Confession: I'm a bad deck builder.

[Versus]

Unless you're ultra patient or amazingly helpful, I would just back out now. (Only mods get to use red text. - Bardo)

This is something I should already know. I'm a little embarrased to even bring it up. I figured no one on WotC would help me out, so what the hell. If there are some good articles you can point me to on the subject, that would be more than sufficient.

There's a reason I play mono-colored decks. I'm too fucking stupid to create a balanced mana base. After all these years I blindly throw together dual lands and basics and call it day. This hasn't worked for me in the past and it still doesn't. I'd like to change this. I refuse to netdeck, so what happens is I'll take a mixture of spells I think work well together and then proceed to slaughter the mana base. Worse yet, I'll take a similar Archetype and convert it to match the colors I play. Is Red Death running 4 Badlands, 5 Swamps, 4 Mountains, and 7 Fetch? Okay, then I'll just run 4 Bayou, 5 Swamps, 4 Forest, and 7 Fetch for Green Death, not taking into account that maybe I'm running 8 spells with GG in the casting cost and Red Death doesn't.

I'd like to evolve a little bit, maybe you can help?

Here's the things I know (or think I do because I read it somewhere):

-Every non-land mana source (not costing above 3 itself) should be counted as half a land. Ex: If a MGA deck should run 18 Forest, by adding 4 BoP they can drop their land count to 16 Forest.

-Count the spells up and form a ratio.

-Double cc spells should count as 1.5 so 4 copies of Ohran Viper should count as 6.

I'm just gonna take a Standard deck list as Fetches will confuse the crap outta me right now.

I know the mana curve is effed here as well, but let's take care of one problem at a time...

Creatures-22

4x Birds of Paradise
4x Wall of Roots
4x Tarmogoyf
3x Ohran Viper
4x Doran, the Seige Tower
3X Treefolk Harbinger

Spells-16

4x Thoughtseize
4x Oblivion Ring
3x Crib Swap
2x Profane Command
3X Eyeblights Ending

I add that up to be 27 green, 14 black, and 11 white. I took into consideration the double cc spells and counted the activation of Villages as 1 green.

So that would give me a ratio of 27/14/11 or 51.92% green, 26.92% black, and 21.15% white.

This is where I'm clueless. How do I figure out what to use? What would Gemstone Mine count as? Pain Lands?

Here's the mana base I'm using. As you can guess I gleened it off of someone elses decklist who's running a different mixture of spells.

4x Horizon Canopy
2x Llanowar Wastes
3x Treetop Village
2x Caves of Koilos
3x Gemstone Mine
2x Urborg, Tomb of Yawgmoth
4x Forest
2x Swamp
1x Plains

By my calculations that would be 16 green, 10 black, and 10 white sources. 16/22 is far more than 51.92% though. From those numbers I would only need 11 green sources. Are these calculations all just bullshit?

The deck plays fine by the way, but I'd finally just like to know how the hell you do this properly.

Big thanks to anyone who hasn't fallen asleep at this point.

[Lego]

I'm a pretty terrible deck builder myself, but I'll try to throw a few thoughts out there just in case they help. First, the points you make about why you're a terrible deck builder are great. Now just stop doing those things.

Try not to be so analytical about your approach to creating manabases. That's not always the best way to go about these things. The ratio of colors in your deck isn't always the most important. If you're playing 4 colors, but all of your manafixing is Green, you're going to want more Green in the deck, even if maybe you have more of other colors. If you're playing Green/Black with an equal number of Green and Black spells, but all your green are early game cards and all your black is late game, you're going to want more green mana because you'll have time to find the black. Keep things like this in mind and it'll help a lot in building your base.

Finally, and this is, I think, the most important thing: Test, test, test! Throw together a rough list of what you think is a good manabase, then start testing it. Goldfish first, make some changes, and then test against things you expect to face or want to do well against, and continue to make changes. It'll take a while to get a base you're happy with.

(Incidentally, Urborg doesn't belong in that deck. You're running a single spell that costs BB, and it's a late game spell. Swamp would pretty much be functionally identical to Urborg in this case, without powering up opposing EEs. Also, you're playing GW, so there's little reason not to run Windswept Heaths, and then play a couple Bayous and replace the Caves with Scrubs. At that point you can get rid of the Gemstone Mines. Those are pretty terrible in a deck like this anyway.)

[DragoFireheart]

By the sound of your post it seems like you have a good grasp on the mana bases as you are looking at ratio of spell colour-to-land. Honestly I don't see how fetch-lands would confuse you.

Lets look at the Red-Death example:

4 Badlands, 5 Swamps, 4 Mountains, and 7 Fetch.

The Fetches would most likely be Bloodstain Mire and Wooded Foothills.

Think of fetches as a "wild" card in terms of what land it really is. Because this deck is running badlands, all of your fetches can get you a dual-land.

So the real land configuration would be something like this:

5 Swamps, 4 Mountains, 11 Badlands. It's because of the unique interaction between fetch-lands and Dual lands is why they are both so sought after. When you break down the mana-base of a deck like Red-Death you can get a more clear view of how other decks' manabases should look like to avoid color-screw.

Of course we will get those crazy-decks that need rainbowlands like City of Brass and Gemstone Mine but I don't believe that is what you are trying to understand here.

Back to the Red-Death example I was using...

We can see that ~50% of the mana base all of the colors used in this deck while the other ~50% will net you one color or the other, with 50% of that making black and the other half making red.

Lets say that lands that can color fix any card in your deck is "M" and lands that only give you one color is "S".

So the best ratio for a mana base would be 50% M - 50% S

If your deck has more black spells as opposed to red spells then you would adjust your S ratio to include more black lands as opposed to red lands.

For decks that start to get into 3 colors things start to become a little more crazy. [never mind expensive due to the number of duals/fetchs].

Did any of what I say help?

[Silverdragon]

The best way to do this imho is to first get a rough estimate of the manabase using the process you already showed and then test the hell out of every possible configuration that seems reasonable.
Although there might be a mathematical formula to arrive at just the right manabase it'd take far longer to figure out than to just test every iteration until you reach a point you're satisfied with.
For example you can never calculate the impact an opponent has on your manabase (either directly with Wasteland or Sinkhole, or indirectly by using landwalking creatures or free spells like Massacre or Submerge). You also can't calculate how important bluffing might be (Tropical Island -> Daze for example). One last thing that you can partly calculate (but is not set in stone) is when you want to cast your spells (Cabal Therapy for example is rarely a spell you play turn one even though it only costs 1 mana).
Generally I use the "George Baxter Analysis" built into MWS to get a first impression of what my manabase needs to function on the most basic level and then tweak it from there by goldfishing and testing against a Green Beats deck (no Disruption, just critters and pump spells). From there I go on against MonoR Goblins (no disruption except for manadenial) and finally Red Death (disruption.dec). If the manabase and deck as whole survive most of these games I go on to test against the rest of the gauntlet.
Hope that helps.

P.S.: I should add that this is only when I build a deck from scratch (which doesn't happen that often) so don't think I'm not the laziest person ever because I am
edit: @DragoFireheart I think you'd rather have all your manaproducers produce every color you need but because of the drawbacks of nonbasics you eventually have to cut them down for basics. Wasteland aside I'd definitely want a manabase of 20 Badlands over 11 Badlands and 9 Basics. Even considering Wasteland Red Death needs at most 2 Swamps and 1 Mountain so if could I'd play 17 Fetchlands.

[Versus]

Lego:

...but all your green are early game cards and all your black is late game, you're going to want more green mana because you'll have time to find the black.

I always neglect this. Lets say a 16 card deck has 4 counterspells and 4 Wrath of God. 4 are UU and 4 are WW. I would probably make the mistake of running 4/4 Islands/Plains when it should probably be more like 12/4 as WoG is a late game play.

(Incidentally, Urborg doesn't belong in that deck. You're running a single spell that costs BB, and it's a late game spell. Swamp would pretty much be functionally identical to Urborg in this case, without powering up opposing EEs. Also, you're playing GW, so there's little reason not to run Windswept Heaths, and then play a couple Bayous and replace the Caves with Scrubs. At that point you can get rid of the Gemstone Mines. Those are pretty terrible in a deck like this anyway.)

Of course I would run duals in Legacy, that would probably be easier for me than Standard. That deck above is just a Type 2 example as it's the deck I'm currently building and all the different duals, Gemstones, Man lands, ect confuse the issue for me. The Urborg inclusion by the way is two-fold; for one it stops some of the pain from all these crap "duals" in standard when used for B (colorless even with Horizon Canopy), it keeps Gemstone around longer, and doubles in LD when opponents are relying on it for Korlash/Tendrils.


Drago:

5 Swamps, 4 Mountains, 11 Badlands. It's because of the unique interaction between fetch-lands and Dual lands is why they are both so sought after. When you break down the mana-base of a deck like Red-Death you can get a more clear view of how other decks' manabases should look like to avoid color-screw.

I wasn't really thinking of it like that. So, to use that and your "M" and "S" examples Gemstone would be classified as "M" (I get that), but would you say that a simple B/G dual would be "S" or is that a third variable?

Silver:

tweak it from there by goldfishing and testing against a Green Beats deck (no Disruption, just critters and pump spells). From there I go on against MonoR Goblins (no disruption except for manadenial) and finally Red Death (disruption.dec). If the manabase and deck as whole survive most of these games I go on to test against the rest of the gauntlet.
Hope that helps.

That does help a lot. So far the above deck can withstand full out Aggro. Lots of stalling with Walls, Vipers, and removal until Goyf and Doran come down. However, I haven't tested against any LD or mass removal which this deck is susceptable to (Teeg is in the SB). I've actually lauded or dismissed entire decks based on their performance against only half the decks I'd be up against. Only to find that the decks it did beat were the minority of the environment.

Test, test, test!

That's really the answer in a nut shell I guess.

I appreciate all the replies. I realize it's not as easy as simply throwing down an equation and something I just need to do on my own.

All of this is a good start.

[Bardo]

If there are some good articles you can point me to on the subject, that would be more than sufficient.

By the esteemed Richard Feldman

The Constructed Manabase, Part 1: Mana Ratios

The Constructed Manabase, Part 2: Mana Ratios, Continued

The Constructed Manabase, Part 3: Color Balancing

The Constructed Manabase, Part 4: Detail Work

These are all free now, and while some of the specifics don't apply (karoos, signets, etc.), the lessons are worthwhile.

[Versus]

That's the shit right there! Thank you sir.

I've read through them once. Will do so again when I get home from work. One particualr paragraph JUMPED off the screen at me...

if you are taking your deck seriously, you should be spending a lot more time on your color base than just plugging some numbers into a formula. In the long run, Magic is far too complicated to forgive such behavior

[Citrus-God]

Another way to practice deckbuilding is to Draft a lot.

[Hoojo]

EDIT: Posted AFTER Bardo's post, so this may not be relevant.

I tell you what I do and you can decide for yourself.

First, I count up the mana symbols on all the spells. Eternal Witness gives me 2 , Rebuild gives me 1 , etcetera. Colorless mana does not count. I then divide each mana total in half, and that is the number of sources I need for that color. I count duals as 1 source for each color they produce, and fetches count as 1/2, so two polluted deltas is 1 black mana source.

This method can produce some high and low numbers for mana slots. I never go over 25 mana sources in a deck, and I rarely, if ever, go below 18. Using that threshold, I determine how "mana hungry" the deck is through playing. I consider mana to be cards that provide more mana than they cost to play. You have to use some discretion here, as some cards like Exploration and AEther Vial provide no mana, but they can still accelerate a mana base.

As a general rule, I start my testing using this chart:

Most cards in the deck are 0cc = 20 sources of mana
Most cards in the deck are 1cc = 21 sources of mana
Most cards in the deck are 2cc = 22 sources of mana
Most cards in the deck are 3cc = 23 sources of mana
Most cards in the deck are 4cc or higher = 24 or more sources of mana

This isn't a bulletproof method, but is has worked for me and doesn't require a lot of number crunching to come up with a quick list to test.

[Bardo]

if you are taking your deck seriously, you should be spending a lot more time on your color base than just plugging some numbers into a formula. In the long run, Magic is far too complicated to forgive such behavior

This is the best of all possible advice. Let me also say that building the proper manabase is in art itself and I have deep respect for people who can really pull a complicated one together.

Being the analytical person I am, when I first started building manabases, templating off old Sligh/Schneider lists, I would list all of my cmc's, number of colored mana sources and subject all of those figures to a formula to produce a baseline manabase. I think that's a good way to get started, since it forces you to think and consider what you're doing.

At the highest levels, all of those formulas and shit are sort of worthless--you really need to rely on good judgment and thorough testing. That's how I ended up with the Perfect Threshold Manabase. Lots and lots of rigorous testing with very little theoretical exercises.

An important caveat: I was kidding about the Perfect Manabase. Every manabase needs to be built with the deck and field in mind. The kinds of things that are possible in 1.x these days could not have been done pre-2005 rotation where Wasteland left the field. Also, non-basic hate wanes and ebbs overtime, so you never know.

The only way to get it right, is to test it and then test the mana some more. Also, take notes when goldfishing how often you hit three land on turn 3. How often do you find your spells uncastable due to color requirements, etc.

And everyone has their nuances, re: risk. I tend to err on the side of one or two too many land, where others may be more risk prone and can shave a land or two here or there where I wouldn't. I figure it's easier to salvage a game you're losing due to flood, than it is when you're mana-screwed. Not even Finkel can bluff or get their game together when they're getting clobbered with one land on the board.

[DragoFireheart]

Something else I would suggest is that mana-makers like Birds could be counted as 1/2 a mana source.

@ Versus: The 'M" ratio would be the number of lands that give you the mana fix for any spell in your deck while the 'S" ratio is for lands that only give you mana for one color of your deck.

3-Color decks are a bit different and would have 3-variables.

As you can tell getting the right mana fix is based on the deck. There isn't any one formula that you can take your 1337 Calculator and just plug numbers into. It's a bit of both math and some creativity.[Aka, super secret tech]

[Versus]

Another way to practice deckbuilding is to Draft a lot.

Another thing I suck at!

Most cards in the deck are 0cc = 20 sources of mana
Most cards in the deck are 1cc = 21 sources of mana
Most cards in the deck are 2cc = 22 sources of mana
Most cards in the deck are 3cc = 23 sources of mana
Most cards in the deck are 4cc or higher = 24 or more sources of mana

That's a good starting off point.

Also, take notes when goldfishing how often you hit three land on turn 3. How often do you find your spells uncastable due to color requirements, etc.

I'll start doing this for sure.

Thanks again everyone. Very helpful as usual.

[Versus]

One more question if you guys don't mind. I want to figure out the mana base on my own, but let me know if I'm heading in the right direction with the points I've made below the decklist. I want to do it right and know the reasons why it's right. Not simply that it works, so it must be okay.

Forget the OP list.

Creature-21

4x Saffi (GW)
4x Tarmogoyf (1G)
4x Ohran Viper (1GG)
4x Doran, the Seige Tower (GBW)
3X Treefolk Harbinger (G)
2X Shriekmaw (Evoke 1B is all I'm concerned with)

Spells-17

3X Griffin Guide (2W)
4x Thoughtseize (B)
4x Oblivion Ring (2W)
2x Profane Command (XBB)
4X Nameless Inversion (1B)

This gives me:

1cc=7
2cc=14
3cc=15
X=2 (used late game as finisher if possible)

-I dont want to miss my first 3 land drops.
-I want to be able to get B or G on turn one.
-GW, 1G, 1B on turn two.
-GBW or 1GG by turn three.
-White is essential by no later than turn 4.

Would that be the proper assesment? HAve I hit the mana curve correctly at least?

[Cait_Sith]

Okay, some problems:

1) This list is obviously T2. You are putting mana demands on your list that could irk Legacy, and it has 5 allied fetches and 2 sets of 10 duals (allied and enemy), plus a wide variety of color fixing outside of the aforementioned.

It cannot be reasonably done in Standard.

2) Your curve is top heavy, but with 26 slots open you could probably fix that. If you have any hope of reaching your requirements, then you need to make room for 23 lands and 4 Birds of Paradise, so cut something.

Right now your curve is just under too much stress because of the color requirements. I would recommend avoiding CC requirements outside of a single color and including some serious color fixing, or lightening the demands on your mana base.

[Versus]

Okay, good.

I had BoP and WoR in the original list, but some other folks lurred me away from them. Walls I dropped on my own. I didn't want too many 0/X creatures that rely on Doran to be any kind of threat. The first list I posted was doing very well, but I let others opinions sway my own.

Viper and Command are the only 2cc spells. I can see Viper becoming an issue. Profane Command is pretty late game where I never had a problem getting BB durring testing. Vipers could be cut, however they provide me CA or can stall a turn or two. Would adding the Birds back make playing Vipers more stable?

Saffi was a late addition, but her synergy with the Evoke mechanic is strong as is the ability to save Doran in a pinch.

I guess Griffin Guide could go, but T2 is fucking WoG/Damnation happy and this decks requires you overextend a bit.

Now we enter the other problem with me and deckbuilding...cutting cards. I'm afraid it's a lost hope.

[Cait_Sith]

Cut Command. It is the weakest part of your deck. You can use it to bolster Griffin Guide and to make room for BoP.

Saffi is the weakest card in your deck now. If you fear WoG, Gaddock Teeg defeats that easily. Go for it. If you are running Saffi solely because of how well she interacts with a 2-of, there is a problem.

[Versus]

Good points. I'll test it like that for a few hours

I did have two more evokers in the SB. That Hurricanish Cloudthresher for Faerie decks and Mesa Tokens. This is also where Teeg is btw, so I probably don't need him and Saffi, you're right.

Thank you for the advice. I really shouldn't even bother you guys with this crap. It's a Legacy board for christ's sake.

[Cait_Sith]

Thank you for the advice. I really shouldn't even bother you guys with this crap. It's a Legacy board for christ's sake.

These lessons apply equally to Legacy. Having a deck that wants BB turn 2, GG turns 3, and WW turn 4 is clearly a bad deck or is going to employ SERIOUS mana fixing (even then, it is a bad idea.)

The ability of people to make a deck with a stable mana base is part of the reason Goblins could dominate for so long. If the base has a weak point, Goblins could hammer than into oblivion.

[Versus]

Agreed. Not to mention over 50% of the lands I'm using cause me damage. All around it's pretty terrible. I guess Doran isn't an optimal Standard creature.

What I should do is play mono-red and run 4 Magus of the Moon as EVERY frigging Standard player are overextending their mana bases similar to my own not even including a single basic. There's one guy on there running Korlash in a 3cc deck that has 2 actual Swamps and 4 Urborg to support it. That's another story altogether.

I hate Standard, but it's all I have in the area. I spent $200 to make a Rack Deck I got to play for 2 months. Now that Bob is gone it's useless. Bah!

[kirdape3]

Actually B/g Rack is considered to be one of the great winners out of the Ravnica rotation.