A good curve is something that is often brought up in deck design, but I don’t often see explanations for it, except maybe the reasoning that you should have something to do in each part of the game. Well, its a bit more in-depth than that.
I was at Rotterdam Team Limited GP last year. I didn’t win many games, because I got the rest of the pool. My deck was horrible. One of the very few games I won, my first four turns looked like this:
[draft]Minister of Inquiries
Sage of Shaila’s Claim
Not exactly the curve you want, but it didn’t matter. The simple fact I was able to curve out and use all my mana during each of the early turns, gave me a good headstart and I was able to win.
Okay, so why is curving out so strong?
Games are about resource management. You all know why card advantage is good and disadvantage is bad. Its the same thing here. Mana is a resource and you want to manage it right. However, in this case, the resource is different. Whereas you can keep your cards in your hand, if you want to, and they want go anywhere (well, you know what I mean), you have access to your mana during a specific window. After you untap, the mana you didn’t use during the previous round, is lost. So, use your mana when you can.
This should be the basis of all deck construction. You want to use mana as effectively as possible, so you have to at least give yourself a chance to do just that. Shouldn’t that mean you should have a similar amount of cards in each part of the curve?
Not really. First, you might get manascrewed, in which case you want more early drops. However, casting two spells, before your opponent, is very strong. This can be most easily achieved by a concentration of cheap spells. Being able to play a creature to an empty board and then be able to destroy opponent’s next play is a very good move, or stealing the initiative from draw by killing the opponents two-drop and playing your own after it. You need cheap spells to do that.
On the other hand, there aren’t usually that many playable one-drops in Standard. There are usually some, but often that one-mana spot is mostly used for setting up something else, like through [card]Attune with Nature[/card], [card]Oath of Nissa[/card] or simply playing a land tapped, which is a sort of one-drop in your curve.
You should also remember that certain cards don’t lend themselves very well into your curve. You might not have a chance to play counterspells on curve, because you might not be able to take off a turn to cast them or you might not have valid targets for them.
So, what does a good curve look like?
Well, that’s not an easy question to answer. According to Salvation Wiki, a basic curve should look like this:
8 at one, 7 at two, 6 at three, 5 at four, 9 at five, 25 Land.
Well, playing nine five-drops isn’t really ideal. This doesn’t take into account any of the points I’ve made before about double-spelling, mana-screws, lands that come into play tapped and so forth. The thing is, its not really an exact science. We haven’t even talked about ramp or card draw yet (and I’ll save those for a different article anyhow).
So, how to proceed? The best way is to start by benchmarking (or copying) someone else’s curve. There’s plenty of people, who have thought about these questions long and hard. Most importantly, they have tested these things thoroughly. Because, well, you can make any amount of calculations you want, you still need to test things out. Simple calculations will get you pretty far, but they do have their limitations.
The idea of the curve was first put into use in this deck deviced by Jay Schneider, although it commonly carries the name of Paul Sligh:
4 Brass Man
2 Dwarven Trader
2 Goblins of the Flarg
4 Ironclaw Orcs
3 Dwarven Lieutenant
2 Orcish Librarian
2 Orcish Artillery
2 Orcish Cannoneers
2 Brothers of Fire
2 Dragon Whelp
1 Black Vise
4 Lightning Bolt
2 Dwarven Ruins
4 Mishra’s Factory
4 Strip Mine
The idea behind this deck is that you do your best to use your mana all the time. Counting the [card]Black Vise[/card], the deck has 9 one-drops. I don’t count the burn spells here, because with this deck, you didn’t generally need to use the removal that early and you would often rather save to use it on your opponent anyhow.
It also has nine two-drops, six three-drops and two four-drops. It has more lands than you would have in a deck these days based on the curve, but since you have so many ways of using that mana, you could easily play more.
Brothers of Fire
Back when Abzan Aggro was a thing, the players would often play up to 26 lands, even though the curve stopped at four (in the main deck), although you could argue that [card]Den Protector[/card] was a sort of five-drop in these decks and later on, when these decks became ubiquitous, they would one up each other with [card]Wingmate Roc[/card]s.
Here’s a version by Brad Nelson from Pro Tour Dragons of Tarkir (which he went 9-1 with):
2 Warden of the First Tree
4 Fleecemane Lion
4 Rakshasa Deathdealer
4 Anafenza, the Foremost
4 Siege Rhino
3 Surrak, the Hunt Caller
3 Dromoka’s Command
2 Ultimate Price
1 Valorous Stance
3 Abzan Charm
4 Hero’s Downfall
3 Caves of Koilos
3 Llanowar Wastes
1 Mana Confluence
4 Sandsteppe Citadel
2 Temple of Malady
1 Temple of Plenty
2 Temple of Silence
2 Urborg, Tomb of Yawgmoth
4 Windswept Heath
The curve looks like this (without the removal): 2 1-drops, 8 2-drops, 4 3-drops and 7 4-drops, with the removal its 2 1-drops, 14 2-drops, 11 3-drops and 7 4-drops.
Again, it has 26 lands. You’ll often draw more than you’ll need, but at the same time you just want to play out your powerful cards early and reliably. In a three color deck, this often means that you have to play more lands (Andrew Boswell used to play a full set of [card]Mana Confluence[/card]s to make sure he always had mana of the color he needed). In the deck above, your usual first turn play is just a land that comes into play tapped (there’s nine of them), but after that, you want to use all your mana for the next turns.
There’s also no shortage of ways of using excess mana. [card]Warden of the First Tree[/card] can be levelled up, [card]Fleecemane Lion[/card] can become monstrous, [card]Rakshasa Deathdealer[/card] can be pumped and [card]Abzan Charm[/card] can draw you cards. Again, this can help you with the optimal use of your resources.
In economical terms, we are talking about opportunity costs. In most of our decisions, there’s an opportunity cost included. It basically means that if you choose a, it will limit your ability to choose b. For example, if you are studying at a university and decide to choose a certain course, it will limit your ability to take part in another course that has classes at the same time.
Building a deck is a series of similar trade-offs. You (usually) have a limit of 60 cards and whenever you add a card, that’s one of those resources spoken for. Now, there’s different kinds of opportunity costs attached to each card. Some are not that powerful, but you can play early, some are very powerful, but require more mana, which means that you can’t play it early and you’ll have to add more lands into your deck. At other times you might want to play efficient removal, which might not be able to handle all the creatures your opponents is actually playing and sometimes you have cards like [card]Drake Haven[/card] in your deck that don’t have immediate impact on the board, which means you’ll use mana or maybe a whole turn just to play it.
You have to take all this into account while designing your deck and the curve is a good tool here. By having an idea on what your curve should look like, you can take opportunity costs into account at least to a point. Of course, there’s still going to be luck involved, but that’s partly the opportunity cost. You don’t want too many high costing cards in your deck, because you might not draw anything besides those in the early game.
Again, you can’t completely take out the role of luck, but you can mitigate it with a good curve.