Sullivan Library - The Big Lie of "Good" Cards
I don’t generally like to write "rebuttal" articles. Sometimes though I just can’t help myself. Out in the real world I’m pretty prone to swearing (though I’ve gotten better at it in recent years). But in my articles I try to maintain a sense of decorum. But reading Tom LaPille’s most recent article actually got me angry enough that I uttered a "What the f***!" out loud as I read it. Heck even in my USENET days I only managed to say the f-word once and I’d like to publicly apologize for that September 1st 1997 post. If this article comes across as especially harsh I’m sorry. It comes out of having my buttons pushed by Tom’s article and not getting my caffeine early enough in the day. I’m not trying to smack Tom down for the sake of smacking him down I just feel that he is so far off the path I don’t want him to lead anyone else down it as well.
So let’s get into it. What was it that Tom LaPille said that got my goat?
It started out with this in his discussion of Merfolk and Faerie decks:
The various tribal decks on the other hand had things like Pestermites and Mystic Tideshapers. I do understand why cards like those two find their way into Constructed but I think that even linear tribal decks can and should still do their best to emulate normal decks and play as many "good cards" as possible.
The basic thesis here is that those cards that are "good" in a vacuum are the ones that we should be paying attention to and playing with. The unspoken part of this thesis is that cards that aren’t good in a vacuum aren’t worthy of being played with another voice adding in the whisper "and you are doing something wrong if you are playing them."
This is the egregious message of the article. The basic theory of the article is wrong but the underlying message is one that can undermine the development of a deckbuilder. As authors we all have a certain degree of responsibility to our words. If we write an article about a deck that is bad that sucks to some degree but it doesn’t generally do any lasting damage. People will play it and it will get discarded once the harsh light of reality shuts it down. It’s the strength of empiricism that I’ve talked about in that the real world will provide experiences that will dull the message of a bad deck. What’s more damaging by far is bad theory. Theory can often be hard to prove but when given the tone of authority (by both the writing style of the author and the fact of its publication) it is lent a tone of credibility. It is made all the worse when it is linked to by people with authority as an article worth reading virally spreading the misinformation. Leading people down the wrong path theoretically can shape the beliefs of people who employ it for a long time until things are righted.
What is a good card? What is tech?
There can be any number of measures of what a good card is. It’s ability to give us card advantage or tempo. Its inherent effectiveness (casting cost to effect ratio). But soon we leave the vacuum and go into other spaces. It’s relative power compared to other available cards. It’s placement within a metagame (if Islands are very good perhaps so is Boil). Even the ease with which a card can be cast (GG as compared to 1G) is a non-vacuum measure – it relies on the existence of other cards within a deck to judge its worth.
This is kind of where I get to where Zac Hill is wrong. Here is how Tom characterizes Zac’s position:
Zac Hill has said over and over again that there is no such thing as a "good card" and that there are only cards that do what you need done and cards that don't do what you need done. This is a clever way of presenting the idea because it makes it hard to attack.
(Or perhaps it is hard to attack because it is mostly true?)
Where I think my friend Zac is wrong is that I think that there are good cards. But really this is just semantic. I know what Zac means (and verified it by calling him just to make sure). He means that it is more important to have a card that does what you need it to do than it is to have a card that is good. His statement was a little hyperbolic; he does believe that there are good cards and a bad cards. Knowing which is which is a part of what we all call "tech." And as Zac later retorted to Tom in the forums it is "why you playtest" – just another example of the value of empiricism.
This is a highly subjective process. Even if we all agree that Ancestral Recall is better than Concentrate it is very hard to measure just how much better one card is than another. Does the card’s instant-ness make it twice as good as a similar sorcery on that measure alone? Three times? 1.13961 times? Any answer we give is subjective. Comparing Sudden Shock to Violent Eruption can illustrate this point. Violent Eruption can do twice as much damage but it costs twice as much (or more depending on how you count color symbols). Eruption can be split up into numerous targets and be cast via madness. Sudden Shock has split second. Which one is better? By how much? We can answer the question but we’re kidding ourselves if we call an answer objective.
This is the big reason that it is important to be able to have a good if subjective read of the value of good cards. Aaron Forsythe put it well: "Tech is to be distilled in basements and stored in sun-proof bottles and traded for diamonds missiles and real estate. Tech is to be guarded for months and then unleashed upon scores of hapless players in a scourge like a biochemical bomb. If tech was free it wouldn't damn well be tech."
Tech is knowing what is good and what is bad and knowing it when other people do not.
Cards have a real value to them. But this value is not static. Icy Manipulator is a good example of this. Here was an incredible card way back in the day before tourney Magic was a big part of the equation. It took a hit for a short while and then came back to being a good card with the advent of Prison decks in 1996 or so. Then for about 10 years it was a bad card again until Eminent Domain gave it a short life again as a reasonable card. At this point it’s probably back to being a weak to middling card.
The issue here is that all of this is contextual. Cards are good in context. If you doubt this to be true play your best Block deck from one Block against a good deck from another Block. Change nothing about them. If you do this several times you’ll note that there is not a direct correlation between how the decks interact. I remember playing Cabal Rogue’s version of High Tide against a friend to demonstrate how sick High Tide was. He smashed me. Why? He was playing Brain Freeze. Context f***ing matters (oops I swore again… I’ll let Craig clean it up).
Context and Synergy and "Having the guts"
Context doesn’t just matter with regards to one deck and another deck and the environment swirling around them that we commonly call the metagame. Context also matters in the cards within a deck. We can call it synergy if we want. It doesn’t matter. What matters is that we are having two cards interact with each other.
Take a bad interaction. I’m sure that you’ve had this happen when you were building decks. I remember building a deck running both Bushi Tenderfoot and Samurai of the Pale Curtain at one point not realizing that if both were out the Samurai wouldn’t let the Tenderfoot flip into Kenzo the Hardhearted like I wanted. It’s a cute example and I’m sure you can come up with your own if you’ve spent any amount of time working on decks.
Even good decks sometimes have this "non-synergy." Every deck with Wrath of God and a reasonable amount of creatures has it. Every time that you play Pernicious Deed with cards that can be blown up by it you’re looking at the possibility of a negative situation should the right context come up. It’s part of the reason that Sakura-Tribe Elder has higher value in a deck with Deed and part of the reason that a weaker card like Rampant Growth (just as an example) might make the cut in such a deck. It might be weaker but you’re looking at higher synergies.
Way back in the day I was hanging out after the 2003 Wisconsin State Championship with my friend Adam Kugler. He had largely designed the version of Blue-based Affinity that most of the Madison guys had taken to States that year. He played it himself. But he had this other Affinity deck that he thought was better.
"I just don’t have the guts" he said.
It was a deck with "bad" Constructed cards like Disciple of the Vault and it ran Ornithopters not as a means to draw cards but as a way to attack pumped up by Bonesplitter and the like. He kept smashing most of the decks that we had tossed at it. He didn’t play it because it ran "bad" cards.
This proto-modern Affinity deck didn’t have Ravager or Skullclamp. It was still pretty damn good. And it ran a bunch of garbage. Or so it seemed at the time. These days Affinity runs all kinds of garbage. But who cares if you can win with it? Careful Study and Breakthrough are both "bad" cards. But again who cares? If you’re playing Dredge they do what you need them to do and they do it well.
Take this deck:
Aaron Snover used it to win the Detroit PTQ just a couple of weeks ago.
I hope you forgive me for this but I’m going to go out on a limb and call Spark Elemental Lava Spike and Jinxed Choker "bad" cards. I’m going to call Keldon Marauders Chain of Plasma and Flames of the Blood Hand "marginal." I’m probably a bit more forgiving than most. This puts us at nineteen cards that aren’t "good."
I have a feeling that if Snover had listened really listened to what Tom was saying if he’d tried to make this deck have more "good" cards he would have ended up with a really bad deck.
Take Tarmogoyf. Good ol’ Tarmogoyf. Remember when Magic Storyteller (hear hear!) Evan Erwin asked people at Valencia about him? Well Zac naysayed Tarmogoyf as the best creature ever which is fine but a huge percentage of people did consider him that.
So why isn’t four Tarmogoyf the rule for say Spirit Stompy? As I wrote in my article on the deck it simply isn’t big enough regularly enough to support it. Three became the de facto number in most versions. Simply knee-jerking four Tarmogoyf is wrong. Who cares if it is the best creature? Does that matter when you are balancing it all out? What matters is how the deck empirically performs. Kowtowing to the pressure of maximizing the numbers of your "best" card in the face of empirical evidence pointing to other conclusions is just cowardice. Who cares if people snicker at your "bad" three-count? For that matter who cares about your "bad" card choice. Is it the right card? Then have the guts to run it.
What about in Standard? If synergy is something not worth valuing why did Ryohei Masuno not choose to run Tarmogoyf in his Elf deck for Shizuoka? Taischi Fujimoto did run four in his Elf deck but Ryohei Masuno’s zero is an important number to take note of.
There is a tension between pure power (power in a vacuum) and synergistic power. Sometimes it can be hard to discern where the line is. When preparing for Pro Tour: Chicago a million years ago with Zvi Mowshowitz our group Cobra Kai’s testing for the event had come to the conclusion that River Boa at that time the best two-drop creature in history was not worth it. Its pure power did not match the needs of the deck. Synergistic power won over.
And so I come to Tom on Zvi’s Faerie deck.
Choosing Sides Between Tom and Zvi
Tom writes this bit of commentary coming just shy of getting it:
[…]it's actually a rather beautiful deck in that it plays a lot of strange-looking cards that happen to work very well together. The Cloud Sprites and Nightshade Stingers ensure that Spellstutter Sprite is able to counter things on curve much of the time and give Mistbind Clique small faeries that can be championed painlessly. They also turn Familiar's Ruse into an extremely efficient counterspell that does an awfully good job of emulating Counterspell itself when the only extra cost is replaying a random one-drop and they are reasonable offensive threats on their own with Scion of Oona backup. So what's the problem?
He’s almost there. Look at that analysis! It’s spot on. The one-drops do do all of the stuff that he is ascribing to them. They do a job and they do it well. So yes what is the problem?
The problem is obviously that you are actually playing Cloud Sprites and Nightshade Stingers in Constructed. These are not "good cards" in the abstract by any stretch of the imagination. When you develop with Nightshade Stinger into Spellstutter Sprite into Scion into Familiar's Ruse you are going to look like a genius.
The emphasis above is mine.
It almost sounds as though Tom would be embarrassed by the little Sprites. They aren’t good! Good lord what will people think! Oh but when it works out you sure look great but the other times you’ll look like a fool.
Last week I recounted my loss to Alex Borteh’s Fish deck. The scary moment of that match was turn 1 game 1 "Merfolk of the Pearl Trident." Merfolk of the Pearl Trident is a lot worse than any of the one-drops being considered today for decks. But I knew. I was going to lose.
His deck was "a gooder" to some (the sarcastic disparaging tone very evident in their voice). How could a non-reanimation deck running Vodalian Merchant be any good? A deck with Darting Merfolk? Those are bad cards! Alex didn’t care though. The synergies were enough. He didn’t care if some people thought the deck looked bad. What mattered were the numbers. The winning.
So it is with Zvi. Do you think Zvi gives a damn what people think about the look of his deck? If you have any sense of who the man is you’ll let loose a resounding "No." Do you think Zvi gives a damn about winning? Again if you have any sense of the man you’ll be saying loudly "YES."
We all miss the mark from time to time but Zvi’s list wasn’t an example of him missing the mark because he was running cards that were bad in a vacuum like his one-drop Sprites. He chose those cards because they were the ones that were needed to get the job done.
Tom’s criticism of the deck that I highlighted above is both harsh and shallow. He goes on to compare it to the "improved" faerie deck of Yuuta Takahashi saying this:
The upgrades from the previous model are so massive that it almost doesn't look like the same deck […] Zooming out what has changed? The simple answer is that the deck plays cards that are good in the abstract instead of cards that are bad in the abstract.
This is massively unfair. Zvi worked with the tools that were at his disposal. Bitterblossom wasn’t even available yet. I’m sure that he would have dropped all of the one-drops (or nearly all of them) if it had been. I can’t be 100% certain but I’m pretty close to that level of certainty. It would have been fair to term the new deck an "update." Instead it sets up Zvi’s deck as a straw man.
The Takeaway – Correlation and Causation
The other day Paul Krugman said on the Colbert Report that "The truth is complex." And it’s true. The truth is complex. Simply looking at something like "play good cards and you win" is a grotesque simplification of how Magic works. Subscribing to this philosophy will lead you away from wins.
The big problem is that it will limit the possible cards that you will consider for a deck. Does it really matter if a card is strange looking if it does the exact job that you need it to do? Synergy can be abundantly obvious sometimes. We can see how Affinity or Goblins cards are jam-packed with synergy and how this increases their value by virtue of their interaction with the other cards in the deck. It can also be less obvious. Fixating on whether a card is "good" enough in a vacuum will inevitably lead you away from decisions that could have been the right decision.
When people are trying to sell you a one-shot fix-all solution in Magic be wary. Card advantage is not the end-all be-all. You can have lots of card advantage and still lose. It is the same with tempo. And it is the same with "good cards." All of these things are tools that we should employ. We should cultivate the recognition of what is and what is not a good card but we shouldn’t be a slave to only running those cards that we’ve deemed or others have deemed are "good." More goodness doesn’t equal more winning any more than more card advantage or tempo does. Do not confuse correlation with causation.
Until next week!