Land/Mana Ratio Help?
Deck Help forum
Posted on Feb. 23, 2012, 10:59 a.m. by Miasma
Hi. I know there is a statistical formula somewhere that will help get just the right amount of land/color producers in relation to others in your deck, to get a great land ratio. Does anyone know the formula, and can explain it to me? I would really appreciate it. Thanks.
There is a formula used for calculating the chance of drawing a certain number of cards by a certain turn, but I don't think there is an actual formula that takes into account anything except chance of drawing a land. 24 lands is generally accepted as good for enough to hit most of your land drops without getting flooded (depending on how aggressive the deck is). The chance of drawing any particular card involves using a hypergeometric distribution and is explained here: http://www.kibble.net/magic/magic10.php
February 23, 2012 11:56 a.m.
21 lands for 0,1,2,3 cmc Agro decks. 22 lands for 1,2,3,4,5 cmc decks. 23,24 lands, is the most lands you should have in your 60 card deck. Anything more then 23 lands, in a 60 card deck, will mana flood screw you.
February 23, 2012 2:13 p.m.
that is incorrect mike. Many control decks run 24-27 because land is crucial these decks often finish extremely well as well. Any more than 28 and your running a Land legacy deck. 28 your running Valakut. 24-27 your control. 24 and under is tempo aggro and midrange.
February 23, 2012 2:29 p.m.
askewnotion says... #6
I keep a base of 20 lands as my standard unless I'm doing something a little crazy. From here, I add the ceiling level average CMC of my deck as lands. This number is then reduced by my mana accelerators.
For stays-in-play non-creatures, I subtract .75 if MC <=1, .5 otherwise. Creatures I treat as .5 and 1/3, respectively.
For exploding fetches of permanent mana sources (something that fetches land, for example), I use only 80% of the values above.
For temporary mana sources, I use 60% of the values above.
This makes things ready for a rough play-testing. You may have to tweek up or down a land or two once you get a feel for how the deck plays. Quick agro decks can get away with a little less while decks that run more mid range or thrive in mid-late game may need an extra land or two.
When deciding what land to use, I always try to keep my mana in proportion to the number of symbols on the card. For activate abilities on the car, I count the symbols as half. If you are playing a splash color and are seeing a lot of double/triple symbols on it, you may need to shift the ratio to favor your splashed color a bit more.
I also weight my dual lands according to the color intensity of my deck. For example, a Drowned Catacomb in a deck that is 80% black and 20% blue would count as .8 black and .2 blue. In a deck that is 60% black, 20% blue, and 20% white, this land would only cover 80% of the decks color needs. Thus it would be .6/.8=.75 black and .2/.8=.25 blue. Doing this helps me overcompensate for splash colors by adding more support for them, leading to less dead hands.
There is no real science, but these ratios have always worked for me. Just use some rough guidelines, and play from there.
February 23, 2012 3:14 p.m.
MasterFlinter says... #7
24 lands in a 60 card deck has become the norm because of the power of 4 drops in standard recently. Cards like Day of Judgment are extremely important to hit on turn 4 because if you wait another turn a decent aggro deck may kill you. By using 24 lands you are making 40% of your deck be lands. By turn 4 when on the play you will have drawn a total of 10 cards. 40% of 10 = 4. If your curve tops out at, say, CMC = 3 and it isn't important to consistently hit 4 lands then you only want for 3 of your first 9 cards to be lands, or 33%. 33% of 60 cards is 20 lands. This isn't the end-all-be-all but it is a decent way to get started.
Choosing how many of each color source to use isn't as easy and there are many more factors to impact your decision. You need to look at how many cards you have of each color, how many colored symbols there are of each color, how high the total casting cost of each card is, and whether it is important or not to be able to cast the card immediately or if you can wait and still use the card effectively.
February 23, 2012 6:02 p.m.
Example scenario's that often happen with 20,21,22,23,24,25 lands
20 lands: Opening hand 1,2 lands, quite OFTEN. No,1 land Mulligans: OFTEN. Mana FLood: Rare. Turns 1,2 no land, Turn 3 no land, turn 4 finally draw a land. Occurence: Semi Often.
21 lands Opening hand: 1,2 lands:between semi uncommon to semi often. Mulligan: Semi uncommon to Semi often. Open hand average: 2 lands. Turns 1,2: 1, no land, turn 3: land, turn 4: no land. Turn 5 1 land, no land turn 6: 1 land. Mana Flood: Rare
22 lands: Open Hand ave: 2,3 land. Mulligan: Uncommon to Semi uncommon Mana Flood: uncommon to semi rare. Turns 1: no land, turn 2: 1 land, turn 3:no land turn 4: 50/50 on 1 land, no land. Turn 5: 1 land if no land turn 4.
23 lands: Open hand Ave: 3 land. Mulligan: Uncommon to semi rare. Mana Flood: Semi uncommon to uncommon. Turn 1 50/50 on get land. Turn 2: land. Turn 3: no land turn 4: land. 4.5, 4,5 lands by on turn 4.
24 lands: Open hand Ave: 3,4 lands. Mulligan: Semi rare to very rare. Mana Flood: Semi common to Semi Uncommon. Turn 1: land. Turn 2: no land. Turn 3: land. Turn 4: no land. Turn 5: no land Turn 6: land. 5.5, 5,6 land by turn 4. 6.5, 6,7 land by turn 6.
25 land: Open hand Ave: 4 land. Mulligan: Extremely rare. Mana Flood: Common to Semi often. Turn 1: land, turn 2: no land. Turn 3: land. turn 4: no land turn 5: land Turn 6: no land, turn 7: land. 6.5, 6,7 land by turn 4. 8 land by turns 6,7.
Now based on those common average scenario's, which would you want on average? I would want the 23 land. 23 land will consistently get you 3 lands in your staring hand. usually no Mulligan, usually no mana flood. and usually 4.5, 4,5 lands by turn 4.
And 23 land can work with almost any deck. I have played semi aggro to semi control decks, even some control decks with 23 lands. And 23 lands consistently works with most everything, and should be the standard base amount of lands.
21 land for aggro decks, 22 land for semi aggro and 23 lands, for base, standard, and 24 lands for semi control to control decks, and 25 lands for extreme high cmc extreme control decks
anything higher then 25 lands, like 26,27 lands, = hard core mana flood, no get cards to cast.
February 23, 2012 7:53 p.m.
26 or 27 is only used in control for that reason, it has enough draw spells to be ok with getting that much mana. But yeah, I agree with the analysis above.
February 23, 2012 11:08 p.m.
I'm sorry guys. I seem to have asked the wrong question. I mean ratio in relation to other lands. I already know how many lands I want. I just need to know how to get the perfect combo within the amount.
Here's an example:
Currently I have 3x Hinterland Harbor, 2x Glacial Fortress, 4x Sunpetal Grove, and 2x Razorverge Thicket, 4x Island, 4x Plains, 3x Forest, and 1x Evolving Wilds. (total = 23)
I want to know the formula for deciding which lands to use in relation to the mana symbols in the spells I use. (like whether I want 4x Hinterland Harbor, and only 1x Glacial Fortress)
Thanks guys for the previous information.
Cableguy says... #2
No real formula or algorithm but for me it depends on the deck.
Tempered steel decks can get away with running 19-20 land because they have a low converted mana cost plus Mox Opal
Control decks can never miss a land drop and interact with the opponent much more often as such they typically run 24-27 land depending on the cost of the cards you have.
As far as distribution of colors goes it can be much more complicated depending on how many colors you run. In a two color deck you want max double lands like Drowned Catacomb
and Darkslick Shores
and X number of colorless lands. After that count how many you have left to add.
I had 12 land left to add for basic lands and my blue cards made up 52.94% of the deck (the number of blue mana symbols) and black made up 47.06%.
Now it isnt an exact science but I did "12 x 52.94%" and got 6.3528 and did "12 x 47.06%" and got 5.6472.
I than rounded up to the next whole number for the largest number aka blue to 7 and kept the lower number at its whole number which was 5.
Again not exact but as you can see pretty darn close.
UB Control (Revised Revision)
February 23, 2012 11:22 a.m.