The mana base of a deck simply refers to the lands and spells that help provide the mana for that deck. Defining exactly what is and isn't part of the mana base is a good place to start.
Anything that actually produces mana or accelerates your mana is an obvious inclusion. In current formats these include cards like Orochi Sustainer, Kodama's Reach, Sakura-Tribe Elder and Rampant Growth. The one obvious thing to note is that these cards are all green. Fortunately this isn't always the case. There are plenty of artifacts like Fellwar Stone, Honor-Worn Shaku and Wayfarer's Bauble that offer acceleration to other colours as well. You sometimes see other cards that can function as part of your mana base in certain circumstances, too. Cards like Vedalken Engineer and Gift of Estates can also be considered to be part of your mana base for example. For the purposes of this article I'm going to lump all of these cards into the general category of “mana generators”.
There are other cards that do provide mana or allow you to draw additional lands that I wouldn't consider part of a mana base. This is either because they cost too much mana themselves or because the mana they provide is only temporary.
Cards like Seething Song do accelerate you but they only do it for a single turn. In constructed that turn can make a huge difference; an Arc-Slogger on the third turn instead of the fifth turn is a massive advantage. In limited, however, the games take more turns to complete and the format is much slower in general. Card advantage also tends to be much more important in limited. While you might be able to use a card like Seething Song to get your Flowstone Crusher out on turn three, in general you don't want to lose card advantage by using a Seething Song to achieve that. A much better plan is to use a Rampant Growth to provide a permanent increase to your mana base while still bringing out the Crusher a turn earlier than normal. In general, cards like Seething Song aren't playable in limited because of the inherent card disadvantage you can't help but gain through using them.
Cards like Journeyer's Kite and Seek the Horizon do provide extra lands but they don't really allow you to run fewer lands, simply because they are so expensive to cast in the first place. You can't really include these cards as part of your mana base for that reason. While they can help in creating some card advantage and also help in gaining access to a splash colour they should not be included in the deck in lieu of lands because they require so much mana to activate or cast in the first place.
The first thing to consider is the quantity of mana that will go in your deck. In general this means the lands, but it also includes the other mana generating cards in your deck. If you're playing a 9th Edition draft and have two Fellwar Stones, a Llanowar Elves, a Rampant Growth and two Wood Elves then you probably don't want to run 18 lands. If you did run 18 lands you'd find yourself getting mana flooded quite a lot as a number of the spells you drew would have little to no impact on the actual game. If you get to a mid-game point where things are fairly stable and your opponent draws ‘Land, Dark Banishing, Land, Hollow Dogs' against your draws of ‘Land, Fellwar Stone, Land, Rampant Growth' then even though you've drawn the same number of spells over those turns you're obviously going to be way behind on the board at the end of those turns. This is one of the inherent dangers of including too many mana accelerators and is something that has to be considered when constructing your deck.
The general rule to stick to (ignoring any mana generators for the time being) is to play 17 lands in decks with average or low mana curves, and 18 lands in decks that have a number of five or six casting cost spells. You can very occasionally get away with 16 lands but that should only be considered in decks that have mana costs consisting almost exclusively of one, two and three mana spells, with maybe a couple of four mana spells at the top end. In a deck like that you can get away with 16 lands.
Heavy requirements in both colours will often require extra lands or other mana fixing to compensate
On the flip side, you might want to increase your land count to 18 lands because your colour commitments are so high. If your deck contains Foul Imps, a Thieving Magpie and an Aven Windreader then you might want to play 10 swamps and 8 islands to give you a better chance of getting the coloured mana you need for those cards. Even though the mana costs of the spells in the deck don't dictate the inclusion of the extra land, perhaps the coloured mana requirements do. The same can be true when you're splashing a colour; if you have Samurai of the Pale Curtain, Moonbow Illusionist and are splashing for three powerful red cards, then you might want to consider upping your land count to 18 to better support your colour commitments.
The other reason you might want to play a lot of lands is if you have something to do with them, or if you think your deck is just so good you want to try and minimize the possibility of mana-screw. A recent blue-white 9th Edition deck I drafted had Slate of Ancestry, Trade Routes, Wrath of God and two Thought Couriers amongst its highlights. This deck had ample ways to gain back card advantage and three cards that could cycle away excess lands to draw more spells. Although this deck had a low mana curve with not one single spell costing more than four mana, I felt that it still warranted 18 lands. It had so many things to do with those lands and ways to recover any card disadvantage lost through mana flooding that the extra land to prevent a loss to mana screw was important in this deck.
The general rule of thumb when choosing the number of lands you play in decks that contain other mana generators is to remove one land for each two or three additional mana generators in the deck. If you have a Rampant Growth and a Llanowar Elves you can probably run one less land than you were planning to. If you have five additional mana generators you can probably lose two lands safely.
The reason why you just can't trade these out on a one-for-one basis is because you still need sufficient lands in your opening hand to cast your accelerators. Dropping below 15 lands would considerably increase the chances you'd be forced to mulligan. You also don't want to miss too many land drops in the meantime. Consider a situation where you cast Rampant Growth on turn two to fetch a land, then miss a land drop and cast Wood Elves on turn three to fetch another land and then miss the land drop next turn again. In this situation it's turn four and you still only have four lands in play. Your opponent has been developing their board, possibly with second, third and fourth turn creatures whereas you have the same number of lands as them but with only a pathetic 1/1 on your side.
The general rule of thumb is to remove one land for each two or three additional mana generators in the deck.
The main point of mana generators like the ones I've discussed relates back to that favourite topic of mine: tempo. While they do lose you tempo when you cast them – a Rampant Growth on turn two instead of a Grizzly Bears puts you at an obvious board disadvantage – the theory is that you regain that tempo and then some by being able to deploy your more expensive spells before your opponent. If you don't also draw the extra lands you need naturally then you don't really gain any of that lost tempo back again as you're forced to cast spells of the same cost as your opponent. This is why you still need a respectable number of lands to go along with any additional mana generators you might play.
By quality, I basically mean colour: how well does the mana base you're selecting support the casting of your spells?
There are some decks that make it very easy. If you only have spells with a single coloured mana in their casting cost, and you are only running a two colour deck, then it's a fairly safe bet to run nine of one land type and eight of the other. The extra land would go to the colour that either had the highest number of total cards or the highest number of early drops.
The other most common land split is a 10-7 split which you would do when you had a lot of double coloured casting costs in one specific colour. If you were blue-black with Wicked Akuba, Befoul and Death Denied then you really want at least 10 Swamps to cover those cards. In that sort of deck you really want to avoid any double-blue cards like Shapestealer or Hinder. Something higher up the curve like Sire of the Storm is more acceptable simply because by the time you hit six mana you'll hopefully have something like four Swamps and two Islands on average.
It's rare that I'd play a two-colour deck with an 11-6 land split in favour of one colour. That's a pretty extreme balance and the sort of deck that had this split would probably be something like a black-red deck that contained Consume Spirits, Looming Shades or maybe a card like Grave Pact while only really containing 6-8 red cards maximum. You would consider running eleven Swamps and only six Mountains in that deck. Having only six lands of a single colour really increases the risk that you won't draw one so your colour commitments to the main colour really have to be very significant indeed to consider this split.
When drafting you should be paying attention to the coloured mana requirements of the cards you pick up. If you are drafting white-blue for example and already have Ismaru, Hound of Konda, Samurai of the Pale Curtain and Moonwing Moth then cards like Shapestealer basically become unplayable. If the card in question is very powerful and not necessarily needed early in the game – something like Kiku's Shadow for example – then it can still make the deck but you have to be prepared for the possibility of it sitting in your hand for a few turns until that second Swamp turns up.
Things get tricky when you have to consider splash colours as well. The general consensus states you want access to three off-colour mana sources for 2-3 off-colour spells and an extra mana source for each extra spell. It's this area where the mana fixers really come into play. If you have two Rampant Growths, for example, you can certainly get away with just two off-colour lands and possibly get away with just one if you feel you need to. If you have a deck with cards like Elvish Warrior and Trained Armodon you'll want access to your splash colour without necessarily hurting your main colour. Cards like Rampant Growth excel under these circumstances as they both accelerate you and find whichever colour you need.
A lot of the other green mana generators only provide access to green mana. Llanowar Elves, Orochi Sustainer, Overgrowth and the like can therefore obviously only count towards your green mana count when deciding how much of each colour to run.
The other consideration is the colourless land issue. Often you'll draft lands that you want to play, but those lands will only generate colourless mana. Lands like Quicksand and Miren the Moaning Well will often warrant inclusion in your deck but they will have a negative impact on your colour commitments. There are two choices here really; the first is to consider them lands and accept the consequences of the decrease in coloured mana. The second is to consider them spells, which means adding them as an additional land to the mana base you would otherwise use. Often Quicksand's effect will be equal to a removal spell you would consider running, so treating it as a spell instead of a land simply means cutting a weaker spell for it. This has the side benefit of giving you a slightly increased land count and decreases the chance of mana-screw as a result. This second method is often the best way to go unless you have a deck that has very low coloured mana requirements.
Putting it into Practice
That's enough of the theory, let's look at a real-life example.
This is from a recent 9th Edition draft I did. It's a green black deck and the core cards that will definitely make the cut are:
1CC: Llanowar Elves
2CC: Elvish Warrior, Enfeeblement
3CC: Razortooth Rats x2, Rootwalla, Coercion, Dark Banishing,
4CC: Gravedigger, Bog Wraith, Giant Cockroach, Giant Spider
5CC: Hollow Dogs, Kavu Climber, Gluttonous Zombie
6CC: Craw Wurm, Force of Nature
That's a total of 17 cards. The mana generators I've drafted amount to:
And the remaining cards include the following playables:
So there are several questions here that need considering:
- How many mana generators would you run?
- How many lands would you run?
- Which cards do you choose to fill out the deck, and does the mana base of the deck influence these choices?
- What balance of forests and swamps do you go with?
Before I tell you my answers, have a look at the cards above and think about how you would finish the deck with that pool of cards. Once you've decided what you'd go for, click on the links below to see my own answers.
In this deck I think I'd actually run all of them. With the Force of Nature in there and the other high casting cost spells I think you want all of the Overgrowths and Wood Elves. This deck is also lacking second turn plays and the Rampant Growth and Fellwar Stone fit into the curve nicely there. If there was a third Overgrowth I'm not sure I'd include that but two seems good for this deck. This takes the deck up to a total of 22 cards now.
Now that you know how many mana generators you're running it's time to decide how many lands you want to run.
In this deck I'd go with 18 normally, if there were no additional mana generators. We have access to up to five additional mana generators and there's already a Llanowar Elves in the deck. If the mana curve were a little lower here we could consider dropping three lands as there are six mana generators in total, but with this deck the mana curve is very high and I think you can only realistically drop to 16 lands. We also want to make sure we hit three mana to be able to cast the Overgrowths and Wood Elves and 16 lands helps.
So now the quantity of the mana in the deck is set. The next thing is to finalise the remaining cards you want in the deck as they will probably impact the number of swamps and forests you want to run.
There are a number of alternatives in this deck. I'd relegate the Naturalize to the sideboard initially. As the deck is a little slow early on, I wouldn't want to suffer the life loss of the Serpent Warrior and so I wouldn't include that either. This deck does have a lot of mana available to it, and if it happened to draw one of the Overgrowths alongside another of the mana generators it would only need to see five lands to be able to cast the Scaled Wurm. That seems achievable to me and I think I'd include the Wurm as a late game card. The next choice is between the three black cards. Highway Robber is an average card but in this deck finding the mana to cast it might be difficult. Enfeeblement has the same problem of course but that is an automatic inclusion because it's a cheap removal spell, but the Robbers are a four mana 2/2 which isn't great at all. Consume Spirit is the best card by a long way but that has an even bigger problem when it comes to black mana. It requires to do anything and at least to be useful against most creatures. Even though that is a big problem, it is still a removal spell and with Rampant Growth and the possibility of the Fellwar Stone generating black mana I think it's still a better choice than the Spineless Thug. I'd round out the deck with Scaled Wurm and Consume Spirit, which takes the total spell count to 24.
The land count and the spells are finalised. All that's left now is to decide on the appropriate balance of the land types.
There are 16 lands to decide on. Green mana is obviously extremely important to this deck as not only does it aid the casting of most of the creatures in the deck it's also a requirement to cast almost all of the mana generators. I think ideally you'd want to see ten Forests in this deck, as you also want to be able to cast the Elvish Warrior on turn two. Including ten Forests leaves room for only six Swamps and I don't think that's enough. There are two double-black spells in the deck and they are important as they are both removal spells. There's also the consideration of the two Razortooth Rats as these will often be the three-drop of choice. As a result I think the deck really needs that seventh Swamp. This leaves room for nine Forests which I think is acceptable as just one will allow the casting of Llanowar Elves, Overgrowth and Wood Elves and any of those will give access to at least which allows the casting of everything except the Force of Nature.
1 Llanowar Elves
1 Elvish Warrior
1 Wood Elves
1 Razortooth Rats
1 Bog Wraith
1 Giant Cockroach
1 Giant Spider
1 Hollow Dogs
1 Gluttonous Zombie
1 Kavu Climber
1 Craw Wurm
1 Force of Nature
1 Scaled Wurm
1 Dark Banishing
1 Consume Spirit
I've separated the deck out like this so you can see that it only really contains 19 ‘real' spells. The Wood Elves might possibly trade with a Goblin Piker or something but in general this sort of deck has a lower number of spells than the standard '17 land/23 spell' deck. That's just something that comes with the territory – one thing it does have on it's side is much more powerful creatures. You won't see a standard draft deck playing multiple five, six and eight casting cost creatures like this one does.
I hope this article has given you a good overview of the things to consider when deciding on the mana base for your limited decks. While it's impossible to cover every single example in a single article, most of the general areas you need to think about have been covered and now you have some standard rules you can apply when building your own deck.