Information Processing In Magic

Ari Lax frames his article around the best five decks he’s ever played competitively and why they were good. Can you read your opponent’s hand without looking?


In my mind, the point of theory articles isn’t to be a one-size-fits-all answer to how something works. The point is that someone could potentially read it and find a new pattern of thinking that to them makes sense and allows them to make solid decisions and have a plan.

I wanted to end this article on a list of the best decks I’ve ever played, but it turned into something more. Let’s start with the aforementioned list:

5. Dredge (Lorwyn Extended)

4. Elves (Shards of Alara Extended)

3. ANT (Rise of the Eldrazi Legacy)

2. Twinblade (New Phyrexia Standard)

1. Faeries (Shards of Alara Standard and Scars of Mirrodin Extended)

These are the decks that made winning easy, where I would sit down and wonder what my opponent was doing playing any other deck. The common factor? These were the best decks at generating information and having relevant decisions.

A while back, Wizards ran a poll on this subject. The scenario they posed was two players playing a sixty-card mirror match. Assume that on all decisions, they make the optimal play given what they know. One player is playing normal Magic; the other, unknown to the other, has perfect information. The question asked: What percentage of games will each player win? Obviously this is just the opinion of a mass of people, but the results were still staggeringly in the favor of the player in the know. The main argument people used to argue the odds back down was that, assuming the other player was perfect, much of the advantage could be negated by how close they could get to full information just by observing the game.

The concept of hidden information is just as fundamental of a rule as drawing one card per turn or playing one land a turn, but people often don’t refer to it as such. When you make a decision about what card to play or how to attack, you logically base it off what you know. Since the way to win at Magic is to figure out how to bend and break the rules, how can you do this? There are all sorts of articles about how to figure out things you shouldn’t know logically based on the flow of the game or through things like reading people, but quite frankly half of that stuff is not going to be a big help. I’m going to be brief on this here, as I want to talk about it more later, but simply put, most of these things result in decision trees that are just too complicated.

So, within the context of “normal” Magic, how can we gain an edge in the war of information? Obviously sometimes you have the Peek because “You just gotta know,” but how does information in general apply to decks?

Proactive Information:

Quick. You are playing a random beatdown deck against a control deck and have to make a decision on turn four as to what you want to play. What are the two main decision factors? One is going to be damage output, or the clock you present. The other is going to be playing around your opponent’s ability to return the game to a stable state.

You see this most often with decks people traditionally label as aggressive, but when you are being proactive, you gain an information edge by minimizing the number of relevant cards your opponent can present. Instead of having to focus on the value of a ton of different cards like what happens in a control mirror sometimes and altering valuations based on continuously changing board states, you make things simple. You know the format; they can have card A, B, and C. Nothing else matters.

As a result of this, you get to make much stronger educated guesses towards your decisions. This aspect gives beatdown decks the traditional stigma of being hee-haw decks because the game’s lines are much simpler. Either one action happens, or you win with a few situations like having to decide what resource you want to fight. There is one easy defining moment that decides each game as opposed to always having to figure out which decision you made over ten turns that led to the way the game played out. When you make a misplay, it’s more of an obvious bumble causing missed damage. The thing people don’t realize is this also leads to aggressive decks being some of the most reliant on high-level skills like derived information. All of your decisions have to be made in advance of the actual reveal of what your opponent is going to do, forcing you to lean on indirect information to make them.

Indirect information, as I said, is HARD. How do aggressive decks avoid having to derive this at every point in time? Basically, how do they minimize the number of potential interactions the opponent has that they have to think about? There are a few different tropes that pop up from format to format. For example, if you play Red, you primarily lean on the fact burn spells allow little interaction. The game progresses much faster to a point where your opponent runs out of actual options that matter. In the current Standard, you see Tempered Steel leaning on the fact its namesake blanks every reactive spell that doesn’t interact with the enchantment, and Tokens plays a lot of single-card threats so that it can keep representing a board state where its opponent needs a specific answer to win.

Reactive Information:

Now, imagine you are on the control side of things. What are you doing? Trying to figure out how to play your answers to their threats?

How do you make this decision? You don’t go into a turn thinking, “I’m just going to Doom Blade whatever they play.” No, you sit there, wait for them to play it, then make a decision after you know what happens. From the proactive deck’s viewpoint, this is like those playtest games where you walk into an obvious blowout and say, “Wait, I guess I should have done this; let’s just go back.”

Being the reactive deck inherently provides an information advantage. Your opponent plays a spell, then you choose what you want to do. There is some assumed derived information about what answer lines up best with each threat and what matters, but in general these are decisions that are less contextual and more absolute. You can go into a game against U/W Humans knowing “If I have Mana Leak and Doom Blade, I want to Doom Blade a Hero of Bladehold so I can Leak a Mirran Crusader”; less so “My opponent has played this three spells so far. I have this much power on board, and their life total is another number; do I play another guy or can I race him with three cards having these two removal spells if I don’t?”

Information that Matters:

You are playing Goblin Charbelcher combo. You Gitaxian Probe your opponent and see that they have a Force of Will. Now what?

If you want to get into semantics, Belcher does have Empty the Warrens based plays that can circumvent a Force, but most of the time, knowing whether they do or don’t have it doesn’t change anything. You are going to cast your spells in the same order as if you didn’t know and hope it is enough.

This is obviously an extreme example, but it proves a point. You can only capitalize on information if you can alter your decisions because of it.

On the proactive side, you see this a ton. Yes, you could not play that creature so that their Day of Judgment doesn’t kill six of your guys, but is that one guy going to make the difference if they do have it? You aren’t going to beat the sweeper anyway; does it matter if you know about it or not? This also gives the proactive decks a potential edge on the reactive decks. By choosing your actions to force your opponent’s responses, you effectively remove the advantage they had by knowing what you are doing before acting. When I talked about the weaknesses of Standard U/B Control a few weeks back, this is what I was talking about. If you just forced their actions early, their deck was no longer consistent, as they didn’t have the option of timing any card selection into their curve. This is also a huge part of why the old school instant draw spells were so powerful (i.e. Fact or Fiction and Thirst for Knowledge). Your opponent had to force your actions every single turn; otherwise you just spent the one correct end step reversing all of their work.

On the reactive side, this means a lot more for deckbuilding. Remember Caw-Blade? Part of the reason it was so powerful was that you could pass the turn representing one of a million different things. That said, even if you have a million options, you need a default profitable option should nothing occur. This is the Limited scenario where you pass the turn with Rebuke, and your opponent just does something else. If you are just going to pass with one reactive option up and nothing to do if your opponent doesn’t do anything, you need to be making the decision that if they have nothing, your ideal interaction or close to it would also be doing nothing (i.e. just getting another land drop to hit a Titan). If that isn’t the case, you need a secondary option that is profitable, like a Think Twice. If you are trying to be reactive, one of the most important things is having options that matter so that you can make decisions that also matter.

Processing Power:

So, why is it that indirect information is such a bother, or weighing irrelevant information even an issue? Why don’t you want to consider everything at once in case you incorrectly determined something to be irrelevant?

You don’t have unlimited processing power. No one does. There is a cap to how much information you can actively process in a given time span. You cannot consciously analyze every aspect of a game of Magic and expect to come up with the optimal line of play. When you start trying to figure out whether they have a card you can’t play around regardless, your attention goes to there instead of another relevant decision. Enough distractions and you start missing simple things like on-board kills. I’m not even talking about stamina over the course of the day; I’m talking about over the course of a turn or two where multiple actions are performed.

So, how do the masters do it? How do they find these corner-case plays or see these small lines or tells that provide more information or deny it to their opponents? I hinted at this at the very start: Most of the mid- and low-level processes are automated or completed via shortcuts. They have been through the scenario before; they already know the answer, and they execute their play without allocating mental energy to it.

When possible, your goal should be to establish these default responses to situations in order to maximize your ability to concentrate on new situations. There are times you can get stuck in convention, but more often than not, you should be going with a set line unless something sets off a red flag in your mind instead of piecing things together as you go. Yes, there will be a time where whether they have that card or not decides the game and then you can sit there and try to figure it out, but those moments are what you are banking your thought time for.

Most importantly, you should be generating plans and sticking to those. If you are stuck, play for board position, card advantage, life total, or just any one specific thing. Contrary decisions are the quickest way to be incorrect, as even if your decisions are right at the time, you can negate gained advantages if you fail to continue capitalizing on them. If you shortcut all your decisions to something like “I want to attack Ramp’s threats and ignore mana advantages,” you set a subset of cards that matters and eliminate a ton of card valuation legwork you would find yourself bogged down in through the midgame. Commit to a plan, and if you find it doesn’t work over several games, try something new. Finding good shortcuts involves time, but you win because of them.

So, going back to the start, why was each of those five decks so insane?

5. Dredge

This was the perfect example of an optimized proactive deck. Your opponent only had a couple of cards that were even close to interacting with you, and because of Cabal Therapy, you could just eliminate those. Transplants to various other formats (notably Legacy) failed, as opposing decks suddenly had options to interact before you got Cabal Therapy online. You went from a deck with only a few cards that mattered to a whole slew of interaction possible, leaving you in scenarios where you would have to choose between playing cautiously and possibly losing to their clock or going for it and dying to a counter.

4. Elves

On top of having proactive draws where you could just force them to have the one card that mattered or they lost, Elves was the best deck I have ever played at forcing the opponent’s plays. The ability to just Glimpse kill someone from basically nowhere meant they had to keep up interaction at all times, even if you were just beating them down with 1/1s. You entered a game state where they had to do nothing to not die, but you got to do minor things like attack for 2-3 or play another 1/1. You also had Wirewood Symbiote that forced their hand on removal and Mirror Entity to force them to have answers not just to Glimpse of Nature but to end-step Chord of Callings.

3. ANT

Again, this is another extreme of minimizing the number of cards you have to think about. You also have relevant things to do at almost every point with the information Duress and Thoughtseize give you because of how good cantrips are at sculpting your long game. Beyond the basic consistent functionality of the deck, this last part is what sold me on this deck and is something rarely found in proactive decks.

As an aside, this deck also lets you do a whole ton of shortcutting on simple things. Even just going through the combo with Dark Ritual as a +2 instead of thinking you are spending B and gaining BBB eliminates an operation and makes things that much easier. Ill-Gotten Gains isn’t this weird counting game. It’s 12 mana total (including rebought Rituals) if Infernal Tutor is the fifth spell of the turn. Add accordingly for other things like a Daze or needing an additional Tutor-Tutor cycle to Storm up.

2. Twinblade

When Twinblade was legal earlier this year, there was for all intents and purposes one matchup that mattered: Caw-Blade. We actually first turned to Twinblade for the Invitational because of how miserably complicated the Caw-Blade mirror had become. Before New Phyrexia, you basically had two game states: before you had the ability to establish a Planeswalker, you tried to get a Sword or Hawk advantage, and once you could establish a Walker, you did that. There were a lot of angles to figure out because of all the mana allocation options every game, but you were rewarded for having a plan.

Suddenly, because of Batterskull, everything changed. Just putting a 4/4 into play provided an almost insurmountable advantage, and the mirror turned from executing a plan to trying to mise with situational answers. Gitaxian Probe was suddenly an all-star because otherwise you had to essentially randomly play around a handful of singletons that could each wreck your day. I’m sure there were things we could have learned to gain edges, but quite honestly we didn’t have time to come up with those.

Instead of fighting that fight, we just moved to Deceiver ExarchSplinter Twin as the mirror trump, and it was savage. Talk about forcing their moves. Again, like Elves, your opponent always had to respect the fact you could just randomly kill them at any point. Unlike Elves, instead of just letting them sit there afraid, you forced them to burn their potential interaction with Jace and Stoneforge. If they happened to not stop one of those, you just killed them, and if they fought you there, you would eventually run them out of cards that mattered and attack with infinite 1/4s. Instead of worrying about marginal interactions with situational cards, you could just keep presenting threats in increasing power until they ran out of actions. The combo, despite being reasonably easy to interact with, essentially made fewer cards relevant by eliminating the need to be afraid of them generating board presence. Winning the game and even just playing your turns became much easier, and that was why this deck was so much better than the alternatives.

1. Faeries

There are just so many reasons here that show why Faeries was such a dominant deck. You had the reactive edge on all of your spells, so that when you didn’t need to one-for-one them, you just played a threat and started killing them or sat there content with letting Bitterblossom do all the work. You had the nut draw that forced them to deal with a Mistbind Clique immediately or die. You could at any point not only shift what you were playing but how long you planned on letting the game continue, giving you relevant reactions to almost anything thrown at you. Finally, the concept of hidden information barely applied. At any point you were one Thoughtseize or Vendilion Clique away from taking a peek and just knowing everything. The level of freedom you had to make decisions and the amount of information you had to do so were beyond that of any deck I had ever played before.

Not every format is going to have a deck that functions as smoothly as these decks did. In fact, few do. Instead there will be bits and pieces you will pick up on, and those should be the factors cluing you in to what the best deck is. Have a plan, block out the noise, figure out what matters, and spend your time tanking on that.