Building A Consistent Manabase

By Guest Author FlyingDelver


The printing of Assassin’s Trophy has given a breath of fresh air to GBx decks, and their popularity is growing. In GBx decks, there is a lot of wiggle room when it comes down to which spells you want to run, and how many copies to run. However, when people are discussing decklists, I often hear these sorts of arguments:

Risk Factor is not a card you want in Jund.”

“You certainly want to have Tasigur over Bob in Rock.”

Anafenza is too narrow to run in Abzan.”

Ultimately, when talking about decklists, it is way less useful and constructive to talk about exact numbers of spells since all of those things might be tied to a certain metagame and personal preference. GBx is an archetype which wins on thin margins, and a big aspect of this is making educated hedges for your metagame. So is it really useful to convince somebody that two Kolaghan’s Command are better than one without any metagame awareness? What we can break down to a science is the way a manabase should be built to satisfy the color requirements of the cards you want to play. This should be the first and foremost focus upon reflecting on existing decklists.

Every GBx player has been in a position where they’ve died with a ton of spells in hand. In that moment, it didn’t matter if you had two Kolaghan’s Command in your deck or one, because the deck didn’t function. It may have been because it didn’t curve out well, or stumbled, and inconsistent manabases can be at fault for this. The way to ensure this does not happen is to have a consistent deck, and a consistent deck is fundamentally defined by the manabase.

“Liliana, the Last Hope” by Anna Steinbauer

I follow the rules which Frank Karsten has given us concerning mana sources, mana count, and mana curve. Magic is a game of strategy and statistics, and the latter is particularly important when deck building.  Frank Karsten does a great job of breaking that down in his articles here and here.

When it comes to examining statistics in deck building, the differences can be extremely small, to the point where they’re almost indiscernible, but GBx decks operate on those extremely thin margins. The point is to gain every single competitive edge, no matter how insignificant it may seem. The punishment for stumbling is much harsher in today’s modern, and that is why it is so important to focus on the little details. You should play with this mantra in mind: If you stumble, you lose.


“Thoughtseize” by Aleksi Briclot

There are always certain requirements for a Jund manabase which will never change from a mathematical point of view (the bold numbers represent the highest requirement for each respective color):

These are normally the limiting factors for the color requirements of a Jund manabase, so this should be the basis from which we build our decks. We should fulfill these requirements in order to maintain as much consistency as possible.

There are other factors  to consider as well. The exact number of lands to run in a Jund deck is a big debate going on in the community with some feeling 24 lands is correct and others 25. I think the answer is rather simple:it depends on the curve of your deck.

If you run only a few one-mana spells, but many two and three-mana spells, as well as a several four-drops, you want to play 25 lands. This way, you’ll have a higher chance of hitting land-drops four and five on a consistent basis. This should allow you to curve out so that spells don’t jam up in your hand and ensure that you get the most value out of your mana and spells. If you have a clunky mana-curve, you need to make sure you run an appropriate land count.

The rule is simply this:

If you run 10-11 one-mana spells, seven three-mana spells and a playset of four-mana spells, then you want to run 25 lands.
If you run 12-13 one-mana spells, six to seven three-drops and three to four four-drops, you can be fine with 24 lands.

DO NOT cross this line with 24 lands. I once saw a list running four Bloodbraid Elf and seven to eight three-drops in a 24 land manabase. This is greedy as hell.

The land count is also intertwined with the amount of creature lands and other potential tapped lands. The rule is simply:

When you run 24 lands, play three creature lands. When you play 25 lands, play four creature lands.

Creature lands in particular can be awkward or even devastating in an aggro matchup. You don’t want them to hinder your curve early on. The only exception to playing four creature lands in a 24 land manabase is IF and only IF you expect a heavy midrange or control meta. Since games go longer in these matchups the extra creature lands don’t hurt you as badly.

Here is the manabase shell that you can play around with:

The number ranges can be explained by this:

The Twilight Mire substitutes the ninth fetchland (Bloodstained Mire), so you have to decide between those two. When you run two Swamps and two Forests, run a fetchland distribution of four Verdant Catacombs, two to three Bloodstained Mire and two Wooded Foothills. When you play two Swamps, one Forest and one Mountain, choose four Verdant Catacombs, three to four Bloodstained Mires and one Wooded Foothills as your fetchlands.

Now with this recipe in mind, let’s look at a couple of different examples.


This version contains 18 black sources, 18 red sources and 18 green sources (17 green if you don’t run the 25th land in the form of Treetop Village). So this manabase is perfect if you want to play Anger of the Gods in the sideboard and also pretty good when you want to play Finks as well (for the 25 lands build).

You run no basic Mountain, which means you are a bit weak to Field of Ruin (FoR), but a basic Mountain is pretty awkward to have in the opening hand, so your opening hands are better without it.

You run the ninth fetchland which means you’ll have the required consistency for the red sources, but more fetchlands means more pain from your manabase. Overall, this manabase is really consistent but is weak to aggro and FoR.


This version contains 18 black sources, 18 red sources and 17 green sources (16 if you run 24 lands and 17 if you run 25 lands, again the cut is the Treetop Village). This manabase is also perfect if you run Anger in the board, but can’t quite support Finks.

You run a basic Mountain, which means you have better game against FoR decks and also have a good pain-free red source to fetch against aggro. Basic mountain can reduce the quality of opening hands though.

You only run eight fetchlands with a Twilight Mire in place of the ninth fetch. This really helps illustrate what a more pain-free manabase can look like. Against aggro decks this matters but similar to the Mountain, Twilight Mire can be awkward in opening hands. This manabase does lack a bit of fulfillment for Finks, but overall it’s probably the most pain-free way to build it in an aggro-driven meta.

You can weigh these two example manabases against each other and even experiment with numerous variations in between like these:


Land combination
Mana sources
2 Forests + 9th fetchland
high consistency, weak to FoR, painful
18 black, 18 red, 17-18 green
1 Forest, 1 Mountain + 1 Twilight Mire
lower consistency, strong against FoR, pain-free, nonbo with Damping Sphere
18 black, 18 red, 16-17 green
2 Forest + Twilight Mire
average consistency, weak to FoR, rather pain-free, nonbo with Damping Sphere present
18 black, 17 red, 17-18 green
1 Forest, 1 Mountain + 9th fetchland*
average consistency, strong to FoR, rather pain-free
18 black, 19 red, 16-17 green

*This version is a very special one, as you can potentially run a Treetop Village in a 24 land manabase here. Since you have 19 red sources naturally, you can just exchange one Ravine (going down to two copies) and bring in a Treetop Village. You go down to 18 red sources that way, which is still enough to cast Anger out of the board. This is another route to take if you want Anger, but still want 24 lands and Treetop all together.


“Assassin’s Trophy” by Seb McKinnon

The basic theory applied to Jund also applies to Rock as well. Here are the typical mana requirements for a Rock manabase (the bold number represents the highest requirement for a given color):

  • 18 black sources for Liliana on turn 3.
  • 18 green sources for Finks in the Sideboard.
  • 14 black sources for Discard/Fatal Push on turn 1.
  • 14 black and 14 green sources (in addition to having at least 21 lands that either produce green or black) for a turn 2 Decay/Trophy.
  • 13 black and 13 green sources (in addition to having at least 20 lands that either produce green or black) for Maelstrom Pulse.
  • 13 green sources for a turn 2 Goyf.
  • 16 black sources for a turn 4 Damnation.
  • 16 green sources for a turn 4 Thrun/Baloth etc.

The limit of this manabase is the 18 black sources and then depending on which green cards you run, between 14 and 18 green sources. Most Rock decks want to run Field of Ruin, which is obviously great, but you need to be aware that FoR does not count as a mana source for any card with CMC one through three. This means FoR does not help with casting Liliana on turn three, making the manabase really tight.

The creator of Rock, Sol Malka, does not advocate for running FoR at all, but rather Ghost Quarter. His reasoning is that he builds his deck with the understanding that it’s not a Tireless Tracker deck. He doesn’t care about getting up to five or six lands. It also lets him interact with opposing lands for one mana instead of three, and that can be very relevant against Tron. Also, since UW Control plays both FoR and Paths, you will be out of basics quickly making your own FoRs worse. His point is that when you are not trying to go as large (four Trackers), you should play Ghost Quarter, but this is still up for debate.

One difference between the Jund rules and Rock rules for manabases is that Rock decks have always been able to run four creature lands, regardless of whether they have 24 or 25 lands. The reason for this is that the manabase is overall not really painful. You can afford to fetch for basics more often and rely less on your creature lands for mana fixing. Of course, they are still important considerations since they can hinder your ability to cast spells on curve by entering the battlefield tapped. The overall shell for a Rock manabase should be the following:

  • 4 Verdant Catacombs
  • 0-2 Black Fetchlands
  • 4 Blooming Marsh
  • 3-4 Field of Ruin / Ghost Quarter
  • 3-4 Swamp
  • 2 Forest
  • 2-3 Hissing Quagmire
  • 2 Overgrown Tomb
  • 1-2 Treetop Village
  • 1 Twilight Mire

If you play 24 lands, then only run three FoR/GQ. You can’t get to 18 black sources otherwise. If you play 25 lands, then you can run four FoR/GQ or a fifth black fetchland. When you run the fourth FoR/GQ, then you have to run three Hissing Quagmires in order to maintain 18 black sources. If you run the fifth fetchland you can run a 2/2 split of creature lands while also achieving 18 green sources. If you run Ghost Quarter in general, you can play a fetchland instead of the fourth Swamp, which generates another green source.

So basically, the best manabase for a 24 land build is as follows:


This manabase runs 18 black sources and 17 green sources. You basically don’t have much wiggle room here. This manabase only loses a smidge of consistency when it comes to casting Finks. You can get away with that by playing a bunch of Dark Confidants, which can help dig for lands, in conjunction with Twilight Mire which can produce double green from a swamp. In that sense it kinda feels like 17.5 green sourcesand since Kitchen Finks is usually in the sideboard, it’s admissible.

When it comes to 25 land manabases, you have two options:

The first is to have four FoR instead of three. You’ll have the same amount of black and green sources while maximizing your FoRs; it’s also the most pain-free version.

The other thing you can do is add a fifth fetchland instead. This is a way to achieve 19 black sources and 18 green sources so that everything gets cast easily. However, the extra fetchland makes your manabase slightly more painful.  You also now have the freedom to change one Quagmire into a second Treetop. This would net you 18 black and 18 green, which is perfect.

Here are some other variations that you can experiment with that consist of 25 lands:

Land combination
Mana sources
4 FoR/GQ + 4 fetchlands + 3 Quagmire
average consistency, pain-free
18 black and 17 green
3 FoR/GQ + 5 fetchlands + 3 Quagmire
very high consistency, painful
19 black, 18 green
3 FoR/GQ + 5 fetchlands + 2 Quagmire + 2 Treetop
high consistency, painful, stronger creature land power
18 black, 18 green


“Lingering Souls” by John Stanko.

The last GBx deck I want to talk about is Abzan. The principles of building a good manabase with Abzan are the same as Jund and Rock. Here are the typical required manasources for Abzan (the bold number represents the highest requirement for the colour):

  • 18 black sources for Lilianas.
  • 14 black sources for turn 1 Discard/Push.
  • 14 green and 14 black sources (in addition to having at least 21 lands that either produce green or black mana) for a turn 2 Decay/Trophy.
  • 13 green sources for a turn 2 Goyf.
  • 12 white sources for a turn 3 Lingering Souls.
  • 14 white sources for a turn 1 Path to Exile.
  • 16 white sources for a turn 4 Gideon, Ally of Zendikar.

So depending on what exact cards you play, your white source requirement can actually be really different. You are fine with 18 black sources and 14 green sources pretty much all of the time, but white is not clearly definable. It depends on how heavy your white splash will be. There is the option to only splash for Lingering Souls and sideboard cards,and then there are versions like Abzan Traverse that splash more heavily for cards like Path to Exile.

The required amount of white sources needed when building an Abzan deck that is only splashing white for Souls and sideboard cards would be 12. However, some people are also running a couple of Paths in their lists. For them, I would crank the requirements up to 14 sources. Since the best versions include Tireless Tracker as a top end threat, you should be playing 24 lands. Builds with 25 lands are pretty uncommon to run in Abzan, as there are not many four drops that are popular enough to support in the maindeck. Cards like Siege Rhino see play, but it’s usually only a one-of Gideon, Ally of Zendikar played in sideboards. Here is the basic recipe for an Abzan manabase:

Most of the time three creature lands in a 24 land manabase will be best. As for the basics, there is not much wiggle room. If you don’t run Path, you can theoretically run a 2/2 split of Swamps and Forests, but beyond that, the extra Plains is needed as a white source for Path. So this would mean a 2/1/1 split of Swamp/Forest/Plains.


As for the fetchlands, the distribution should be the same as Jund. If you run two Swamps and two Forests, then run four Catacombs, three Marsh Flats and two to three Windswept Heath. If you run two Swamps, one Forest and one Plains, then run four Catacombs, four Marsh Flats and one to two Windswept Heath. Also note that we do not run less than nine fetchlands. This is due to Tracker being way more impactful with more fetchlands. If you opt for more Blooming Marshes, then you need to run nine fetchlands. Otherwise, you can run upwards of 10.


This build has 20 black sources, 18 green sources and 14 white sources. This list is capable of casting anything except for maybe Gideon out of the sideboard. This version is fine if you don’t want Gideon or other white cards, as it’s rather pain-free when you run three fastlands. There are two adjustments you can make if you want:

You can run a 2/2 split of basics. That way you can play 10 fetchlands to obtain white consistently when you don’t need white for Path.
Or, you can keep the Plains and go back down to one Forest instead. This would increase the white count to 15, which will enable you to cast white cards like Gideon.

Here are some possible variations to play around with.

Land combination
Mana sources
9 fetchlands + 2/1/1 split of basics + 3 Blooming Marsh
high consistency, pain-free
20 black, 18 green and 14 white
10 fetchlands + 2/1/1 split of basics + 2 Blooming Marsh
very high consistency, rather painful
20 black, 18 green and 15 white
9 fetchlands + 2/2 split of basics + 3 Blooming Marsh
high consistency, rather pain-free
20 black, 19 green and 13 white
10 fetchlands + 2/2 split of basics + 2 Blooming Marsh
very high consistency, painful
20 black, 19 green and 14 white
“Traverse the Ulvenwald” by Vincent Proce.

Of course, there are some key differences to building an Abzan Traverse manabase. When it comes down to these kind of builds, the important thing to mention is that they typically run three Traverse the Ulvenwald and four Mishra’s Bauble. For that reason, 19-20 lands is usually sufficient. Traverse itself is counted as an actual land in this build, which increases the land count to 22-23; four cantrips adds another, making a total of 23-24 lands (cantrip rule of thumb: about four cantrips being played in a deck means you can run one less land). Here are the typical mana requirements for cards that are run in Traverse builds (the bold numbers represent the highest requirements for each respective color):

  • 18 black sources for Lilianas.
  • 14 black sources for turn 1 Discard/Push.
  • 14 green sources for turn 1 Traverse.
  • 14 green and 14 black sources (in addition to having at least 21 sources that produce green or black) for turn 2 Decay/Trophy/Grim Flayer.
  • 13 green sources for turn 2 Goyf.
  • 14 white sources for turn 1 Path.
  • 12 white sources for turn 3 Souls.

With this in mind, the recipe for this build looks like the following:

  • 8 Fetchlands
  • 2 Overgrown Tomb
  • 1 Godless Shrine
  • 1 Temple Garden
  • 2 Blooming Marsh
  • 4 Shocks
  • 1-2 Creature lands

Here, the manabase is very tight and again there is not much wiggle room. You have to run the basic split of two Swamps, one Forest and one Plains because you want to grab every color basic with Traverse the Ulvenwald. Otherwise, you can’t count it as a white mana source. The only real wiggle room is the choice of playing 19 or 20 lands, and typically, a second creature land counts as the difference.

Here is a possible manabase you can build in Abzan Traverse:

The 20th land to play is the Shambling Vent. Overall, this adds up to 19-20 black sources, 19 green sources and 15-16 white sources. Note that these numbers are very high, but they aren’t 100% consistent like actual lands. The three Traverses are counted as all three color sources, but won’t add anything to turn one plays and the same goes for Bauble. The requirements for the turn one plays (now excluding the traverses and baubles) are 15 black sources, 14 green sources and 11 white sources. So we can see, the manabase is quite dependent on the Bauble and Traverse package to function. Without it, you can only cast Path consistently on turn three (which you may want anyway). With this in mind, the addition of the 20th land does not quite help with the turn one plays, but does help to curve out better when no Traverse is involved. It helps to reach a higher consistency and makes the manabase less dependent on Traverse for mana fixing.


Marsh small
“Blooming Marsh” by Adam Paquette.

Next time you evaluate a deck, instead of looking at how many Dark Confidants they’re playing or whether they have Leyline of the Void in the sideboard, look at the manabase, and whether they’re playing the appropriate amount of colored sources for their spells. Remember that the spells are always tied to expected metagame whereas the manabase is not. The manabase is tied to the spells being played. Help other GBx aficionados increase their consistency and deck performance by following the guidelines. Thin margins are important and as I said earlier, you don’t want to stumble because when you stumble, you lose.




FlyingDelver is deeply involved in the G/Bx Modern community as the author of MtGSalvation primers for Jund, Abzan, G/B Rock, as well as the administrator for the G/Bx Midrange Discord, and part of the administrative team for the G/B Rock Facebook page. Please find relevant links in the external resources section. You can support FlyingDelver through his Patreon page.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s