Spoiler: At the bottom of this article.
The other day I was conversing with some mages in a forum, and I was questioned about the reasoning behind the numbers of each card in my Four-Color Control deck. In answering this question, I realized that this is a topic I have not seen discussed lately. It is a fundamental element of deck building on which 98% of players don’t have that strong a grasp, including many strong players, and even some strong deck builders.
To start, what is the most common number of a card to play?
If you said “four,” take a second. Stop and think about it again.
No, it’s not “one” either.
The most common number is “zero.” That’s right, zero is going to be the right number of a card to play the vast majority of the time. You may not think this says much, but I suspect upon further reflection you will see the wisdom that lies within.
It is more than just every card needing to justify itself; it is the fact that when you are thinking about how many copies of a card to play, always consider that it might actually be zero. This is especially important in a deck or a format with a lot of sacred cows, cards that you have to play (just ask anyone).
It’s not too hard to see that playing four is often right, as you will often be in a situation where the card is one of your better cards and you want as many copies of it as possible. As such, this is generally the next most common number in Non-Vintage formats.
The next level of theory people come to realize is that with a single card you give yourself access to some kind of effect somewhere in your deck, while minimizing your chances of drawing it in all the matches against which you don’t want it. For instance, you might have a Tormod’s Crypt for your Trinket Mage; a Squee, Goblin Nabob for your Survival of the Fittest; a Sundering Titan for your Tinker; or a Tendrils of Agony to eventually find with any of your tutors.
One-ofs make a lot of sense when used this way, but when used without any library manipulation, it is most commonly a single late, late game victory condition. Or it could be a card with extraordinary diminishing returns, or a legendary land. Maybe it’s restricted (in Vintage), or perhaps you’re in need of a fifth copy of something (which we will discuss in a moment).
Sounds simple enough… but here’s where it gets tricky. When do you play two copies of a card? What about three copies?
This is one area where I learned a lot from Adrian Sullivan. While I did not always agree with his choices of numbers for cards in a deck, I eventually came to realize that he was not the contrarian I thought he was, as there was actually sound reasoning behind the often surprising or seemingly eccentric card numbers.
The number one reason that you’d run twos and threes is to manipulate the odds that you’ll draw a card. Not every card is something of which you want as many copies as you can get. This is where the concept of diminishing returns comes in. Some cards get better the more of them you play, like Accumulated Knowledge, Lava Spike, and Muscle Sliver. However, many cards get worse the more you play.
Technically, every card sees diminishing returns at some point (except maybe Rocket Powered Turbo Slug), but generally, if this point is much greater than four, it is not particularly important other than dealing with trying to produce 5th, 6th, 7th, and 8th copies of cards by using a weaker version in addition to the one you really want.
Take a card like Cruel Ultimatum, for instance. This card is clearly very powerful and can be game-winning. However, this doesn’t mean you just stick four in your deck, even if your mana can support it. In fact, most Cruel Ultimatum decks that have been successful used 2 or 3 Cruel Ultimatums, since you actually get hugely diminishing returns on each additional Cruel Ultimatum.
During the course of a game, you will generally only have to cast Cruel Ultimatum once in order to win. This makes the second and third copy you draw worth less than the first. Looking at it the other way, we see that drawing the second Cruel Ultimatum actually hurts you even more than the first when you are getting beaten down and only have a couple land on the battlefield.
Here, the question becomes at what point is the increased harm from drawing Cruels early too costly when compared to the ever-decreasing value of drawing more Cruels. You may think that this is just a function of the expense of Cruel Ultimatum, but it is not really about that.
Look at Terminate, a fantastic tournament staple creature-kill card. As good as it is, the more Terminates you play, the less they are each worth. Adding a fourth Terminate actually only increases your chances of drawing one by turn 3 by 23%, whereas adding the first Oblivion Ring gives you chances to ever draw it. This is not to say that your deck should be all singletons, but rather that attention needs to be paid to the relative value of each copy of each card. Often, particularly in aggressive decks, the point where additional copies of a card are not worth it is above four. For instance, we would clearly play five Arcbound Ravagers in our Affinity deck, right? However, nothing is sacred. Sometimes you will see control decks with 3 Tarmogoyfs or 3 Baneslayers. How can this be if those are the best creatures, or even the best cards?
The first Baneslayer in your deck generally adds more of an element to your game than the fourth, though generally people still play all 4 Baneslayers since, even though the fourth is not as good as the first, it is still better than most of your cards.
Another common reason to play the twos and threes is to break the Rule of Four. Basically, sometimes you want to run five or six or seven copies of a card. One way to accomplish this is by running some other version of the same cards. Kathari Remnant is not as strong as Bloodbraid Elf, but we really wanted to play at least six Bloodbraid Elves in Honolulu, so much so that we played two Kathari Remnants. It wasn’t that we were worried about drawing too many copies, nor did we need to draw one copy… we just wanted six cascade spells at the four spot, and it was the next best option.
Your deck will often have requirements that will be at odds with just playing a list of four copies of the nine best cards. Imagine that you’ve determined that you have a mana curve that wants fourteen creatures that cost one mana. If this is true, and it may well be, you are not going to be able to play four of each, generally meaning three cards that are four-ofs (presumably the best ones), and you play two copies of the weakest.
Obviously another classic reason people run three copies of a card is to stick the fourth in their sideboard to access with Cunning Wish, Burning Wish, or Living Wish. Still, this is really just sort of the same principle as the one-ofs, but slightly differently applied.
It should be kept in mind that if your deck has a lot of library manipulation, whether it is Divination or Impulse, you will see more cards over the course of a game. This means two things. First, you need fewer copies of cards to find them (particularly if you don’t want to draw a bunch of them). Second, decks with a lot of library manipulation tend to not have room for four copies of every type of card, since so many of their cards just manipulate their library or make mana.
If you want to be able to sweep the board, have multiple types of spot removal, creatures with which to win, game-breaking spells, Planeswalkers, card draw, and permission, you are going to run out of room if you use all four-ofs. This is why so many control decks end up playing twos and threes, such as with Five-Color Control decks.
To give a glimpse into some of the reasoning that goes on when deciding on the numbers of each card, I will break down my Four-Color Control Deck from States.
4 Lightning Bolt – Lightning Bolt is cheap, efficient removal. The added versatility of being an instant, hitting Red haste creatures, hitting Planeswalkers, and combining with other direct damage to form a back-up road to victory condition had me wanting to max on Bolts, rather than play something like Deathmark. I needed to recoup lost tempo from lands that enter the battlefield tapped, so having one-cost removal is important, and Bolt is just a stronger card than almost all of my cards, though I am not sure I would play five.
3 Earthquake – I use three because I really like drawing a sweeper against some decks, and I can usually find a use for at least one against most decks, but I start getting diminishing returns pretty fast. Drawing two or three Earthquakes can be potentially very bad in some matchups. In control decks, three is a very common number for effects like this, and experience has shown it to balance the tension between really wanting it against Boros and Eldrazi Green, but not being able to handle drawing two against Red and Jund.
When playing control decks that draw a lot of cards and dedicate most of their deck to drawing cards and adding mana, you begin to lower the numbers of all the types of effects. The idea here is that rather than play four of each answer, you play three, and all the extra space is spent on Draw-Twos or other library manipulation. Then you end up with a better chance of drawing the answer despite playing fewer copies, since you are looking at more cards. This also has the practical offset of leading to some card advantage, which you sorely need on account of having the potential to draw the wrong answers alongside having so many lands in your deck. It is not that card advantage wins games, it is that it gives you more options, which you need to make up for having bad options, needing the correct options, and having fewer to begin with on account of running so many lands that don’t give you the option of using them as spells.
2 Terminate – I used two Terminate as a function of the number of Lightning Bolts and Wall of Denials I had selected, and because I wanted nine spot removal spells for creatures, based on my experience with decks like this in general, and from recent playtesting. Nine most closely approaches the ideal balance between “I really want to draw one against almost everyone,” and “I can’t afford to draw three against some people,” and “I can use Earthquake, Flashfreeze, and Ajani Vengeant to take pressure off these slots.”
The other reason I selected 2 Terminate and 3 Wall, instead of 1+4 when I think Wall is better, is a combination of mana curve considerations as well as the fact that Terminate can solve some problems that Wall of Denial cannot, such as Dauntless Escort when I have Earthquake, or Vampire Nighthawk or World Queller. Creatures like World Queller make me want to be able to find a Terminate at some point, and with all my Draw-Twos, it is reasonable to think that I could find a Terminate eventually. The second Terminate increases the chances seeing one in the first 20 cards by 70% (compared to one Terminate), whereas the fourth Wall of Denial only increases my chances of having a turn 3 Wall against Thrinax or Ball Lightning by 23% (compared to three Wall).
3 Flashfreeze, 1Double Negative – Flashfreeze is clearly the best maindeck counterspell these days, but I opted for a single Double Negative instead of the fourth maindeck Flashfreeze for a few reasons. First of all, once again there’s diminishing returns, as drawing more Flashfreezes makes them worth slightly less. Second of all, this decreases the odds of drawing two against non- Red/Green decks, as per the discussion on three-ofs above.
Finally, I would generally like to draw a Double Negative and a Flashfreeze rather than two Flashfreeze, and one Double Negative is not terrible to draw, but I really, really don’t want to draw two Double Negatives and zero Flashfreezes. With the fourth copy being a Double Negative, I give myself chances to draw it without ever getting “screwed” with two. In addition, it is not just that the Double Negative is the fourth “cheap counter,” it is also doing double duty as the fourth “hard counter” (in addition to Traumatic Visions), as I determined that four counters that could hit non-Red/Green cards was the minimum I wanted to play here.
Dropping to three is a decrease of 24% by turn 5, a drop I am not willing to accept, whereas the increase to five does offer an increase of 14%… but I am not willing to accept the associated increase in drawing too much countermagic. Nor do I have the slots, as I have had to shave numbers across the board to make room for everything I wanted to play.
It should also be noted that I could help both of these aspects by playing more Double Negatives, but I opted not to do that as I think Double Negative is a far weaker card than either Flashfreeze or Traumatic Visions. I am only playing the one as a concession to space restrictions, as well as because of the non-zero value of having access to one. When your opponent knows you do not have Double Negative, they can play more liberally, whereas the constant threat of it makes it more difficult to play against you.
4 Esper Charm, 2 Divination – Esper Charm is one of the absolute best cards in the deck. I played four because that is the limit, and I would play more if I could. In fact, I play two Divinations because I want more Esper Charms so badly I am willing to settle for a couple of Divinations to circumvent the Rule of Four, as we discussed above. Why six? This is mostly the result of lots of playtesting, as I experimented with four, five, six, seven, eight, and nine.
3 Wall of Denial – I initially tried four, as I played in Rome. It is the best creature removal against Jund, Mono-Red, and a number of others, plus it was great against Boros, and at least good against most people. Why not four? Wall of Denial is another case of diminishing returns, as every Wall beyond the first is actually not that good most of the time. My mana curve was definitely a consideration as well, but basically, I wish I could play 3.3 Walls. I decided after much experimentation that three was better than four at approximating this.
2 Ajani Vengeant – I played three in Rome, and they had been nothing but good for me. However, something had to give as I only had 60 slots. I can definitely imagine more being correct, but I opted for only two as there is more at play than just the diminishing returns of drawing two Ajani. I have so much power at the top of my curve, so I want to err on the side of cheaper cards in general, if it comes down to the fourth Bolt or the third Ajani.
Now you may ask why not cut the Nicol Bolas that we will discuss in a minute for a third Ajani. That is a perfectly valid question and it may be correct. However, after a lot of playtesting, I determined that I wanted a powerful way to take over the game besides just Cruel Ultimatum, especially since occasionally I will end up in games going long that Cruel doesn’t 100% put away. Basically, I thought I needed a way to go ever bigger, and I prefer Nicol Bolas to Obelisk of Alara. While Ajani works well with Wall and is just a great card, it is a main phase four-drop, so there is some tension there.
3 Traumatic Visions – I had four in Rome, but it was just a bit slow, and I wasn’t using the second Traumatic Visions I drew as a counterspell enough to be worth it. In addition, I am primarily using the Visions to fix my mana, and since I have two more Divinations than I did, I believe that the mana fixing from them is enough to make up for the lack of a fourth Visions, especially since I upped the Rupture Spire count this time around to ensure the right colors. The downside to playing more Ruptures and Divinations is that I must make up for the lost tempo, hence my adding Terminates where I had a O-Ring and a Wall, and cutting a Visions to help make room for a fourth cheap counter, as at Worlds I only had three.
3 Sphinx of Lost Truths – At Worlds, I only played one, and had two Sphinx of Jwar Isle main. I resisted for a long time, but Michael Jacob continually pointed out that Sphinx of Lost Truths was helping more consistently, whereas Sphinx of Jwar Isle was usually only helping more against Jund, though that was the most common matchup. While they are boarded out a lot, they have added value game 1 since they let you filter the wrong answers, and they make your Cruels far better.
3 Cruel Ultimatum, 1 Nicol Bolas, Planeswalker – I only had two Cruels at Worlds and wanted a third, so I added it, as it is very important to draw 1 eventually. I wanted to draw it so badly that I would be willing to play a fourth (but not a fifth), even risking drawing two. However, I opted for Nicol Bolas instead of the fourth Cruel, as I think that the number of times where Nicol Bolas will offer me an avenue of gaining an advantage that I would not normally have had, such as dealing with an Eldrazi Monument; Emeria, the Sky Ruin; or against Thought Hemorrhage, that it was worth the small decrease in chances of having a Cruel on turn 7 when I need it. This is case where I actually think I might have to play at least two of these bombs to win a game, so I am not afraid to play four, since only playing three actually increases the odds that I won’t see a second copy to an unacceptable level.
3 Island, 3 Swamp, 2 Mountain, 1 Plains – I thought it important with my Traumatic Visions and fetchlands to have enough basics to cast Cruel Ultimatum. I wanted a Plains so as to make my Traumatic Visions able to find White, though it doesn’t cast Cruel obviously, so I opted for only one despite Baneslayer Angels in the board. Finally, I added the third Island from a combination of wanting one more land that entered the battlefield untapped, untapped my Glacial Fortress, wanting more Blue to activate Traumatic Visions, wanting to be able to get more than just enough Blue for Cruel later in the game, and wanting to make sure my Scalding Tarn can hit.
4 Arcane Sanctum, 4 Crumbling Necropolis – I found that the Trilands were fantastic and vital to the manabase working. I don’t have much on turn 1 anyway, and a single Triland goes a long way to make my mana work, with a second one usually ensuring that I am golden. The tension is obviously that drawing all tapped lands is awful, but I try to make up for this with tempo plays like Bolt and Terminate. Also, I have found it much worse to not be able to cast my spells than to always be a turn behind.
3 Rupture Spire – Obviously playing three Rupture Spires starts to flirt with danger, as they are so slow, but I really don’t want to be stuck being unable to cast my spells. As long as I only draw one, I am usually pretty happy to see them and don’t mind tapping out turn 2, since I can recover from Thrinax with Wall and Blightning with Divination. Sometimes I hold up the Flashfreeze turn 2, rather than play the Rupture, usually when they have no creatures yet, i.e. no Leech and on the draw. I only use three since I really like drawing one, but the drop in relative value of the second one I draw is horrible. This is a good example of how different cards have very different rates of diminishing returns.
3 Scalding Tarn – This land is totally awesome, and I wouldn’t mind the fourth except that in testing I ran out of lands to fetch once too often when I needed them. In addition, there is a small amount of tension between Scalding Tarn and Esper Charm, since you often get Red early which makes Esper Charm more difficult. However, you can usually plan around this. I would play a fourth if it did not come up so often that I run out of targets.
2 Glacial Fortress – This is a nice way to get White mana without messing up Cruel. I don’t play more because there are so many Islands already that it sucks to get stuck with six lands in play where three are Islands plus Glacial Fortresses, as they might as well be Plains for Cruel. I chose 2 Glacial Fortress as I found that I wanted 19-20 lands that helped with Red and Black for Cruel. The first Glacial Fortress is decent, though not a superstar, and the second one is below average sometimes, hence two.
1 Exotic Orchard – This is a little bit of greed, and a little bit the fact that I wanted another Red source and thought one Exotic Orchard would be better than an Arid Mesa, as I played at Worlds. It is Red most of the time, and against people that are not Red I Traumatic Visions for Mountains. It is very bad to draw two against Mono-White or Mono-Green, and I am not willing to risk it, plus it is actually bad to draw two against a lot of people since it will often only be Red or only Black or only Blue/White, making casting Cruels off it awkward. Still, it is usually at least one of my colors, and often two, and it is untapped, so I gave it the nod.
Now on to the sideboard…
2 Baneslayer Angel, 2 Malakir Bloodwitch, 1 Sphinx of Jwar Isle – An important part of this strategy is adjusting your kill cads to be the ideal ones to get the job done. Sphinx of Lost Truths is more of game 1 card since it filters the wrong answers. However, the guy you want after boarding varies a lot depending on the matchup.
Baneslayer was good against creature decks and graveyard based combo decks, plus it is just better after boarding since they always have less removal. I played two instead of four since it makes their removal worse. With four, they can just keep all their removal in without fear of it being dead. Baneslayer Angel is so good I could totally imagine playing more, though I did not think it was worth the space, since I rarely want more than three or four creatures total, and usually I want Bloodwitches, Jwar Isles, or Lost Truths as well. Also, having different names is important to fight Thought Hemorrhage.
Bloodwitch was for every non-Boros White deck, and some Boros decks. Jwar Isle was mostly for Jund decks, though it was randomly good against some reactive decks of the era. I only used one copy as it has a lot fewer applications than the other two. I also found that I want to have only three or four creatures at a time, so this configuration lets me optimize the mixture.
1 Flashfreeze, 1 Essence Scatter, 3 Negate, 1 Hindering Light – Just as it is important to adjust my creatures after boarding, it is important to adjust my countermagic. Not everyone was playing Red-Green, and not everyone is totally reliant on creatures. This mixture gives me the greatest chance to have the answers that I want in any given matchup, while having a slant towards Negate, as I need to be able to take out creature removal against people that don’t use many (or any).
Negate is the obvious card to put in against such decks. I use the fourth Flashfreeze because that is the limit, and the best. I use Essence Scatter and Hindering Light over the fourth Negate as I want to have a variety of options, making it harder for opponents to play against me, and because I think there are more situations where the value of a first Scatter or Hindering Light will be worth more than the diminished value of a fourth Negate. For instance, against Jund, the threat of Hindering Light is a stronger one than Negate, and I wouldn’t play four Negate against them anyway.
In addition, I might face a Ranger of Eos deck, making Essence Scatter gain a lot of value. When I play against someone where I want the fourth Negate, I won’t miss it as much as I would miss having a second cheap counterspell that can counter Ranger. I don’t play Double Negative here because it is not powerful enough. It is mediocre against most, rather than especially good in a few key matchups.
1 Thought Hemorrhage, 1 Pithing Needle — High-powered answers to very specialized situations. More diminishing returns, as one copy is far more than zero, and I don’t have room to play as many of everything as I would like, so I play one of each to get the most value out of each slot. Using one Needle and Hemorrhage increase the chances of seeing it to 50% each after seeing half of my deck, whereas if I used two of one and zero of the other, I would have a 75% chance of drawing one, and a 0% chance of drawing the other.
In addition, I wanted at least two great cards to bring in against graveyard combo decks, but I could not justify Relic of Progenitus, as the graveyard decks were not popular enough. These both hose Extractor Demon, so it is two slots for that. It is more important to have the first Needle against Elspeth, Knight-Errant or Luminarch Ascension than the second Hemorrhage against Control or Magic Christmas. Likewise, the first Hemorrhage is more important than the second Needle, as the second copy of a card only increases your odds of seeing it by half as much as the first copy of a different card. This is the real reason you see masters like Gabriel Nassif play so many one-ofs in sideboards.
1 Oblivion Ring, 1 Deathmark – These are minor ways to tweak the control elements. The main similarity is that they both kill fat creatures that are not particularly vulnerable to Lightning Bolt, since my maindeck relies heavily on Bolt. O-Ring has the whole “single copy is much more than zero” for dealing with things like Howling Mine or Eldrazi Monument. Deathmark is the ideal removal spell for tuning my deck against a lot of the field and I do not play more simply because it would mean cutting a Bolt for a Deathmark which doesn’t have enough value to justify the use of the space. The first Deathmark is much better because in many matchups, I actually want to increase the quantity of removal in my deck by one.
So, what should you take away from this article? Aside from hopefully a better understanding of the thought process that goes into selecting the number of each card, one’s ability to analyze other people’s decks can be markedly improved.
The next you are examining a decklist either from a writer, a tournament winner, your buddy, or a brew you just made, it might be interesting to go through it systematically and look at the numbers yourself. When you do this and ask yourself if the numbers make sense in the light of all we have discussed here, it can provide some valuable insight into the mind of the deck’s designer as well as potentially offering opportunities to potentially improve the deck.
Remember, nothing is sacred, and while it may be a useful time saver to develop shortcuts like “always play Black Lotus in Vintage,” there will be exceptions (Dredge, for instance), and when you have an opportunity to examine a deck in depth with time not a factor, it can useful to question everything. You don’t have to spend a ton of time on every card choice, but a greater understanding of the theory behind the numbers can be invaluable.
When your friend shows you a brew with 3 Baneslayer Angels and 2 Sphinx of Jwar Isles, ask him, “why those numbers?” You see someone with 3 Lightning Bolts and 4 Elvish Visionaries, before just snapping to make the “obvious” swap, take a moment to put yourself in the shoes of the deck builder. Give them the benefit of the doubt! Maybe, just maybe, there is a reason behind their choice. Perhaps they are wrong, but obviously it is not just a case of them not getting the memo that Lightning Bolt is generally a better card than Elvish Visionary.
As an experiment, it might be interesting to look at the Top 8 decks from this past weekend’s awesome SCG $10K St. Louis Open Weekend and see if the numbers hold up to scientific scrutiny.
One last thing before we roll out… I had to mention the “Buy-a-Box Promo Card.” It’s recently confirmed as real:
That’s right, this card is actually as amazing as it looks. It’s not just a good manland, it’s also a good mana fixer. I mean, Sejiri Refuge is seeing regular play, and it is definitely not because of the one life. This card is not just good because of the lack of better options.
It seems like a nice fit into a format where it can be challenging to advance the board against a deck like Jund, when any permanent you play is at risk of getting Terminated, Pulsed, or Bolted. Holding a creature leaves you vulnerable to Blightning. What to do?
Well, Celestial Colonnade lets you advance your board quite a bit without Jund having immediate access to an answer. It is also not incidental that it lets people have a much higher concentration of creatures than previously. It is not unthinkable to imagine a Bant deck with 4 Celestial Colonnades and 4 of the Green/White manland. That is a lot more kill cards, regardless of whether it is a control deck or aggro.
If there is one thing I know about the Japanese, it is that they love them some manlands. Do you think there is any way they won’t be playing tons of these (especially if any of the rest of the cycle can compare to this one)? I also predict there will be even more main deck Goblin Ruinblaster running around.
Okay, obviously this card will be played. Having the option to buy a Serra Angel has got to be far more relevant than gaining one life. Seriously though, just how good is this card on intrinsic power level?
Compare this to Svogthos, the Restless Tomb, which was a reasonable tournament card. First of all, let’s look at the creature Svogthos gives you. What is a one-sided Lhurgoyf worth? Well, Lhurgoyf at four saw a little play but not much, so maybe three mana for a one-sided Lhurgoyf is good.
Would you pay three mana for a Serra Angel!?
So the creature is better, what about the land as a mana source? Svogthos enters the battlefield untapped but only produces colorless… is this better or worse than entering the battlefield tapped and producing two colors?
Back in the day, they used to cost entering tapped as being worth about as much as producing a color instead of colorless. Time has shown that entering untapped and producing colorless is slightly better than a colored mana entering the battlefield tapped. This is why Blasted Landscape was more popular than Remote Isle.
So how much is a second color worth? Quite a bit, actually. Since both options (colorless untapped or dual land tapped) produce one mana each turn, the increased value from a second color compared to a first is actually going to be just about exactly the increase in value from the first color versus colorless.
Let’s say in Candyland (this is where we talk about abstract hypotheticals that don’t involve Lotus Cobra) we value the ability to “produce a color and be untapped” at 100%. Now, let’s say that “producing a colorless and being untapped” is worth 67%. Now, let’s say that “enters the battlefield tapped but producing a color” is worth 50%. That would imply that the a colorless mana is only worth 2/3rds of a colored mana (which as a general rule is not too bad), which would make a land that enters the battlefield tapped and produces colorless only worth about 33%.
Does this mean that Coastal Tower (and by extension Celestial Colonnade) is worth 100% since, a tapped Island would be worth 50%? Remember, you can’t tap it for both at the same time. In addition, we are concerned with how much the color is actually worth and in our make-believe example, we said that color was worth 33% (compare colorless untapped to Island). That would mean that Coastal Tower is actually only worth about 83% of an Island (and Crumbling Necropolis is worth somewhere around 116% in a three-color deck, making it not surprising that the Trilands are so popular in their respective shards).
So basically the question is whether or not the option to buy the service of a 4/4 flying vigilance creature for five mana is worth 17%. Obviously other factors have to be considered too, such as “I need to fix my mana, one way or the other,” and “I can only play so many lands that enter the battlefield tapped.”
What can we use as a parallel? Faerie Conclave has seen a ton of play in just about every format, so let’s compare them. How does 1U for a 2/1 flier compare to a 3UW 4/4 flying vigilance creature? All things considered, those are actually fairly comparable in Constructed (though heavily slanted in Limited). So is Celestial Colonnade on the same curve as Faerie Conclave?
Oh that’s right, Faerie Conclave only produces one color.
Celestial Colonnade is totally absurd.
Happy Holidays! See you next week.