fbpx

Rule of Law – Defending No Bannings

The DCI decided not to ban any cards in Standard, despite rumors and pressure from many in the community. Regarding the Delver deck—which was the focus of nearly all the rumors and pressure—the DCI wrote:

“The DCI looked at the results of competitive Standard events. We found that while a high percentage of the participants played WU Delver decks, that the win rate of those decks was very close to par. For instance, in a recent MTGO PTQ, the win rate of WU Delver decks against non-Delver decks was a bit under 51%. In general there are decks that the Delver deck is strong against, and decks that it is weak against, but on average the deck tends to get results close to average. Additionally the number of people playing high level Standard events is the highest ever.”

I love the fact that empirical evidence was cited in support of the decision. Agree or disagree with the decision, we have been given some concrete tools with which to discuss the problem. Does playing Delver give someone an automatic advantage over those who don’t? Well, not in practice (or at least not by much). Does Delver take a lot of the fun out of Standard? Well, not if you believe people vote with their feet and have been showing up in record numbers.

That’s a pretty good start to the explanation. With limited space, The DCI conveyed a lot of the reasoning behind the decision to leave Standard alone. Some arguments they didn’t have time to articulate, other arguments were perhaps implied in what they did say. I’d like to pick up where they left off today and provide some additional explanation (and some counter-arguments where appropriate).

Publication Bias

i.e. What About the Decks that Don’t Get Played—What’s Their Win Rate?

I (and others) have often been troubled by “metadata” that combines the results of many published research papers into an easier to digest “big picture” of a given topic. A good scientist is very careful about drawing broad conclusions from metadata, but not every reader is. One main problem with published research is often called the “file drawer problem,” or, more generally, “publication bias.” Studies that don’t achieve a significant result or aren’t interesting to publishers for any of a number of reasons don’t see the light of day. They are condemned to remain in the researcher’s “file drawer.”

Similarly, decks that beat Delver only 25% of the time mostly get left at home—not registered in big Standard tournaments. Only the underprepared or masochistic Standard player would even consider playing such a deck. Given this state of affairs, is it entirely fair to cite UW Delver’s 51% win rate against non-Delver decks as a sign of the health of the format or the fairness of that deck? I’d say no, it isn’t fair.

The aspect of it that is fair is that by looking at the statistic we make sure we aren’t in a world in which people know they need to beat UW Delver and just can’t do it. I suspect that this was the case during the [card]Jace, The Mind Sculptor[/card] + [card]Stoneforge Mystic[/card] Caw-Blade days of Standard. So this snapshot of the current metagame is a legitimate vital sign, in that narrow way.

The unfair part is that we still don’t know just how good UW Delver is against the deck you want to play, or the deck you played a month ago, but gave up on after UW Delver kept beating you. Those decks are mostly excluded from the 51%-win-rate sample. In addition, it is possible that they looked at too small a range of data, or didn’t weight the data correctly—but I’ll give them the benefit of the doubt since I would guess the figure isn’t too far off after these minor corrections.

So why do I still think that the DCI got it right? Well, since we can’t hope to objectively get a view of the decks people left at home (we could do a survey or look at message boards, but what we find would have plenty of its own biases), we need to come up with another measure. What we’re measuring is how much UW Delver is warping the metagame. I think a fair way to measure this is to look at diversity among the non-Delver decks.

Even if Delver only wins 50% of the time in the format, is every successful non-Delver deck an anti-Delver deck? If so, that reveals that many decks are being forced into the file drawer, and not getting a chance to shine due to Delver. If not, then the diversity of other viable decks means different styles, colors, and speeds of deck are represented, and fewer decks are hopeless to play. As it turns out, we see Zombies, Wolf Run, Pod, UW Control, Solar Flare, and other decks making Top 8s. That’s a pretty diverse group of decks, both in number and style, which combine to win around 50% of the time against UW Delver.

Maybe It’s Fair, but Is It Fun?

The fundamental problem with looking for answers to “is it fun?” among a chorus of individual opinions is that some opinions tend to be expressed more often and more forcefully than others, by the very nature of the opinion. If we ban a card whenever people complain about it, nothing powerful or popular will ever survive. We need a more objective measure of whether people find Standard fun. The DCI has chosen to look at attendance figures for large Standard events.

Of course, this isn’t a perfect measure. That Standard with Delver is popular does not preclude the possibility that Standard without Delver would be more popular. To illustrate that more specifically, people might really like playing with [card]Huntmaster of the Fells[/card], [card]Geist of Saint Traft[/card], [card]Birthing Pod[/card], and [card]Restoration Angel[/card], whether or not [card]Ponder[/card] and [card]Delver of Secrets[/card] and [card]Snapcaster Mage[/card] are legal, and in fact they may even like it more without Ponder or Delver or Snapcaster.

Another important limitation to this type of analysis is that it can justify almost any aspect of the status quo when attendance is good. In the same way, increasing Magic sales have been used to justify the Titans cycle, investment in Duels of the Planeswalkers, planeswalkers appearing in nearly every set, and many other aspects of the status quo. Unless you believe the status quo is perfect, you should be troubled by this pattern.

You see the same thing happen with the National Football League (bear with me international or non-sports-inclined readers). How can they change the rules to make the game safer when the game is so incredibly successful with the rules it has in place? How can they change the way the draft works to disincentivize tanking among the bad teams when it doesn’t seem to hurt TV ratings all that much? Once you justify one aspect of the status quo, you’ve justified them all.

Again, why am I OK with what the DCI did and said given these problems? Well, tournament attendance did go down when Affinity was the best deck in Standard by a wide margin. People do vote with their feet, so to speak, and refuse to participate in formats that really aren’t a whole lot of fun, and the Affinity era (and perhaps more recently the Caw-Blade era) provides evidence for a connection between health of the format and attendance.

We can’t rule out that attendance would be even higher if Delver was gone, but if The DCI is happy with current attendance, there is some wisdom in not rocking the boat. Banning comes at a cost. Confidence in your future products at their release goes down (“I wonder which cards in this set they’ll have to ban”) and you risk a similar format, or a worse format, emerging post-bannings.

That’s really a point that should be emphasized: The DCI should err on the side of not banning anything. It seems to be common sense, but how common is it given what you hear in the community?

A New Set and a Rotation Are Coming Soon

The DCI also stated in its announcement:

“Looking at the Magic 2013 card set, it appears that there may be more tools for other decks than for the UW Delver deck, though time will tell if this bears out. The DCI will continue to observe how this plays out, but is taking no action.”

This is really just more weight for the “err on the side of caution” argument. If things might correct themselves organically soon, or shift to a new menace as the case may be, letting that happen is preferable to banning something now.

The counter-argument here is perhaps that only the DCI’s unreasonable delay in taking action explains why it is now late to the party, and it should still do the right thing and save the few months it can. Perhaps, but I’d be more inclined to criticize and move on in a few months than have short-term bannings that aren’t necessary.

So Is There Nothing to be Concerned About?

Although I think the DCI made the right call this week, it doesn’t mean I’m not troubled by Delver and other recent trends in the development of the game. I’ve been a vocal opponent of hexproof, much of Avacyn Restored, and other designs and implementations I find sloppy and harmful. Designing and developing Magic cards is a tough job, and I think the current teams do a lot of things right and a lot of things wrong.

Duels of the Planeswalkers and more flavorful design (starting with M10) were great steps in the right direction, so great in fact that they are masking many things that aren’t as positive. If you own a barbeque restaurant, and your ribs are so good that you’re making money hand over fist, it doesn’t mean you can’t improve your coleslaw. If someone says your coleslaw sucks, showing them your business’ healthy balance sheet won’t prove much about the coleslaw.

Zac Hill wrote in his [card]Cavern of Souls[/card] preview article that he wasn’t “…going to sit here and look you in the eye and tell you that [card]Snapcaster Mage[/card] is a fair Magic card. Nor am I going to blame any of my colleagues for the problem: I worked with Tiago Chan to design it, and by the time we realized exactly how powerful it was in concert with the abundance of one-mana cantrips in Standard, the card was already out the door.”

A fair question is “how the hell does this happen?” Several cards that are powerful at nearly everyone’s first glance, like [card]Snapcaster Mage[/card] and [card]Primeval Titan[/card], end up being too powerful given how they interact with at least somewhat obvious other elements of the Standard metagame (like [card valakut, the molten pinnacle]Valakut[/card], [card]Mana Leak[/card], [card]Ponder[/card]). [card]Aether Vial[/card] is way better than it looks. [card]Umezawa’s Jitte[/card] too. To be fair, [card]Ponder[/card] is too. Is [card]Snapcaster Mage[/card]? Is [card]Primeval Titan[/card]? Bannings shouldn’t fix these problems, but I’d hoped we’d have fewer of these problems to fix in the first place. Get it together guys, and stop using the health of the balance sheet to explain to us how great your crappy coleslaw is.

-Matt Sperling

mtg_law_etc on Twitter

Discussion

Scroll to Top