average mv matters a lot more in deck building than I think most people realize. You can make a 60 card tech with 17 lands. If your average mana value is one. I do agree with you, however, that the pre-cons have their numbers off slightly, either that or they’re set for a slower gameplay. I think the grid that you had at the end of your is really something that new and advancing players should be using as a template, because Lord knows that even with years under my belt apply 60 card formats, making a Tricolor manabase was why I bought my first commander pre-con, I had no idea what I was doing.
Something I never see discussed is whether hitting every land drop every game is actually necessary? In the 4-player social setting, missing a drop every now & then doesn't seem worse than over-committing to the board early. I feel like sometimes we're too precious about land drops as a hold over from 60-card competitive. (Not talking about the player who's on 30 lands & misses 4 drops in a row.)
I play 60-card casual multiplayer. It still feels bad that I miss land drops early. You still want to get your stuff on the board. Having to discard because you don't have enough land and exceed your max hand size feels bad.
Hitting the first 4-6 or so even is huge and is the difference between double spelling most of your deck or not. I think if you consider it to be almost literally being able to play 2X as much Magic, you can see how important it actually is. Also would argue that leads to more fun gameplay even in a casual pod. But also most players run land light so you may not be seeing that contrast when entire playgroupls are simply running land light and the majority of those pods are missing more land drops than would be ideal
The command zone did a video on this or it was part of their yearly stats episode. Hitting every land drop is the number one statistical indicator of winning a game.
Missing your sixth land drop? Probably fine. Missing your third? Death sentence. Having your game plan be "Rely on the pity of your playgroup so you can be greedier" you're going to find yourself more disappointed than not.
Interesting watch! As someone that naturally gravitates to mono color low curve decks with mdfcs and land cyclers intuitively, it's funny to hear I'm just good, baybee
I run most decks at 34, anymore than that and I start to get mana flooded. I never run more than 36 unless the deck cares about lands then I don't go over 42. For my cEDH decks I run between 24-26. I keep my casual decks at 33% lands so on average I draw a land every 2-4 cards. I am pretty sure all "pro" calculations only consider having 4 lands by turn 4. If you think about turns 7-13 you will most likely be flooded and hoping to draw something not a land if you run 38-43. This is just my experience and how I build all decks.
do you play with endless free mulligans in your casual pods? 34 lands is terrible advice for most beginner decks, unless your curve is super low and/or you have tons of 1-2 mana drawspells and scry/surveil effects.
My personal rule after another video I watched is 37 physical land cards (physical meaning like actual, pure land cards). That video calculated 37 as the breakpoint number at which you have above a 50% chance to start with 3-5 lands factoring in one free mulligan. I do really like the 43 number as a target to futher pad my physical lands, though. I frequently add in MDFCs (and I love the Bounce lands), and Cyclers weren't something I'd considered. Sounds like 43 is a good target for all sources put together as a sort of "land-in-hand" count, separate from ramp and rocks and such.
I actually run 40 lands in all my commander decks at the start of the deck! It's great for testing and figuring out a direction for the deck before I fully commit to upgrading it. ^_^
Very interesting to see math and thought behind how many lands you want. I've actually had to rebalance a few of my decks recently because I didn't have enough lands in them, haven't played them out in actual games but the play-testing seemed a little better. It does ultimately come down to individual wants but knowing the 'best' number of lands is good to know, gives a good base. Also none of this matter to competitive players because that is a completely different beast. Think of it like this if casual EDH is a game of pickup basketball in a neighborhood then competitive EDH is like Naruto, you have to worry a lot more about BS teleporting and breaking the laws of physics. Remember competitive games are all about playing less Magic because your goal is to win as fast as you can, not have fun..
I don’t know. 40% is always enough for most decks that are built properly. As long as your mana curve is not too high or low. 3.4 is the average curve of a good deck, thus 40% land you should in 77% of games have 5 land by turn 5. If you have more than this the likelihood of being flooded is very high.
I think there is a weird psycological effect where people are much more afraid of mana flood than they are of screw even though it should be the opposite. People already understand why tapped lands are bad as they slow you down and missing a land drop is just that but a hundred times worse. At least if you're flooding out you should have things to do if your deck is built reasonably well. Tl:dr- Play more lands!
Flood IS worse than screw. With screw, drawing one or two more lands means you're free to unleash a hand full of pure gas. With flood, drawing one or two spells leaves you with... A spell or two.
To be fair flood is significantly worse than screw. If I'm drawing nothing but lands once the gas in my hand is gone that's it. In the case of screw if I'm stuck on 2 or 3 lands as soon as I start getting land drops i have a ton of stuff I can do. And in a game that goes long missing a land drop won't kill you
I go with 34-36. it depends on the mana curve, the colours, and style of play. 38 would be either a mana starved deck or a deck thats all about playing as much land as fast as possible (Like azusa) 43 is taking out a lot of utility, and bordering on mana flooding. I feel comparing limited to commander is a bit off, because limited is faster. EDH is all about that slower, bigger, flashier style of play, it also has bigger focuses on ramping
this is why i'm a fan of commanders who have abilities that let you cycle through your deck. I run a high mana base with a Raffine deck, raffine connives and once i have enough land, i start ditching the lands.
My Bant lands deck and my Dimir zombies deck both run 34 lands and 5 MDFCs. (Dimir collects rocks, and Bant Ramps). You've made me want to raise this to 42 lands. If I flood out I'm coming back hear to flame you tho 😊
Here's the title of the Channel Fire ball article I based my land and ramp/rock/ritual count on. What's an Optimal Mana Curve and Land/Ramp Count for Commander? By Frank Karsten
The 36 comes from a much more aggressive 14 lands in a 40 card deck, leading to around 20 lands in constructed and 35 for commander. It's not out of the blue, but it IS based on hyper aggressive play that's less applicable to casual commander than, say cEDH where you're looking for a very low curve win con.
Love this video! I'm a 38-42 land boy and my playgroup thinks I'm weird. I'd rather slightly flood than have "non-games" where I don't hit land 3 or 4 until turn 6 and have to twirl my thumbs for half an hour until I can join playing with the rest of the pod.
My (intoxicated) equation is begin with 50 (fifty) lands. Every 3 (three) mana rocks count as a land. Every 2 (two) draw more than replace count as a land. Every 3 (three) dork counts as 2 (two) lands. Every 4 (four) land search counts as 3 (three) lands. Every dual land counts as 1.2 (one and one fifth) lands. Every triome counts as 1.4 (one and two fifths) lands. Any quadrome(?) or all color land counts as 1.5 (one and a half) lands. Every wincon counts as -1 (negative one) land. I usually run about: 9 (nine) rocks. -3 ~2 (around two) draw. -1 6 (six) dorks. -4 4 (four) land search. -3 10 (ten) dual. -12 5 (five) triome. -7 6 (six) any color. -9 9 (nine) combo pieces that never stick, so -9. -9 This brings my total land count to 2 (two) lands in just about every deck I build. Check my math if you'd like. I have a couple beers in me right now. Taking suggestions, though!
Also keep in mind that you can get away with fewer lands if you are playing a deck filled with cheap spells or a commander with built in ramp or draw. This is a pretty good video for most commanders that don't have that and probably play a notable amount of 3 and 4 drops. MDFC's and land cycling are really good includes, solid advice.
Land Rule that applies to Casual and cEDH # of Lands = 31.42 + ( 3.13 x Average MV of deck ) - 0.28 x (Number of Draw spells and ramp spells less than 3 mana)
I have been playing a (joke) Ghazghkull Thraka deck I call Plateau which has 60 spells and 40 lands. Each cmc from 0 to 5 has exactly 10 cards in it for a flat plateau of a mana curve. One this I have noticed is that the 40 land count means that my deck always has something to do and can even make big plays later in the game despite only have 5 cmc cards at the top end. In fact, because my deck tops out at 5 cmc. It feels even more important to make sure that I play my cards on tempo as often as possible. Delaying my game plan and slow rolling will only put me behind compared to other decks that are establishing game winning big mana engines. A combination of a low curve, consistant land drops, and as many spells that refill one's hand or board keeps the deck relevant even into the later turns. I do cheat a little by counting Ascend from Avernus and other X spells as their default cmc for the purpose of deck building. This deck wins about 1/4 of its games, which is about par for a good casual deck I think, but consistently gets 2nd or at least stays relevant at the table for the whole duration of the game.
I think the mana base ahould typically be 35-40 lands. However depending on your play strategy, the counts can go up ir down. For example landfall decks can go up to 53 at the highest.
Very interesting video! I generally play a p low mana curve for my decks so my land count is around 30-32 including MDFCs. Since my commander and cards are p low to the ground I p much never struggle with mana.
I always played 39 or 41 with esper decks, people judge me, but esper don't have good ramps or a smooth mana curve to allow me to lower my mana base. Since I started doing it, I begin to win more games.
I don't have the numbers to back this up, but I feel like the average land count of opponent decks should be a variable in this formula. First I would assume that by increasing land density I would be increasing consistency in reaching a healthy mid game board state while reducing consistency in early game snowballs. If I am at a table with three greedy gooses with low consistency in seeing lands BUT A HIGHER DEGREE OF SNOWBALLING when they DO see those lands, I can't look at their decks as I would my own. I would need to look at the probably that at-least one greedy deck snowballs. Even given low 70% odds (or even 60%) of a deck to hit all lands by turn 4, if 3 other decks at the table have 70% odds, there is a 1 - (.3)(.3)(.3) = .973 AKA 97.3% probability that at-least one of those greedy decks will hit 4 lands. So that may be a factor in pushing you to play less lands even if not as likely to hit lands on turn 4. That's part of why I like to play less lands than 40 in low-to-mid power control decks, I need to have multiple spells availability in case I need to stop one early snowball. I can't just hit 4 lands and not have multiple pieces of interaction.
Dont forget to account for draw, not just ramp. With enough draw, you can make sure you never miss a land drop each turn, even if you're on the low side like 34.
Land count depends on your deck, simple as that. cEDH Rog/Si can rock 24 lands without an issue and is one of the best decks in the game. More control based cEDH decks run something like 28/29 lands and obviously a lot of fast mana. In casual is also depends heavy on your amount of draw and how low your curve is. I can 100% miss a landdrop or two if I just run more efficient cards and I will miss less land drops if I draw like 5 cards a turn cycle.
17 lands in a 40 card deck is for draft formats where you have next to 0 ramp and card draw is much more limited. I try to keep 50 pieces of “ramp” including lands in my deck which usually isn’t 37 lands and 13 ramp spells or rocks and works wonders
Most of my decks I built had 40 when I first started in EDH. However after I collected more efficient cards I did start to cut lands. But normally still run at least 38 for midrange decks.
You also need to factor in game speed and number of opponents into the equation. Commander games last longer than Standard and other formats in general, so you also have to balance the disadvantage of late-game land draws where you would prefer to have action. The mulligan rule also allows you to attempt multiple "first draws" to ensure you can land at least 2-3 guaranteed lands to start off the game. We are really talking about turn 4 and 5 as make or break. This is generally the point in the game when someone says they are "mana screwed". I've often seen these same players come back to win because they weren't perceived as a threat.
Thank you for this video! One of my goals this year is to explore going against common deck building practices and try new things out. This lines up with something I’d like to try
I mean with the amount of man of ramp that are index currently anyway, the magic number is 42 right, you start getting rid of like two or three of those for MDFCs, and soul ring arcane signet and all the other package that everyone uses and has in their deck, all the sudden it gets to 38 and it makes sense this video didn't tell anyone anyone didn't already know
man, you followed almost exactly the same thought process as me when I made and adjusted my first commander deck the past couple months. I saw the 17 recommended for 40 card decks and calculated 42-43 for 100 cards, then I found that's too much, so I ended up on 40 cards through trial and error
This video doesn't apply to cEDH yall keep cutting lands
I mean we can count mana positive rocks as lands, and most cedh decks run aroung the same number as many commander decks
Honestly this video is pretty dumb. Its depends on what your deck is. Running my karn legacy reforged i could probably cut out a bunch of lands as he ramps my mana quite a bit. Depending on your mana curve. What kinds of things your deck needs to do. The list goes on and on. Weird you say its the perfect amount for you in the video. Then go on to say its what everyone should run in your title. Go back to begging on the streets. Im sorry your parents never gave you enough attention.
@@gauwal another issue is that most mana rocks cost a certain amount of mana themselves, and if you're missing land drops because you've cut too many lands, you might not be able to cast the rocks either. This obviously scales inversely with the cost of said rock (even with 28 lands you can still cast Sol Ring on turn 1 about 100% of the time) but the effect is exponential; in a deck with fewer than 43 lands, a Sol Ring is worth about as much as a colorless land, an on-color signet is far worse than a dual, and high cost rocks like Relic of Legends are nowhere near worth a land drop.
Lily mentioned other issues like rocks costing your mana for the turn. I agree with her assessment that rocks and ramp spells are for just that, ramping to expensive commanders or combos or Storm turns or whatever ahead of curve, not just because they have a tempo cost for setup but because you also need lands to even pay that cost.
Yeah even yuriko runs, checks list 22 oh (it's not too bad it's 33 with mdfc and fast mana)
I run 34 lands and like 18 rocks, getting to the 3/4ish mana I need is no biggie. But my deck is just an Esper tutor shell for thoracle. Side note: In "casual" commander is it as important to hit a perfect mana curve? Like really, if you miss a few drops who cares, its not like you're gonna get killed or anything anytime soon. If you play a bit sub optimally people will pity you. And its SUPER curve dependent. Commander isn't as linear as other formats. You have high odds of getting wiped, or setback anyways, meaning playing on curve will lead to be just as slow as if you missed one or two drops. If your game is going to turn 8, 10, or even like 20 turns, missing a couple lands won't change anything. I kinda thought the point of "casual" commander was to play weird decks that take a while. From my small understanding on kitchen commander, doesn't the person who pops off first usually get beat down? And what are you trying to curve into? Do you HAVE to have a 5th land on turn 5? Can you not spend a turn dropping a sol ring/arcane signet, or casting a Kadoma's Reach? I may just not understand enough about casual gameplay. I can only think of a few cards I'd even use that are above 3 mana (FOW and the other free spells don't count as their actual cost in this case). What are you rushing to? An approach? A bola's citadel? Craterhoof? Tooth and Nail? I'd assume you'd want a boardstate before casting those so you're already not playing them the turn you get the mana too. Usually when one of my friends explains their kitchen decks to me its like 30 synergistic cards and you can win with like 20 of them on the field and you'll get infinite turns or dozens of extra combats or 100m tokens and skullclamp them to draw your deck until you hit this pet card of yours that you can then cast and it makes the heavens sing and the winning lottery ticket falls onto your lap. You're already playing off curve, you're already playing janky sub optimal decks. you're trying to see who can assemble a Rube Goldburg machine without getting interrupted first. Your Atraxa superfriends will probably function if it misses a drop or two and your sage of hours will probably get to its 5 counters in your ezuri deck. You already chose to walk the path of the inefficient, just live it up with your 31 lands and free muligans and whatever.
actually it's 36, the amount I arbitrarily decided to include the first time I built a deck and have not researched or moved from since (joke)
this answer is a lot more common than you'd expect lol
@ if there's one thing magic players are more allergic to than reading it's math
I really like this video and think it provides a lot of useful information about alternative ways to look at mana bases. I think one of the most interesting things magic has done in a long time is make lands way more dynamic and useful in various ways. I actually enjoy the push and pull of non-basic/basic/utility lands when building a mana base for my decks and often try to use my lands to help shore up weaknesses in my mono color or two color decks. I feel like a always start with 38 basics and then slowly start to remove basics as I add mdfc’s, utility lands and color fixing lands until I get down to like five to ten basics depending on how many colors I’m in.
My typical decks run with 36 to 38 lands. The video starts from a premise that the average curve of the deck or at least average card is above 3 cmv ... my decks tend to have an average which is below that. Couple this with an amount of cheap draw spells and 43 would just be too high, as you would flood your hand.
The thing with 'the best amount of lands in a deck' is, that there is no one single answer as it depends on the deck. I have landfall decks which do run 44 or so lands, I also have my Jhoira artifact storm deck, which runs 28 lands only (although that deck is illegal since the banning for jeweled lotus and mana crypt... but I refuse to take it apart... I only play it on very rare occasions with close friends anyway).
So no one is ever going to convince me 43 is the 'golden' number... plus we all know the answer to the most important question ever is 42 and not 43 :P
To be fair I play 36 and I dont have problsm it is a very good number
This video: here's why you shouldn't run 36-37 lands
The comments: here's why I'm gonna keep doing it anyway
Well, there goes trying to have a “nonland mana cards to land count” ratio to have 50 cards that care about mana bases while having 49 (or 48 for partner commanders) open for your playables.
@@hanahomemadepizza1424 I shouldn't run 36-37 lands because landfall is my only deck now
@@falway5109 Joke's on you, because my Hazezon Deserts deck only runs 38.
35 be4 mdfcs if your doing 37 you need 10+ ramp sorces.
@@falway5109 my Flubs landfall deck only runs 25, and I'm thinking of cutting some lol
It should be 0.425x99, the commander doesn't stay in the deck. Which is 42, the hitchhikers guide taught us this long ago. 😁
2:36
@@discounthero9972 You are absolutely correct and I agree with this assessment
Comment is at exactly 42 likes. I had to remove my like when I saw this. Cheers!
“Forty-two,” said Deep Thought, with infinite majesty and calm.
@@AmbiguousGoop others went way past. I guess you can like it again
I want you to know that after I watched this video I sighed like a 16 y/o who was just told to clean their room and took out a 6 drop for a 37th land from a deck 😂
This is the way
Clearly the answer is 96.
With one copy of treasure hunt, one Ad nauseam, and one thassa’s oracle. Gyruda doom of depths as the commander to throw people off.
what if Treasure hunt + Ayula's Influence tho? (discard a land, make a 2/2 bear) if you also have a reliquary tower in there, you can just make bears on your opponent's turn when everyone's tapped out, and say "bear with me" as you spawn 90 2/2 bears before your untap step, then proceed to say "hor hor hor hor" as your opponents grumble and scoop.
95, throw in that one artifact that lets you exile your hand while mulliganing.
You'll never stop me from having 33 lands in my izzet storm deck
The method from Karsten's Feb 2024 article that Rebell references is time-consuming, but more on point. Especially if your deck plays a ton of cantrips, eggs, mana rocks, rituals, and MDFCs, which Karsten weights differently from either normal lands or just any old spell. My Grixis storm deck is at 33 also
As someone with a 28 land izzet storm deck, I think you run too many lands. /j
Casual Mardu aristocrats with 32 and an MDFC. Average CMC of 2.19.
Even 36 lands is a remarkably painful proposition.
My modern Storm Deck had 18 Lands in 60. I was still debating on dropping another Land since any landdrop beyond the 3rd that you Hit can screw you.
I'm glad I'm not the only one. But to be fair, we have cantrips that filter out cards and ensure we hit land drops.
I've been playing since 2001, and we've always known the "40% rule."
It was actually hard to get into the habit of playing fewer lands when I first started playing EDH around 2010, but 37 seems to be the sweet spot for most decks, with 38 being the most I have ever run.
With the unique dynamic of a commander table, I've found in almost 15 years of playing this format that you don't realistically get screwed or flooded more often than you expect when running fewer lands than what is mathematically optimal.
You DO, however, get to play with a few extra really cool cards, and that is priceless.
The problem with the 43 lands amount is that you also don't want 50% of your deck to be mana sources. Unless you have a special case, an EDH deck should typically have 55+% of the deck dedicated to strategy and the rest dedicated to mana base. If I use 43 lands, I wouldn't be able to put more than 2 mana rocks in the deck. The calculators used by those pro players works because they aren't casting signets or additional mana sources. They are going right into their strategy, which puts more dependancy on lands for mana. So for them, it is absolutely important to hit every land drop. This isn't the case in EDH.
That too. I find it strange to just copy-paste the numbers from 1v1.
@@Arctanis-vt3hl where are you getting the 45-55% ratio from? Essentially normal Commander decks are midrange ramp decks in every other formats. You need all the mana you can get.
I think increasing the "cards seen" part of the equation is underrated though?
Drawing 1 extra card by turn 4 means you only need 40 lands instead of 43, and drawing 2 extra cards by turn 4 would mean 37 is enough.
For example in a Sythis harvest hand deck you can easily draw na extra 2-3 cards by turn 4 just by casting your spells.
Absolutely
@@ussgordoncaptain Learning this made me see scrying in a whole different way.
Yeah, early game card draw is ever so important, and consistent card draw even more so.
@@Val-rak this is exactly why fetching a surveil land is rising in popularity.
Sure, it's underrated for Sithis, but how many decks draw even remotely close to the same number of cards, especially that early? One card draw spell does not equate to an extra draw - you need to draw (and cast) that spell. So that means you need enough of a density of draw spells to reliably be able to cast them.
According to the Frank Karsten article, it works out that you need ~4 draw spells costing 2 mana or less to replace a single land. So for Sithis, you can definitely cut a ton of lands because every 1-2 mana enchantment counts as a draw spell. But for your average deck, running 12 draw spells (which must cost less than 3 mana) only lets you cut 3 lands.
Hard land count recommendation make no sense. Land count should take into consideration the mana curve of your deck. The mana value of your commander and ofc the deck color/archetype.
99% of all decks are 5-6 Drop Legendary dragons or dinosaurs and the rest of the deck are 45 5-7 drop timmy creatures. Those decks want to run 42 lands and are what almost everyone is playing.
true if you want to see the recommendation per your mana curve check time 3:26 you can compare you value to your recommended number of lands
Something to consider is that since you always have your commander (and potentially a companion) available to cast, flooding out is not nearly as bad as getting mana screwed. I've been perfectly happy keeping many 4/5/6 land hands but have only 2 decks where I would even consider keeping a 1 or 2 land opener, both of which can win without ever getting above 4 mana.
That doesn't make cutting gas for lands in my decks any less painful though.
It’s absolutely crazy how so many comments here and on reddit are hellbent on playing too few lands. It’s been like that for years.
You mention that ramp and card draw dont count toward the land base, but draw has a very different purpose.
Forty three lands is calculated using a deck that only sees one card every turn, but the more low cost selection you add, the higher the sample size you are starting with.
Faithless looting alone makes your opening hand + first draw a sample size of 10 at the cost of one card on the long term.
Ponder adds up to 4 to your sample size, preordain adds 3.
The math is alot more complicated than ponder = land but it will very quickly and easily increase your consistency while also being more useful in the late game than a landcycler that does very little in a real game of commander.
And it's not just blue cantrips and flooting, red has thrill of possibility +2 and all its clones, black has sign in blood +2 and nights whisper +2, white I can only think of land tax and green just has better ramp which can get you to harmonize +3 faster.
Plus the surveil and scry dual lands. I prefer to be able to draw, scry, surveil etc into actual gas but guaranteeing land drops is also a solid backup function xD
Someone factored that in, I think it was 3 draw cards per land cut. Low cost commanders that draw can cut a few though.
43 lands is assuming 1 draw per turn. If your commander guarantees you to draw an extra card every turn, such as on attack or end step then you'll effectively hit those lands earlier/easier and can probably afford to run less.
Something people also forget is that ramp spells pulling lands out from the library reduce your remaining land count, making it harder to hit lands on the draw, meaning you should run extra lands.
The opposite is also true. Cards that remove lands are good because they ensure you aren't completely land flooded when you need other cards.
Turn 3 Nature's Lore and missing a land drop is peak comedy. It happens every single time.
you're describing deck thinning which is negligible at best outside of 40 card decks.
@@TheTexasDice You shouldn't have kept a 2-lander though. 💀
A 3-land hand and nature's lore or other two-mana-ramp means you get to untap with four lands turn 3 and if your deck folds 90% of the time that happens, you gotta go back to the drawing board.
What's more important than the mythical deck-thinning is that in EDH everybody, including the starting player gets to draw 8 cards 100% of the time, instead of just 50% of the time (second player in 1v1) AND everyone gets a free mulligan. And in most casual pods, people just get to mulligan until they have a playable hand.
This kind of disadvantages "honest" players who build their decks correctly. I tend to run 39+ lands in most of my decks (including a couple of mdfc, but I don't like to count them as a complete land because if I'm playing them as a land 90% of the time, I might as well put an actual land in the deck. -> mdfc should always be added on top of your minimum 37 lands in my mind) and I often have a playable hand in my initial draw or after the free mulligan.
Meanwhile some players in my pod rocking the bare minimum of 37 lands or even a bit lower keep frustatedly mulliganing until they finally have a 3-lander or they keep a risky 2-lander after a couple of mulligans that might end up costing them the game. But if everybody gets free mulligans until they have a hand with 3+ lands, some players will just have a higher spell density because their greedy deck-building style doesn't get punished by the pod's leniency.
I've always used 40 lands and 10 ramp/mana rocks, for 9 years I've been trying to convince people that it's optimal and maybe one of the reasons I win so often
I think people undererstimate the fact, that your commander is your 9th card that is almost always available to cast. If your chosen commander is half-way decent, then having it in play will win you the game eventually (draws you cards, does the thing etc.) and if it's not in play (it got killed) then you likely want to be able to recast it despite the ever increasing commander-tax. Besides many lands letting you cast your spells more reliably (including "double spelling") and all that, being able to reliably recast your commander for a third time just makes even that 8th land a good "topdeck".
I think more people lose to mana-screw than to mana-flood simply because it's way harder to be flooded, when you can always play your commander. So many games are lost, where a player has a full grip of cards in hand but only like 5 lands in play on turn 6.
Got a very similar concept. I'm usually using lands + ramp = 50-52
If I have a low curve and a deck that starts doing its thing early than ramp/ritual is more important than lands and if I have big drops (6+) I'm relying on hitting the land drops consistently and land count becomes more important. There's also the consideration of cards seen and draws. A deck that runs 10-15 low cmc card draws is more likely to see enough lands even with a lower land count.
Thank you so much for making this video. After watching Sam’s video I started playing more land in my decks but couldn’t seem to convince anyone that making a third land drop was better than 2 lands and a rock turn 3
99 lands, Commander is Garth One-Eye. And it's a fine selection of lands as well, such as Den of Bugbear, Fountainport, Khalni Garden, Kher Keep, Kjeldoran Outpost, Mirrex, Springjack Pasture, Urza's Factory, Vitu-Ghazi City tree, Barad-Dûr, and many other lands, such as manlands.
I like the hustle.
One thing i have found with lgs games is that many people follow a ‘gentlemen’s mulligan’ where you can mull basically any number of times to get a ‘playable’ hand. With the length of commander games and the multiplayer nature of the format I get the inclination, but it definitely reinforces the 37 land rule.
I seem to recall a video from The Professor where one pf the guys from the commander rules board said something along the lines of "the 'unofficial' commander mulligan rule is to allow enough mulligans until you have at least 3 lands in hand, but once you have that you must keep the hand", but also followed it up with "well, that can't be the official rule, cause then everyone will just cut down to 28 lands or less and then play super powerful shit they could only expect to run with a consistent starting hand"... which i guess is just what cEDH does anyways? Lol. The point is, if you are gonna allow that, then you really gotta trust your pod won't be willing to abuse it - or if they are, be ready to play a cEDH level game, lol.
@ between four players and up to two starting hands each, it greatly increases the number of games with sol ring in the opener for at least one player.
@@runcmd1419 our gentlemans mulligan, requires that you cycle Sol Ring if its in the hand.
@@altejohlmao no, cEDH players mulligan down to four sometimes, but it often results in a loss. cEDH decks running almost no lands is sort of a result of the dockside meta where the goal was to cheat out or cast or clone dockside to accelerate your mana. Fast mana like the moxen also let you cut down on lands because you can get multiple mana sources out on turn 1 for a t1 rhystic study. Most lists even then still had 30 lands at least. It was only really Roghrak/Silas turbo builds that got some lists down to 24 lands. But it was normally around 27-29. Now, cEDH lists are looking around 32-34 lands a lot because the format is slower. The average mana value is also super duper low, reducing the need for lands. As long as I have several cards that can look for Esper Sentinel/Rhystic Study, I can start drawing 3-4 cards a turn.
I think that’s a terrible idea, and it’s the same as deciding to break any other rule in the game for the sake of it being more “fun”. If you are having to consistently mulligan your hand, there is something wrong with your deck and you should fix it. If you go by the gentlemen’s mulligan, you never have to address that there is something wrong with your deck because you can just keep fixing your hand.
I have been very happy at the 37-38 (including MDFCs) Land count... rarely miss rarely flood
Same. This has felt comfortable in most of my decks. -- And also, card draw for days.
Im curious why you didn't bring up bounce lands since they can allow a 3 land hand to generate 4 mana on turn 4 even if you didn't draw a land in those four turns. Thanks to Thunder Junction a two color deck can now run three copies of this effect (the perfect amount imo). They also have added utility with both MDFCs and cycling lands later in the game as they allow you to cast the spell side or cycle the land.
I didn’t want to complicate a pretty complex video already but that will be something I bring up in the next video for color fixing and duals, which the bounce lands fall into.
I love mdfc but I would only ever run the on color bounce lands. I care about fixing a lot so maybe you care about that less. The two colorless pips and general clunkiness of guildless commons and the otj equivalent make me really not want to play them.
Not sure if u did not mention them because they are bad or if you forgot about them, but you could run up to 5 bounce lands in a 2 color deck if you include the monocolored bounce lands. In my simic landfall deck I actually play all 5
@@devan9197 Yeah if you are playing three+ colors you dont want to run the colorless bounce lands since you can run minimum three of the guild ones, but most two color decks can afford to run them. If you're playing a two color deck and you have consistent fixing issues, the problem isn't the two colorless bounce lands.
@@shideon In two colors decks I have consistent mana fixing because I don't run those lands. Even if you can handle the fixing downgrade you also have realize that you're playing a tapland with tempo loss and you can't play it on turn one. Colorless taplands have to be really good to be playable and I don't think their worth it in two color decks. Mono colored sure you can throw em in there if u want.
Thank you for the help tweaking my Disa deck. They joke at me for having 42 lands. Came in 1st on game one.
Lfg!!! Disa is sick
Was a weird feeling for me. I clicked, remembering an argument two friends of mine had after one scooped for being land-starved. The other insisted he needed 36 lands. he looks at me for validation and asked "how many lands do you run?" I said "I think it's 46".
So, for curiosity, I recounted them and it came to 43. Then I remembered it had been 46 a while back but I found it slightly land-heavy and I shed 3 of the least-preferable combination color lands to hit 43, and it had been at 43 for about 2 years.
I fully expected this video to quote the classic 36 lands, and was hoping to learn something about why 36 is considered the standard, and then the author dropped the "43"
Oh.... ok then.
The advice to use cyclers seems kinda weird. In EDH, I'm never going to cast Troll of Khazad-dum (or any of the others not named Lorien Revealed). You correctly point out that in EDH we want to build engines, and these cyclers don't really synergize with much of anything, or tap into the engine. So realistically, they read "pay 1 for a land". I always want my non-lands to either accelerate, answer threats, or advance my game plan. My gameplan is nearly never, "play over costed creatures with 1 keyword." Kyle Hill also made a joke during a CAH or EDHijinks that people should cut their cycling lands, because nobody every cycles them, and they are just tap lands. I agree with him on that.
None of this controverts your claim that 43 is the magic number -- I'm just arguing against your proposed "solution" to cheat the number.
The idea is not that you cast Troll for 6 on curve as a normal creature. The idea is that you're in the late game, flooded out, head-to-head with another topdecking opponent, and you draw an evasive creature instead of a swamp. It's a reanimate target, a spell, a blocker, an attacker, it's a land you grab with Eladamri's Call, etc etc etc. It's just modal - nobody is saying the creature is "good" it's just that, in a lot of situations, a Swamp is worse.
@cliiwen ok, it's turn 4. You have a land drop and you draw the troll. How do you feel about your troll?
@@colinfowler3021 Just as dead as any 90% of other land draws or six+ drops? Granted a six drop will likely be more useful later, assuming you aren't losing due to tempo loss already.
@@colinfowler3021paying 1 to fetch a fifth land is still cheaper then Farseek or Three Visits, which get tons of play. Sure you don't get the land in play, but in your case, you weren't making the land drop anyway, so there's no downside
@@colinfowler3021 you keep it in your hand for later, or you cycle it on opponent's end step so that you don't miss Land 5. By your logic is Valgavoth a bad card? What if you draw it on turn 4, or 5, or 6? You can't assess cards based on being able to extract maximum value out of them, exactly on-curve.
Saw you on Tolarian Academy and then you popped up on recommended. So glad to find you! Great video!
For arena players, you only need about 10 lands in your deck, because you'll always draw a land when the shuffler wills it so.
This was incredibly informative. Super well researched video Rebell, can’t wait for the future land ones!
I usually run something between 32 and 38, but I think that's mainly due to the fact, that I usually run low curve decks and/or decks that are more likely to lose from mana flood than mana screw.
I'm probably one of those odd ball players that likes to run 40 or more lands. Lol
My current main deck runs 41 Lands. Lol
Why is the turn 4 sample size 12? 7 + 4*1 = 11
Karsten used 12 as well so I must be missing something painfully obvious.
That's a good question
I would assume that it is because you don't draw if you go first in 1v1 formats, such as limited. this also happens 50% of the time so needs to consider the worse/smallest sample size.
@@Reldane then it should be 10 instead of 11. But they did 12
the extra card might represent the ability to mulligan a low land hand in commander? having an extra seven means its not a downside, so it might lean heavier into the calculations than in a 60 card format
@@michaelpsmith3861 then it should be a lot higher than one more card seen
35 Lands - i feel like i'm constantly missing Landdrops, 37 Lands - i feel like i'm constantly flooded. 36 just feels right. Usually running 10-12 ramp spells. Of course there are exceptions (landfall, cedh, etc.)
Landfall when 40 lands:where the damn land I need to trigger my aesi (it still happens and chaos begins)
Curve also matters a lot. When my deck curves out at 3 with maybe a couple 5-drops, I need fewer lands than a big stompy deck lol.
With one single land between those scenarios, there's no way you aren't getting confirmation biased my dude
My flip markov lifegain Deck has an average cmc of 1.89 ... i run 37 Lands and still find myself Not hitting Lands enough lol.
@@falway5109 I'm running 40 lands as well in mine. Fetch effects + "land from the graveyard" effects _really_ get the snake going. Add in some "additional land per turn" effects and your hand is thicker than your library in no time. I also replaced almost all the draw effects with ramp spells, because Aesi.
I love these videos. I remember figuring out 43 was ideal with static paramators other than land count using the hypergeometric distribution calculator but it blew my mind realizing two of the other parameters could be altered by reducing my curve or increasing the number of cards i can see.
I tend to run 42 sources of mana now a days with around 32 being lands 8 being rocks and 2 being some more arcane mana advantage effects.
I almost always run a low curve with advantage engines and playing commander gives that free mull.
32 lands?? O.o
That should mean that you regularly miss your land drops unless you have unlimited free mulligans, which is of course not what is generally assumed.
@lVideoWatcherl I'm not sculpting a hand with 4 lands and a rock.
I want a hand with 2 lands, a rock and either an advantage engine, card selection or an additional land. The ability to see more cards allows me to draw into lands or an engine or to manipulate my graveyard for future value. The ability to consistently hit 3 mana sources is required to play the majority of my cards.
I also Mulligan towards my gameplan so going to 5 can be acceptable as it allowed me to see 3 full grips of 7 cards. Seeing 21 cards although not guaranteed to be unique is still going to be around 1/5 of your deck seen before you play your first card.
An example on how to think about this. When calculating for land drops consider the odds of hitting a rock, atleast 2 lands and an engine or selection card. If I play a looting effect turn 1, I look at 10 cards total with 6 cards in hand and 3 in the yard. If you fetch a surveil land that's 9 cards seen with 7 cards in hand and upto two in the graveyard.
The ability to consistently see just a single extra card finds that next mana source at a rate more valuable than adding many additional mana sources. Also I can only play one land a turn so I tend to favor additional draw to find lands as it also finds engines, combos and interaction at the cost of actually expending your resources. If I ran landfall I'd favor more lands as they become spell effects or id favor more lands if I played cultivate effects.
@@Val-rak sorry but this just isn't good. 60 card, competitive formats with drastically more consistent decks have .33 lands as the floor. unless your decks are very low to the ground, there's no way this works out better than 36-40.
Every time Rebell makes a stats-like video, it reminds me that I'm not nearly as methodical in application of ideas.
I'm just like," Well, I tried that, and it seemed to work okay." 😛
My only gripe with running extra utility lands is that your decks become less unique or explicitly synergistic. Like yeah, they'll be better, but less FLAVORED.
depends on the utility land!
HA! You want to talk “flavored?”
Try running 40 Lands while sticking through a theme that isn’t chock full of “staples.”
(Now, granted, if you forgo Foils and card editions, my deck will cost WAY LESS than the $1000 price tag it currently works with…) but, keeping to such a land count, making certain that it stays on theme, AND it doesn’t go overboard with Staples is much more challenging - when “Skullclamp” isn’t in your 99 of a Marath, Will of the Wild deck.
(However, getting the land base to work is still tricky, seeing that John Avon - Kamigawa Lands - *Retro (Foil)* - AREN’T A THING YET, WHAT IS WRONG WITH YOU WOTC-)
(Ahem.)
But creating a “World Map” theme is a super fun, curious journey to take.
37 Main.
3 Forest, 3 Plains, 2 Mountains, 5 Fetches, and quite a few duels and “utility(?)” lands?
It’s actually kind of fun, when you think of your commander deck, and painstakingly build your land base with the theme.😅
Either way, (still waiting for that John Avon, Kamigawa Retro Foil Mount-) AHEM - I would love to know what you run in your deck.
(I run “Wild Pair…” It’s a crime how the card is $2, yet runs the gamut of half your deck by just checking for creature stats.☝️😦
“I cast Primeval Titan (from my hand), it’s a 6/6, get Ojer-Tac - it’s a 6/6, get two lands, trigger Landfall - Twice, swing with Haste, get another two land(s), get landfall trigger - TWICE, again, cast (X Spell), then pass?” 🙂)
Cards are for playing, not eating.
This is the more interesting question for me...at what point do the diminishing returns of hitting a land drop every turn really hit?
Somewhere after your commander’s mana value and average cost of spells in your deck. We hedge against that by placing more lands that do things so in the event you flood, you have sinks to take some action:
If you build your deck right, never!
Mana sinks (mostly on lands or on permanents as a secondary effect) are very good in Commander. Cards like Warroom are a prime example for this.
Rebell: I'm bad at math.
Also Rebell: Starts doing multiplications.
Me: and here I thought this would be a math-less video.
"36 is the border we should never cross"
*Me over here running every single deck I own at 35 lands and doing just fine.*
You are not doing just fine. You are purposely ignoring the games you missed land drop 4 twice and lost.
I'll bet more then anything that your play group allows more free mulligans.
Yeah, you're doing fine because of endless free mulligans. Completely cool if your playgroup is fine with it. You'll just run into trouble (non-games) more often, when you start playing with strangers in a store and they actually enforce the "up to one free mulligan"-rule.
@@artemi7 I don't have a play group I play at an LGS. I do the standard 1 free then its London Mulligan after that. We rarely have to use it, but I tell people they can mulligan and keep 5 if they need to.
Maybe if you understand mana curves and actually balancing out your mana bases along with ample card draw and ramp, your deck might run more consistently regardless of how much land you run.
@@TheTexasDice The last time I missed a land drop was about 2 weeks ago in Grease Fang. I played a careening Minecart which was ramp in that deck's main strateg, I was able to get a few treasures to smooth out the game. Then hit more land drops and I ended up winning the game.
I just don’t know that Frank’s article really applies to commander in the same way that it applies to other formats. Your average commander deck has 10 to 15 pieces of ramp in it, which is something that is not true in other formats, commander decks also tend to have significantly more card advantage effects, since “dump your hand and beat face” isn’t really viable like it is in limited/60 card. Heck as soon as you get into older formats like legacy, which is a 60 card format these calculations go out the window. That being said I think 43 is it OK number I just think it’s a bit on the high side, somewhere between 37 and 40 is probably more “correct” in a format where every deck has Sol Ring plus 5-10 pieces of cheap ramp that doesn’t have any color requirements, and a decent amount of card draw effects, a lot of the time right in the command zone.
I'll compromise. I typically run 40-41. You say 43. Author Douglas Adams said the secret to the universe is "42," so that's the new number I'll go with.
You should watch the video lol
@RebellLily in process - sorry!
@@RebellLily aha. Complete. Like the idea of using MDFCs & cyclers and do that often. Outside of that, I tend to use the Surveil lands as a way to "dig" ahead for a land drop if my hand is light.
Commie Commander here,
Finally, someone out here praising the real gospel! I rarely drop below 40 anymore and I don't feel great if I'm not being given a really good reason to not LAND (hahhah ugh) between 42-45.
Yo! Ted showed up in the video at 6:27! So that's where Ted went when I was looking for that slice of pizza.
Damn, guess what I'm doing tomorrow? Going through like 12 commander decks and counting lands lol
10 tapland is insane, no more then 3.
That really surprised me, and makes me think that this advice must be aiming at much lower power decks than I imagined.
At casual tables, budget is a huge parameter for tapland percentage, especially because a lot of people want to play three or more colors regardless of budget. So once you grab the budget untapped lands, even I don't tend to find it enticing to spend 5+ dollars on a single untapped land in my 100ish dollar deck
If your pod isnt into proxying original duals and fetches for every Deck youll See those numbers.
Most of my Group has 20+ commander Decks. Doing Goodyear unproxied landbases for those would cost us thousands.
Depends on the budget and colors, I'd say.
I'm building a 75 bucks Ramos deck, so having 10 NON-taplands is more realistic, lmao.
Meanwhile, a 75 bucks two color deck can probably run 0 taplands easily.
Three colors in that bugdet is probably doable below 10, four colors... maaaybe? Haven't tried it yet.
It really isn't when you go for 40+ lands.
Like, take your 3-color deck. You run your Triome, 3 Surveil Lands, 3-4 Bouncelands and probably one or two MDFCs that enter tapped (like Ondu Inversion or Bala Ged Recovery).
That's 10 taplands right there.
Idk if it's just my algorithm, or it's land base winter but to copy a comment from a different EDH land video from a week ago which is heavily relevant to 3:36 in the video;
The #1 commander on EDHRec right now is (still) atraxa, and it's average deck has 34 lands. The top 10 commanders are Atraxa 34, Ur Dragon (9 cmc) 35, Yuriko 29, Krenko 32, Lathril 32, Sauron 34, Kenrith 33, Edgar 34, Miirym 33, Kaalia 35.
Not a single deck in the top 10 is playing 36 or more, and you have to go down to 17th to get to Velociramptor face commander Pantlaza with exactly 36 (the precon comes with 39).
edit: Oh no, just heard 7:05
While I agree with your point, that 40 lands is probably more than most decks need, I'm not sure that EDHRec average decks are a good thing to point at as evidence. I suspect the average deck calculation *may* undersample how many lands are in decks.
@Dynme My belief is actually the opposite, sorry. The edhrec average is WOEFULLY too low.
I didn't elaborate that, so I fully understand how it could look like that.
@@Temzilla2 Ah gotcha. I personally think most decks will function fine at 36 to 38, with 40 being "reasonable but maybe overkill" for non-landfall decks. But obviously there's a lot of variables that should factor into it.
@@Dynme Yeah I think that most of the time you structure your land count based on two major factors, when your draw engines come online, and what turn it's okay to start missing land drops with your goals for the deck. I.e. if you have the most powerful draw effects you can run in your deck, and it's okay to start missing land drops on turn 3, you can run 26 lands like some cEDH decks do, but if you aren't running a ton of low cmc powerful draw engines, and you need to be hitting land drops on turn 6+ you're gonna be looking at 40 or more lands.
Ah, just want I needed. Some cold hard maths in my Magic *breathes deeply* THAT'S THE GOOD STUFF
Your content is great. I think I saw you while waiting in line at Mei Lai Wah on New Year’s Eve, but you were walking quickly so I didn’t want to bother you. Just wanted to say keep up the great work and thanks for keeping us all entertained.
*Me, playing 17 lands in my mono green commander deck, two of which don't tap for mana*
"I'm subscribing, liking, I wanna hear Rebell talk about lands all year"
What, how???
@ I decided it was okay to pay for land drops
@@JACKSONPRYORBENNETT got a decklist?
For what its worth, the lists I've tested with the most consistent land drops are 31 land turbo delirium money pile and 26 land dredge lands but those are some very specific deck churning archetypes that just care about four lands existing somewhere along the first 50 cards of the deck - sometimes not even that (untap source + fetch + w6 moment). And also just the mana bases are worth 300 bucks on each list which is why I don't actually own them in paper. But even as someone who likes having as few lands as possible in a deck, the number (43) tracks in a vacuum.
We can apply this to the previous "Cube Theory" and drop down to 7 x 8. That gives us 56 cards, 43 lands (or ways to cheat lands) and 1 commander. So either drop one face of the cube or reduce the number of cards on any face by 1.
*sideeyes my Purphoros and Xenagos list looking to one-shot people by Turn 5*
No, no, dears, she’s not talking about you.
Loved Sam Black's video on mana rocks.
Every tapland you consider playing should make you compare the land hit probability for one more turn with or without that land to make sure whether or not you should play it (as that land is coming into play this turn but will only be usable the next).
Great video Rebell!
Honestly my gut feeling is if I took out the free mulligan rule me and my friends do for commander it would probably be a lot better to run more lands in my decks. Might bump up the number from 36 at the least.
Mulling to find ramp or important spells is soooooo much better than mulling to get enough lands in your opening hand though. Once you realise this, generous mull policy isn't even enough reason to play too few tbh.
Fantastic video! I'm going to update my Nalia de'Arnise deck! Currently sitting at 36 lands + 3 MDFCs, going to try and get closer to 43
"Hitting a land drop" is about efficiency of having playable mana per turn. As a base line we say one basic land a turn that gives you one more mana available each turn. Cultivate costs you three mana to cast plus a card draw (as if you top decked it). It gives (ramps) you an extra land on the battlefield tapped and one in the hand ; effectively a card draw as you still have the same number of cards in your hand at the end. Sword of the Animist costs two to cast and two to equip and on attack puts a tapped basic into play. So the first attack will let you ramp an extra land but used up four mana to get there. If it survives then the next attack nets an additional land with no extra mana cost. The point is using stuff like this to make land drops is not efficient as just drawing a land. So, the above examples are a poor way to hit a drop but a good way to ramp if you already are hitting your land drops.
Card Draw is the biggest factor in land count. At 40 lands in 100 (I know 99, but easy math) you have 2 lands every 5 draws. You start with 3 in hand. At the end of 5 turns you should have held 5 lands given a perfectly spaced deck. This is where SD (standard deviation) would give you a more accurate percentage. Come turn 6's card draw there would be a 40% chance of getting a land and hitting your land drop. For simplicity lets say someone played a Howling Mine and on turn 6 you draw two cards you have an 80% (simplified math) chance of hitting that land per turn.
you're completely forgetting the value of scrying. being able to scry / draw a bunch of times consistently probably lets you get away with running less lands.
I am a HUGE proponent of playing closer to 42 lands. The thing most people dont talk about when it comes to playing more lands and complaining about mana flood is that in commander you always start the game with one guaranteed spell in hand (your commander). This means that I am much more willing to be slightly flooded. Especially in mid-casual games where the game usually goes a bit longer into the 8-12 turn range.
And especially in a multiplayer game where often times its difficult to just card advantage out the table to win; you are usually winning by double spelling multiple times on critical turns. I find its very rare for me to lose a game with 0-1 cards in hand. Usually at the end of the game, I have a couple cards left in hand and I wish I had more mana to deploy them.
You are also encouraged to run mana-sinks as secondary effects on your permanents (or lands) in order to make use of excess mana. If you flood out, you can always use Warroom to draw a card or Dawn of Hope to make a 1/1.
Frank Karstens updated article actually accounts for cheap (2cmc) ramp and draw and comes down to 40/41 lands for a deck with average mana value of 3 minus 1 land per each 3-4 cheap draw/ramp. According to his formula most of my decks should run 36-37 lands but your individual results may vary.
Karstens TL:DR:
"A good formula for the number of lands in your 99-card deck, counting MDFCs partially in this fashion, is: 31.42 + 3.13 * average mana value of your spells - 0.28 * number of cheap card draw or mana ramp spells. This means that if your average mana value is three, which is fairly typical, then you should start with 40 or 41 lands and cut one land for every three or four cheap card draw or mana ramp spells in your deck"
Finally, someone did the leg work so I don't have to
Compulsion and Blood Sun are great budget cards for those color identities. In a mono, blue deck or a two tone blue deck compulsion essentially gives every card in your hand cycling, which, if it was phrased that way, I think that a lot more people would be playing it.
Blood sun was seen as a throwaway rare in rivals of Ixalan but the fact that it gets you a card when it hits the battlefield as well as making it so that all of your lands, enter untapped can really level the playing field, not to mention the absolute absurdity that happens with bounce lands.
As for the ratio, I did my own math a few years ago and I did come to a different number, however, I think that’s also due to the fact that I was looking at Man of value itself per land as supposed to lands per card. I will definitely however, point out that I’m rather certain that both the pros and myself we’re looking at 60 card formats. In my personal testing, I found following my formula meant that I needed way too many lands to play the cards that I wanted to play, which is probably why, instead of increasing the land count that vastly, I just lowered my average mana value,
Average mv matters a lot more in deck building than I think most people realize. You can make a 60 card tech with 17 lands. If your average mana value is one.
I do agree with you, however, that the pre-cons have their numbers off slightly, either that or they’re set for a slower gameplay. I think the grid that you had at the end of your is really something that new and advancing players should be using as a template, because Lord knows that even with years under my belt apply 60 card formats, making a Tricolor manabase was why I bought my first commander pre-con, I had no idea what I was doing.
Something I never see discussed is whether hitting every land drop every game is actually necessary? In the 4-player social setting, missing a drop every now & then doesn't seem worse than over-committing to the board early. I feel like sometimes we're too precious about land drops as a hold over from 60-card competitive. (Not talking about the player who's on 30 lands & misses 4 drops in a row.)
This 👆🏻
I play 60-card casual multiplayer. It still feels bad that I miss land drops early. You still want to get your stuff on the board. Having to discard because you don't have enough land and exceed your max hand size feels bad.
Hitting the first 4-6 or so even is huge and is the difference between double spelling most of your deck or not. I think if you consider it to be almost literally being able to play 2X as much Magic, you can see how important it actually is. Also would argue that leads to more fun gameplay even in a casual pod.
But also most players run land light so you may not be seeing that contrast when entire playgroupls are simply running land light and the majority of those pods are missing more land drops than would be ideal
The command zone did a video on this or it was part of their yearly stats episode. Hitting every land drop is the number one statistical indicator of winning a game.
Missing your sixth land drop? Probably fine. Missing your third? Death sentence. Having your game plan be "Rely on the pity of your playgroup so you can be greedier" you're going to find yourself more disappointed than not.
Interesting watch! As someone that naturally gravitates to mono color low curve decks with mdfcs and land cyclers intuitively, it's funny to hear I'm just good, baybee
I run most decks at 34, anymore than that and I start to get mana flooded. I never run more than 36 unless the deck cares about lands then I don't go over 42. For my cEDH decks I run between 24-26. I keep my casual decks at 33% lands so on average I draw a land every 2-4 cards. I am pretty sure all "pro" calculations only consider having 4 lands by turn 4. If you think about turns 7-13 you will most likely be flooded and hoping to draw something not a land if you run 38-43. This is just my experience and how I build all decks.
do you play with endless free mulligans in your casual pods?
34 lands is terrible advice for most beginner decks, unless your curve is super low and/or you have tons of 1-2 mana drawspells and scry/surveil effects.
34 is crazy low, your playgroup must be using some crazy greedy mulligans to compensate, for this deficit.
@@PaulSzkibik Just the normal one free.
I agree it is bad advice for beginners but it works for how I build and play.
@@gustavoramirezjr7320 I rarely go to 6 cards in hand and almost never go to 5. Most of the time I find a good hand in the first or second mulligans.
@@galencraig6422 I kinda want to try a low land count deck as well now. 😅
When you dropped the "43", I went from "maybe I close this window" to putting you up on the main monitor.
Super informative and helpful. Looking forward to the one mana rocks.
My personal rule after another video I watched is 37 physical land cards (physical meaning like actual, pure land cards). That video calculated 37 as the breakpoint number at which you have above a 50% chance to start with 3-5 lands factoring in one free mulligan. I do really like the 43 number as a target to futher pad my physical lands, though. I frequently add in MDFCs (and I love the Bounce lands), and Cyclers weren't something I'd considered. Sounds like 43 is a good target for all sources put together as a sort of "land-in-hand" count, separate from ramp and rocks and such.
I actually run 40 lands in all my commander decks at the start of the deck! It's great for testing and figuring out a direction for the deck before I fully commit to upgrading it. ^_^
Before watching: it should be enough that you always have something to discard to hand size and miss your land drop next turn
Very interesting to see math and thought behind how many lands you want. I've actually had to rebalance a few of my decks recently because I didn't have enough lands in them, haven't played them out in actual games but the play-testing seemed a little better. It does ultimately come down to individual wants but knowing the 'best' number of lands is good to know, gives a good base. Also none of this matter to competitive players because that is a completely different beast. Think of it like this if casual EDH is a game of pickup basketball in a neighborhood then competitive EDH is like Naruto, you have to worry a lot more about BS teleporting and breaking the laws of physics. Remember competitive games are all about playing less Magic because your goal is to win as fast as you can, not have fun..
I don’t know. 40% is always enough for most decks that are built properly. As long as your mana curve is not too high or low. 3.4 is the average curve of a good deck, thus 40% land you should in 77% of games have 5 land by turn 5. If you have more than this the likelihood of being flooded is very high.
I think there is a weird psycological effect where people are much more afraid of mana flood than they are of screw even though it should be the opposite. People already understand why tapped lands are bad as they slow you down and missing a land drop is just that but a hundred times worse. At least if you're flooding out you should have things to do if your deck is built reasonably well. Tl:dr- Play more lands!
Flood IS worse than screw. With screw, drawing one or two more lands means you're free to unleash a hand full of pure gas. With flood, drawing one or two spells leaves you with... A spell or two.
To be fair flood is significantly worse than screw. If I'm drawing nothing but lands once the gas in my hand is gone that's it. In the case of screw if I'm stuck on 2 or 3 lands as soon as I start getting land drops i have a ton of stuff I can do. And in a game that goes long missing a land drop won't kill you
me who plays roughly 25 lands watching this video:
"interesting"
I go with 34-36. it depends on the mana curve, the colours, and style of play. 38 would be either a mana starved deck or a deck thats all about playing as much land as fast as possible (Like azusa)
43 is taking out a lot of utility, and bordering on mana flooding.
I feel comparing limited to commander is a bit off, because limited is faster. EDH is all about that slower, bigger, flashier style of play, it also has bigger focuses on ramping
this is why i'm a fan of commanders who have abilities that let you cycle through your deck. I run a high mana base with a Raffine deck, raffine connives and once i have enough land, i start ditching the lands.
My Bant lands deck and my Dimir zombies deck both run 34 lands and 5 MDFCs. (Dimir collects rocks, and Bant Ramps).
You've made me want to raise this to 42 lands. If I flood out I'm coming back hear to flame you tho 😊
Here's the title of the Channel Fire ball article I based my land and ramp/rock/ritual count on.
What's an Optimal Mana Curve and Land/Ramp Count for Commander?
By Frank Karsten
The 36 comes from a much more aggressive 14 lands in a 40 card deck, leading to around 20 lands in constructed and 35 for commander. It's not out of the blue, but it IS based on hyper aggressive play that's less applicable to casual commander than, say cEDH where you're looking for a very low curve win con.
Love this video! I'm a 38-42 land boy and my playgroup thinks I'm weird. I'd rather slightly flood than have "non-games" where I don't hit land 3 or 4 until turn 6 and have to twirl my thumbs for half an hour until I can join playing with the rest of the pod.
My (intoxicated) equation is begin with 50 (fifty) lands. Every 3 (three) mana rocks count as a land. Every 2 (two) draw more than replace count as a land. Every 3 (three) dork counts as 2 (two) lands. Every 4 (four) land search counts as 3 (three) lands. Every dual land counts as 1.2 (one and one fifth) lands. Every triome counts as 1.4 (one and two fifths) lands. Any quadrome(?) or all color land counts as 1.5 (one and a half) lands. Every wincon counts as -1 (negative one) land.
I usually run about:
9 (nine) rocks. -3
~2 (around two) draw. -1
6 (six) dorks. -4
4 (four) land search. -3
10 (ten) dual. -12
5 (five) triome. -7
6 (six) any color. -9
9 (nine) combo pieces that never stick, so -9. -9
This brings my total land count to 2 (two) lands in just about every deck I build.
Check my math if you'd like. I have a couple beers in me right now. Taking suggestions, though!
Also keep in mind that you can get away with fewer lands if you are playing a deck filled with cheap spells or a commander with built in ramp or draw. This is a pretty good video for most commanders that don't have that and probably play a notable amount of 3 and 4 drops. MDFC's and land cycling are really good includes, solid advice.
Land Rule that applies to Casual and cEDH
# of Lands = 31.42 + ( 3.13 x Average MV of deck ) - 0.28 x (Number of Draw spells and ramp spells less than 3 mana)
I have been playing a (joke) Ghazghkull Thraka deck I call Plateau which has 60 spells and 40 lands. Each cmc from 0 to 5 has exactly 10 cards in it for a flat plateau of a mana curve. One this I have noticed is that the 40 land count means that my deck always has something to do and can even make big plays later in the game despite only have 5 cmc cards at the top end. In fact, because my deck tops out at 5 cmc. It feels even more important to make sure that I play my cards on tempo as often as possible. Delaying my game plan and slow rolling will only put me behind compared to other decks that are establishing game winning big mana engines.
A combination of a low curve, consistant land drops, and as many spells that refill one's hand or board keeps the deck relevant even into the later turns. I do cheat a little by counting Ascend from Avernus and other X spells as their default cmc for the purpose of deck building.
This deck wins about 1/4 of its games, which is about par for a good casual deck I think, but consistently gets 2nd or at least stays relevant at the table for the whole duration of the game.
I think the mana base ahould typically be 35-40 lands. However depending on your play strategy, the counts can go up ir down. For example landfall decks can go up to 53 at the highest.
Very interesting video! I generally play a p low mana curve for my decks so my land count is around 30-32 including MDFCs. Since my commander and cards are p low to the ground I p much never struggle with mana.
I always played 39 or 41 with esper decks, people judge me, but esper don't have good ramps or a smooth mana curve to allow me to lower my mana base. Since I started doing it, I begin to win more games.
I don't have the numbers to back this up, but I feel like the average land count of opponent decks should be a variable in this formula. First I would assume that by increasing land density I would be increasing consistency in reaching a healthy mid game board state while reducing consistency in early game snowballs. If I am at a table with three greedy gooses with low consistency in seeing lands BUT A HIGHER DEGREE OF SNOWBALLING when they DO see those lands, I can't look at their decks as I would my own. I would need to look at the probably that at-least one greedy deck snowballs. Even given low 70% odds (or even 60%) of a deck to hit all lands by turn 4, if 3 other decks at the table have 70% odds, there is a 1 - (.3)(.3)(.3) = .973 AKA 97.3% probability that at-least one of those greedy decks will hit 4 lands.
So that may be a factor in pushing you to play less lands even if not as likely to hit lands on turn 4. That's part of why I like to play less lands than 40 in low-to-mid power control decks, I need to have multiple spells availability in case I need to stop one early snowball. I can't just hit 4 lands and not have multiple pieces of interaction.
I feel justified in my land counts creeping up. Except Marvo. That guy shall forever gamble.
Speaking of, dang the price of many MDFCs too.
Dont forget to account for draw, not just ramp. With enough draw, you can make sure you never miss a land drop each turn, even if you're on the low side like 34.
I usually run 35 lands for commanders 6 CMC
Land count depends on your deck, simple as that. cEDH Rog/Si can rock 24 lands without an issue and is one of the best decks in the game. More control based cEDH decks run something like 28/29 lands and obviously a lot of fast mana. In casual is also depends heavy on your amount of draw and how low your curve is. I can 100% miss a landdrop or two if I just run more efficient cards and I will miss less land drops if I draw like 5 cards a turn cycle.
17 lands in a 40 card deck is for draft formats where you have next to 0 ramp and card draw is much more limited. I try to keep 50 pieces of “ramp” including lands in my deck which usually isn’t 37 lands and 13 ramp spells or rocks and works wonders
Most of my decks I built had 40 when I first started in EDH. However after I collected more efficient cards I did start to cut lands. But normally still run at least 38 for midrange decks.
You also need to factor in game speed and number of opponents into the equation. Commander games last longer than Standard and other formats in general, so you also have to balance the disadvantage of late-game land draws where you would prefer to have action.
The mulligan rule also allows you to attempt multiple "first draws" to ensure you can land at least 2-3 guaranteed lands to start off the game. We are really talking about turn 4 and 5 as make or break. This is generally the point in the game when someone says they are "mana screwed". I've often seen these same players come back to win because they weren't perceived as a threat.
This will be a good experiment for the Nikya deck i want to build.
That's crazy. All my decks range from 30 to 33 lands and perform quite well.
I can definitly agree that the 1 mana lotr cycling lands and mdfc make a lot of decks ive played run more smoothly
Every time I run more then 34-40+ I always get flooded depends on the rocks I run 32-36
Thank you for this video! One of my goals this year is to explore going against common deck building practices and try new things out. This lines up with something I’d like to try
I mean with the amount of man of ramp that are index currently anyway, the magic number is 42 right, you start getting rid of like two or three of those for MDFCs, and soul ring arcane signet and all the other package that everyone uses and has in their deck, all the sudden it gets to 38 and it makes sense this video didn't tell anyone anyone didn't already know
man, you followed almost exactly the same thought process as me when I made and adjusted my first commander deck the past couple months. I saw the 17 recommended for 40 card decks and calculated 42-43 for 100 cards, then I found that's too much, so I ended up on 40 cards through trial and error