Evaluating cards is almost the most fundamental level of being good at Magic. "Is this card good?" If yes, you’ll think strongly about playing with it; if no, you’ll write "Snapcaster Mage, 1U" on the back with a Sharpie.

Unfortunately, a lot of people don’t bother evaluating cards for themselves. They listen to what the community has to say and believe it unthinkingly. Others do it for themselves, but go about it in the wrong way. They cling to misconceptions, exhibit confirmation bias, draw inaccurate comparisons, overrate specific interactions, and underrate others.

Two years ago, the victim of all of this was a sorcery for U. Preordain was a great card that wasn’t given the recognition it deserved until months after its release. No one was in the market for a cheap card-fixer, so when said cheap card-fixer came out, the community didn’t give it a second glance. It wasn’t until Michael Jacob made the Top 8 of Pro Tour Amsterdam with four copies in his Modern deck that people started looking more closely, and from there it went from "Standard playable, I guess" to "tournament staple" to "four-of in every blue deck" and eventually "one of the best single blue card-fixers of all time."

But these tragedies can be averted in just three simple steps. Do it for the Preordains of the world with "Primeval Titan" scribbled on the back. And do it for your own deckbuilding skills and your own understanding, rather than Brian Kibler.

Step 1: Don’t figure out why, and then decide. Decide, and then figure out why.

This statement probably took a few of you by surprise. After all, it’s the direct opposite to what we’re usually taught. We’re not supposed to jump to conclusions without prior reason (we think). "It’s the antithesis of science!" your inner eighth grade science teacher is raging. "It’s the very reason people make bad decisions based on emotion rather than cold judgment!"

Well, in some sense, he’s right. Jumping to conclusions is the reason people make bad decisions. It’s also the reason people make decisions. We can’t navigate through complexity with cold logic; it’s like trying to paint Guernica using nothing but a fine-tipped brush or tunnel out of jail with a hammer and chisel. At some point, there has to be a leap of faith.

But hey, let’s try the cold logic anyway, just for funsies. Syllogisms are great, right? If the school system in America works anything like Australia, you probably remember spending grade 10 math class copying out sentences like this:

1. All frogs are green.
2. Bertram is a frog.
3. Therefore, Bertram is green.

Indisputable! Nothing but the hard facts. Let’s try it now in a Magical context.

1. Arcane Melee is good with instants and sorceries.
2. Delver is a deck that plays lots of instants and sorceries.
3. Therefore, Delver should play Arcane Melee.

Hmmm—that didn’t work so well this time. What happened? Maybe we didn’t include enough premises. How about:

1. Arcane Melee is good with expensive instants and sorceries.
2. It’s particularly good with flashback.
3. Forbidden Alchemy and Lingering Souls are expensive instants and sorceries with flashback.
4. Therefore, Esper Control should play Arcane Melee.

This still isn’t working. Also, who calls a frog Bertram?? Seriously.

Alright, one more time. I promise we’ll really nail it.

1. Arcane Melee is good with expensive instants and sorceries.
2. It’s particularly good with flashback.
3. The mana reduction is relevant when playing a deck that wants to play several spells on the one turn.
4. Storm is an example of such a deck, particularly Modern versions with Past in Flames, Empty the Warrens, and Seething Song.
5. Desperate Ravings benefits too.
6. As does Faithless Looting. And possibly we could play Gifts Ungiven.
7. The mana cost isn’t too onerous for a deck with Rituals.
8. Therefore, Modern Storm should play Arcane Melee.

This is perhaps a little less rubbish. But it’s still rubbish. All this reasoning tells me next to nothing about the crucial question: do I actually want to play Arcane Melee? Maybe it’s good under certain conditions and in a certain metagame. Probably it’s not. There are a thousand variables, and we can’t possibly include all of them, let alone know how much to weight each individually.

What I need to do, if I want to decide one way or the other, is pick up Melee Storm and battle my way through four or five different opponents. Twentyish games against a range of different, good decks from all ends of the aggro/control/combo triangle. At the end of that, I’ll have a much better idea of what cards in the deck are good or not than several hours of reflective cogitation could give me. I’ll know whether Arcane Melee is good before I can verbalize specifically why.

Human intuition is a far more powerful tool than our reasoning skills and logical thought. I don’t know why we’re wired this way, but we are. Saying we should reason first is like arguing we should pick up heavy objects using exclusively our fingertips because they’re more sensitive. They are and that’s great, but use your legs and back to lift the darn thing and worry about feeling it later.

That brings me to:

Step 2: You’ve decided. Great. Now work out why it’s the case.

This is where you get to use your fancy logic skills. They’re still there; we’ve just put them on hold for a second.

This is how people make decisions in the real world. In textbooks, you can theorize directly to Bertram being green without being in contact with the real Bertram at any point. But that’s only in textbooks. In real life, you see Bertram and you think, "Huh! He’s green! I wonder why that is?" Your thought process might go something like this:

1. Bertram is green. (The classical "conclusion.")
2. He seems to be a frog.
3. There are a lot of frogs around here. Come to think of it, most of them are green.
4. Maybe all frogs are green.
5. That must be why Bertram is green.

Get it? Let’s try it again with Restoration Angel instead of Bertram.

1. Restoration Angel is green Restoration Angel is good in U/W Delver.
2. That seems to be because it fulfills the "Mistbind Clique in Faeries" role. It’s another thing they have to play around when you leave mana up, and if they play scared and pass the turn, you can run it out for free and gain tempo.
3. It also interacts favorably with Snapcaster Mage.
4. It allows Geist of Saint Traft to hit them through a blocker.
5. It lets you "go big" against opponents who’ve sided in Gut Shots and Whipflares to deal with Delver of Secrets and Geist of Saint Traft. It gives the deck a whole new angle of attack.
6. That must be why Restoration Angel is good in Delver.

Now, when Restoration Angel was printed, for every pundit citing a few of these reasons that it would see play in Delver, there was another saying it wouldn’t. Why? Well, for an entirely different set of reasonable-sounding reasons. It cost too much for a deck that was focused on cheap spells and minimalist land counts. It didn’t synergize with anything except Snapcaster, and that cost even more. It interfered with the deck’s intrinsic resilience to removal spells like Oblivion Ring that lost tempo on Delver or Snapcaster Mage and missed outright on Geist of Saint Traft and Invisible Stalker.

It’s easy to laugh at pundit #2, but let’s face it: your card-predicting track record isn’t much better. It’s probably worse. The fact is that none of us know before playing with them. We’re reasoning and being logical and all that good stuff, but in a game with so much complexity, you can provide reasonable arguments for whatever side you so please. At the end of the day, all these arguments are nothing better than educated guesses.

But it’s useful to develop them after the fact. It helps understanding, which is after all how we get better at this game. When an analogous situation comes along (which is all the time), you’re more prepared. Thanks to experience with previous three-cost planeswalkers, I can look at Ajani and say it’s likely to be good. I know that five loyalty is a heck of a lot to effectively start at, and I know that the creature pump is efficient enough to make a difference. Look at Champion of the Parish and how significant he’s been for Humans decks to see how much difference a +1/+1 counter per turn makes.

Because Baneslayer Angel and Demigod of Revenge both got tons of play, Thundermaw Hellkite probably will as well. Five mana is an appropriate amount to pay for a five-power flying haste creature, and killing Lingering Souls is an exciting bonus. Rancor was very good ten years ago and possibly even better now, thanks to the increased resilience of creatures—compare Strangleroot Geist and Dungrove Elder to Acridian and Simian Grunts.

Am I sure about all this? Well, not really:

Step 3: Once you reach a conclusion, assume you’re wrong until proven otherwise.

In my experience, most people are terrible at this. Once we think we’ve figured something out, we want it to be true. We want it so very badly that we ignore and brush aside conflicting evidence. We look for confirmation instead of looking for contradictions—and when you’re actively looking for something, guess what: you’re probably going to find it.

In 1966, Peter Cathcart Wason performed one of the most famous experiments in the psychology of reasoning. The Wason selection task is very simple—everyone can understand and agree with the solution—but fewer than 10% of people actually get it right when it’s first presented to them. Are you one of the 10%?

The problem is this: you are presented with four cards that are on a table in front of you. The experimenter explains to you that every card has a letter on one side and a number on the other, not necessarily in any sort of pattern. The letters could be anything and similarly with the numbers—as far as you know, there’s no restriction.

The experimenter proposes a hypothesis: "Every card with a vowel on one side has an even number on the opposite side." You are to test (confirm or disprove) this hypothesis by turning over at most two cards. The four cards have A, L, 3, and 8 on their visible faces. Which two do you turn over and why?

…

Almost everyone gets the first half right. Turn over the A card. If it has an even number on the other side, great. If it doesn’t, the hypothesis is wrong. That part is simple.

So let’s say you turn over the A and see a 2 on the other side. So far it’s all going well. What do you do next?

…

Here’s where most people get it wrong. The natural thing to do is to turn over the 8. After all, it’s an even number, and you want to see if vowels and even numbers correspond, right? Hopefully it has an E or something on the back, and then you’ll know the hypothesis is correct.

So you turn over the 8 and find an E. The experimenter looks at you expectantly, and you nod triumphantly. "It’s right!" you say. "Every vowel has an even number on the other side. I checked."

Then she reaches over and turns over the 3, revealing an O. "What about this card? It has a vowel on one side, but the other side is an odd number. Doesn’t that disprove what we said earlier?"

Uh…

I guess it does. Wait. What? How did that happen again?

It’s simple. Instead of actively trying to disprove your hypothesis, you looked for confirmatory instances. The fact that you found one meant nothing. What if that E had been a B instead? It doesn’t matter. The hypothesis still might be true, and it might not. You learned nothing from turning over that 8.

Instead, the correct solution is to turn over the 3. Now, if the reverse side is a vowel, you’ve successfully proven yourself wrong, meaning you know something you didn’t before. If the reverse side is not a vowel, you’ve proven yourself right: there is no other card that could have a vowel on one side and an odd number on the other, which is the only case in which your theory is wrong.

The lesson here is that assumptions are great, but clinging to them is not. You might play two M13 drafts and decide that Goblin Battle Jester is very good. That’s all well and good. That’s a gut reaction, and they should be encouraged. It’s not based on a ton of data, but the process of understanding is all about a series of ever-increasingly educated guesses.

You might then formulate a theory; something along the lines of, "Goblin Battle Jester is good in M13 because removal is sparse and aggression is important." Then you get more experience with the format. Perhaps you play Goblin Battle Jester in a few different color combinations and find it to be lackluster in the exalted-heavy decks. It’s also poor when you have lots of fliers or when red isn’t your main color.

The wrong thing to do here is to cling to your initial theory at all costs. "I know Goblin Battle Jester is good. These results must be abnormalities because they indicate that Goblin Battle Jester isn’t good, and that’s not the case. In fact, come to think of it, he was just fine in that exalted deck. That’s not really a point against it. So it was just those other two drafts, and I got mana screwed four games out of eight in one of them, so that doesn’t count either."

The right thing to do is refine that theory. "Goblin Battle Jester was excellent in aggressive heavy-red decks with a lot of ground creatures. That makes sense; he’s probably underrated in that archetype. Because so many red decks are aggressive, I should make efforts to pick him up early if I think my draft is drifting that way. But he doesn’t synergize particularly with exalted because often they’re chump blocking and racing. And he also isn’t great if you naturally have a lot of evasion. I guess in those decks, I’ll de-prioritize him and look to take other solid creatures like Fire Elemental instead."

So, in summary:

1. Make decisions with your gut.
2. Confirm them with your head.
3. And remember, your own ideas aren’t special. They’re only as good as you make them.

And I’ll see you at the Pro Tour!

Jeremy