The Dragons of Tarkir Spoiler analysis is up. In the article, I go into depth about all kinds of numbers in the format, including the costs and size of creatures. From there, I draw some conclusions about how the format will play out.
In the article, I mentioned that I would go into more detail about my reflection on Fate Reforged, specifically in regard to whether or not the format got faster. I had predicted that the format would be faster, and when I went to do a numerical analysis of the games, I found that that prediction had borne itself out. MTG Goldfish had an even larger sample of games, and also found a slight increase in the speed of the format. However, several pros then began to say that they felt the format had gotten significantly slower in Fate Reforged. This gave me pause. I wondered whether there was some kind of error in the numbers, or if these players were seeing something that I simply hadn’t seen. So I thought through the issue in some detail, and this is what I eventually settled on.
First, I’m generally inclined to believe in large samples over the personal opinions of popular pros. Magic is an incredibly complex game, and pro players have to make evaluations based on their own limited experiences of a draft format. These experiences can be skewed for all kinds of different reasons, but the important thing to remember is that any individual player has such a small sample size of experience that it’s not wise to just take everything that any given pro says without checking it against evidence.
Second, I suspect that there is a difference is definitions here. When I say that a format has gotten faster, I mean that the spread of ending turns has shifted to the right. I am not entirely sure what these pro players mean when they say that the format has gotten slower, but I suspect that a lot of it has to do with their own preferences for archetypes in the format and their beliefs about the relative strengths of different archetypes. What I find fascinating is that although I said the format was faster and other pros said the format was slower, we would often come to the same conclusion about which archetypes were the strongest in the format. While I saw the distribution of turns shift to the right, I also recommend that players react to this by drafting the stronger but less common control decks that would be able to establish control in the early game.
Third, a lot of pros have the problem of getting into echo chambers. When pro players test for a Pro Tour, they usually get a bunch of players in one place and grind out several drafts. These are small sample sizes, but it is a group of players with relatively similar levels of play skill that are trying to figure out the format. This method has strengths and weaknesses, and one of the major weaknesses is that they have inbred metagames. I remember listening to one major pro (I can’t remember who it was specifically) talking about their testing and mentioning that several players on their team were forcing the five color strategy in FKK from the first pick. The problem is that that deck was not drafted with nearly the same amount of popularity in the wild. This skews the metagame towards particular decks. If these pros tended to prefer the slower decks, they were going to gravitate towards playing them, which often leads players to build their decks in order to beat those slow strategies.
Fourth, pro players have very effective voices in the community. When a popular pro with a wide platform espouse a particular theory, like LSV or Paul Cheon for example, those pros are going to have a wide effect on the metagame. If these players say that the format has gotten slower, lots of players are going to listen to them, and the metagame is gradually going to change to reflect those opinions. Because of this, widely popular pros can often shift the metagame regardless of whether or not they are right. This is a fascinating phenomenon, but it happens in Magic all the time. When several popular pros started saying that the format was very slow, players quickly started to draft with an eye towards a slower metagame. Soon, there were definitely a higher percentage of slow decks than there were in the format before the statements.
Honestly, I can’t be entirely sure whether the format ended up slower or faster over the duration of its run. These different factors all contribute to obscure the issue, and I don’t think that there is a good consensus on whether the format got faster or slower. The only thing I can say for sure is that my data showed a slight increase in speed, so I’m going to base my reflection on that data, but it’s definitely possible that the format actually got slower despite some evidence to the contrary.
With all of that said, make sure to go check out the Dragons of Tarkir spoiler analysis. There’s a lot of great data there, and I’m excited for people to read it and then come back with comments for discussion.