Frames (strategic advantages in slanted information states, pt. 2)


A few weeks ago, I looked at the continued effectiveness of the play-action pass in football, how a defender’s incredibly low threshold for deception allows for the play to succeed consistently even as teams put fewer and fewer resources into running the ball. I want to follow up briefly on the framing problem that underlies the ontological tension between pro- and anti-analytics people in football. Ultimately, the groups are often talking about two very different historical experiences of the game at which they both look. For anti-analytics people who were habituated into the game at its lower levels, the run game is the most important part of football. That’s indisputable: it’s what’s coached, it’s what’s called–the entire game is based around running well and stopping the run. And given that success in that part of the game is most correlated with success at all but the highest levels (top-tier college and pro ball), it makes perfect sense why the run game would be the basis of everything even if it isn’t the most analytically efficient. So when those who have been habituated from the bottom up, when they talk about football’, they’re talking about a historical football, a game circumscribed by one’s early and foundational experiences with the game itself.

When analytics folks talk about the game, they’re talking about the game at the professional level, where the competition is so high that every and all edge needs to be explored for potential exploitation. So, from their perspective, if the data bares out that you should deploy X or Y strategy with more regularity because it’s an underutilized efficiency, there’s no tension between ontologies, because their ontology is circumscribed by the highest levels of competition only. And it’s easy to see why pro-analytics people would feel a kind of smugness when debating those who are against analytics: the former are able to be flexible in their understanding of and approach to the game state, because the football’ that they know is one that isn’t embodied and temporal.

Ultimately, I think this debate is a good example of embodied knowledge vs. theoretical knowledge. For those whom football’ refers both to the breadth of their experiences playing and watching the sport–the environment in which they were habituated–as it does to the sport itself, the stakes are high for them as far the discourse entailed by the growing prominence of analytics is concerned. In other words, it is literally more than just a game—it’s a perceived attack on their experience of the world as it is composed.