Friday, January 20, 2017

The Community's Predictions (So Far) for Pro Tour AER

A few days ago I challenged the community to prove how good they were at identifying broken Magic cards by predicting the metagame of PT Aether Revolt, with a free duel deck up as a prize. After nearly 300 entries, let’s take a look at the results so far.

If you would like to play the prediction game before reading this summary of responses so far, go here. Entries will be open for about another day until StarCity begins broadcasting their Standard Open.


Top 8 Predictions


Let’s start by looking at the tiebreaker questions. These are questions I added to the survey to break any ties that could occur if there was a tie in the scoring on the card-by-card predictions. Unlike the main ballot, which asked about the best constructed decks (7+ wins), the tiebreaker section sought predictions about the Top 8 decks only. Also, these questions were optional, so some respondents left these questions blank.




The community predicts a mean of 4.48 unique archetypes in the top 8, with a mode of 5. This would make for a reasonably diverse top 8, but I think the community actually isn’t optimistic enough here. In the past, we’ve seen that even PTs which inaugurated relatively stale Standard seasons (such as PT SOI) had diverse top 8s, due to the early state of the metagame, the inclusion of limited rounds confounding the constructed results, and the sheer skill of certain top players. (Seriously, how many people built that middling Seasons Past deck because Jon Finkel is a beast?) My personal over/under on number of Top 8 archetypes would have been set at 5.5.


Here things get pretty interesting. White and Blue together combine for nearly 2/3rds of the responses here. This makes sense as those colors represent a number of archetypes (Various combos, Tempo/Flash, Control) that can be expected to put players into the Top 8.

I still remember how last year, everyone was all talking about how Magic was all about creatures now and Green with all the high value creatures was just too good and Blue is always going to suck. Or how the year before that people were wondering if maybe RDW was always just going to be the right deck to bring to Pro Tours because it seemed to always feast on the inchoate PT metagames. The lesson here is to remember that WoTC likes to constantly mix things up in Standard and that the pendulum swings around by design.


Mean here is 2.67, which seems reasonable to me just based on my gut impression of past Pro Tours. More and more, we are seeing “stacked PT T8s” reminding us that Magic is at the end of the day a game of skill. Yes, you always need a bit of luck to Top 8, but over and over again we’ve seen how the consistently good players are more likely to put themselves in a position to benefit from that luck.

I made a small mistake on this question and failed to include an option for 0 first time top 8s, which likely skews the results a little. If this question is needed to break a tie, I will be giving the win to the respondent with the answer closest to the true number.

Ok, now it’s time to look at some cards. On the survey I segregated the colorless cards, the monocolored cards, and the gold cards. This was so that cards that could go into any colored deck wouldn't be compared against cards that could only go into decks of specific colors.


Colorless Cards



Walking Ballista is the favorite here, but Heart of Kiran and Aethersphere Harvester aren’t far behind.
Metallic Mimic and Inspiring Statuary come in as dark horse picks among the colorless cards.
These picks are all fairly conventional. In fact, as of this writing, the ranking of the top 4 colorless rares in the survey perfectly correspond with their ranking by market price! (As a Mythic, Heart of Kiran’s greater scarcity of course throws off this pattern.)


Gold Cards


Winding constrictor’s combo potential made it by far the community pick. Oath of Ajani is a distant second.


Monocolored Cards


And finally, we have the biggest chunk of the survey, the monocolored cards. Each submission picked 5 cards, distributed as the respondent desired among the different colors (ie, it’s a valid strategy to stack up on a single color’s cards if you anticipate that color to truly be that strong in PT AER.) There are too many cards here to list, so the table below is all the cards with >20 selections.
The Copycat hype is real. Felidar Guardian was a clear top pick here and frankly, the only top pick that is potentially truly “broken.”
The next top 3 selections were strong answers that have a very good chance of being staples in the format, but aren’t going to be format-destroying cards anytime soon. Sure, we know that Shock is a good card that will probably see quite a bit of play. But we also know it’s probably not a development mistake.
Rounding out the top 5 is another card with some brokenness potential. As both a Cryptolith Rites and a Travel Preparations on a stick, the community thinks Rishkar is a card with some upside.
There’s a natural tension to prediction games like this - to some extent, you can use the conventional wisdom as a guide to your picks. But if you hew too closely to conventional wisdom, then your submission won’t stick out from the crowd in any way and your chance of winning is lower. There is a long tail of dark horse picks among the submissions so far: many cards with 5-20 picks that I haven’t included in the chart above.


What it All Means


As a reminder, one of the objectives of this whole game is to resolve the question of whether or not the community is actually able to identify development mistakes better than the people who design the game, or if complaints about RnD’s incompetence are just hindsight bias. Here’s what I had to say in my post announcing the contest:

If I’m wrong, and there is an obvious development mistake, the community’s picks should concentrate into a few (<3) cards, and those cards should turn out to be OP. If the community’s picks are spread out among a lot of cards, and some of them do turn out to be OP - then I’m right, and there were no obvious development mistakes. If the community’s picks are concentrated into a few cards, and those cards do not turn out to be OP, then I’m still right, and there were still no obvious development mistakes, because the ones we thought were “obviously too good” turned out not to be.

I’m pretty sure I’m right, but hey—maybe you guys will prove me wrong.

Looking at the results so far, I would say the only cards the community has pegged as potential development mistakes are Felidar Guardian and Winding Constrictor. If CopyCat turns out to be a truly degenerate combo, or Winding Constrictor turns out to be oppressive, I think it would be fair to conclude the RnD missed some pretty “obvious” mistakes in this set.

It’s not too late to prove me wrong and win a Japanese Elspeth vs Kiora Duel Deck! The FREE prediction game is open until Starcity begins streaming its first Standard content, which will likely occur sometime on Saturday morning!

No comments:

Post a Comment