clock menu more-arrow no yes mobile

Filed under:

On the metaphysical nature of baseball statistics

As the Tigers enter a new age of analytics, what are the implications for us as fans?

Screenshot from Fangraphs
Screenshot from Fangraphs

The age of analytics is upon us. Here at Bless You Boys, we have not been strangers to these measures; and now, finally, the Detroit Tigers' front office seems to be joining the party fashionably late. The question of whether the Tigers should be concerned with analytics, is moot and obvious -- of course they should.

As a multi-million dollar business, it is their responsibility to do everything they can to improve and streamline their product and to do so within the realm of efficiency. However, as fans, what does our desire to quantify and analyze the game and the players tell us about ourselves and is it on the whole a good thing?

Statistics have always been a part of baseball and there are too many iconic numbers to name. Most baseball fans know what statistic 56 represents and ditto 61, 715 and the importance of .400. Numbers are an integral part of the game, woven into the fabric of baseball.

It is clear that these new measures allows us to know more about the game than we did before. Michael Lewis' Moneyball famously documented how using sabermetrics can show us, for example, that on-base percentage is a better statistic than batting average. That revelation is now accepted dogma, and the statistics created and deployed now far outstrip such simple insights. wOBA for example, creates weights for each offensive outcome and scales it relative to league OBP. Similarly, wRC+ does that and adjusts these outcomes for park factors and league context. On the pitching side, FIP tries to provide a more "accurate" stand in for ERA by looking at what pitchers can control, namely strikeouts, walks, and home runs. xFIP uses steady HR/FB rates instead of home runs allowed under the belief pitchers do not actually control how many home runs they allow. This is not a conversation on these statistics as much as what they represent, and for a full account of these stats please go to Fangraphs.

The underlying trend is these numbers and the way we use things like BABIP to talk about who is getting lucky at the plate, is that they seek not to document what happened, but what should have happened and what we can expect to happen moving forward. The new statistics are about prediction and control more than a simple description of deeds.

When we talk about someone's ERA now, we also talk about what it should have been -- there is an abstraction from actual, real-world events. Park-adjusted statistics remove the corporeal element entirely, taking the game not as it happens in real space but how it would happen in an imaginary neutral setting. Instead of embodying physical spaces like Comerica Park, these measures exist apart in the ether. Eventually there are two games -- the game that we saw and the game that we would typically expect. Often we act as though the expected outcomes are more real and true than the actual events we witnessed.

Wins Above Replacement (WAR), an earlier version which was called Value Over Replacement Player (VORP) at Baseball Prospectus, is now the chief measure that people think of when they think of sabermetrics. Its value, or lack thereof, has been debated endlessly, often pointlessly. It is a good statistic, not a perfect statistic. However, as fans, it is revolutionary in how we view players who are also fellow human beings. Conceiving of players purely in terms of value over an imaginary replacement level player is attractive because it allows us to reduce every player to a single number. Doing so, we can then compare them to others and say definitively who is more valuable.

The problem is that this is a deeply unsettling way of speaking about human beings. We are comfortable doing so when we talk about baseball players, but what about other citizens? We often try to quantify the performance of teachers for example, and it is a gross trend. There is a level of transcendence to human beings that goes beyond value and that is true of baseball players as well. Brayan Pena was frankly not a very valuable player, but he was a rich part of the fabric of the Tigers' 2013 season (NERTS > .4 WAR).

Most people with an intimate awareness of sabermetrics and how they operate are cautious about claiming too much. Bill James for example, in this excellent essay describes the fog and the problems that it presents for our understanding of baseball. Essentially, he says that we should be aware that we know less than we think we know. Sabermetrics often present us with an illusion of compete knowledge and there is a need to invoke Socratic wisdom and the awareness of our own ignorance. We know more than we knew before, but we still do not know that much. And of course, the season ends with the playoffs that are a yearly routine fetishization of stochasticity.

The purpose of statistics increasingly becomes more interesting and elusive. To what extent do numbers supplement the visceral and fleeting experience of being at the ballpark? Do we use them to preserve the past or to understand things about which our eyes deceive us? Are numbers supplementary to the season or the predominant mode of understanding? Which do we take to be the most true -- the random world we see or the invisible world of order? The thought arises that sabermetrics provide a nuanced and thoughtful way of viewing baseball, and we have given too little thought to sabermetrics themselves and what the use of analytics means about us as fans and how we watch the game that we love.