This is a topic I've been meaning to tackle for some time but never quite figured out how to encapsulate all the things I want to look at into one article of reasonable length. Now that the season is officially over for everybody and there is less time needed to devote to daily news, I wanted to start a series of conversations about sabermetrics: what they tell us, what they don't, and how we can use our own favorite team to help us better understand their attributes and limitations.
That was a long-winded way of saying, "what kind of specific case-study have the Rockies been for advanced statistics?"
So, in this space over the off-season, I'd like to look into a number of different individual metrics as well as kinds of stats that attempt, oftentimes imperfectly, to put an objective number on any number of different aspects of the game.
A week ago, I touched on the difficulties in measuring defense and would like to get much deeper into that with some particulars on UZR and the like but for now, I'd like to give a few brief thoughts on the areas I'ill be covering and what the goal of the whole process should be.
Please feel free to add anything else you would like me to look into for this project by posting in the comments section below.
It is first vital to understand (both for me and for anyone reading this) that I have no end-game in mind here, no specific agenda. I do not intend to show a particular result (i.e. Rox prove BABIP is teh lame!) but rather I intend to ask a series of questions that I hope will bring about better understanding between the people who don't trust the statistics and those who have total faith in them.
My position has always been that I like the idea behind advanced stats. I wrote, near the beginning of the season, on the phenomena known as lineup protection. This was an eye opener and would lead to some pretty great conversation here that ended up with a general consensus that lineup protection probably does exist but that it may be impossible to measure.
So let's begin there. Is everything in baseball measurable? Are there components of the game that have a great deal of impact on the outcome but literally cannot be boiled down to any objective number? If so, what are those components, how do we know that they actually impact the game, and is there any sense in talking about them if they can't be measured?
I used to think measuring base-running was a silly concept that no one would attempt, yet when I complained about how bad the Rockies have been at this, a number of people who are much smarter than me linked to information showing that the numbers suggest the Rockies were actually one of the better running teams in the league.
But how are they arriving at the number? And does this leave anything important out?
This will be the subject of each part of this series. Does it really make sense that BABIP and FIP essentially boil down all of the game of baseball into either hitting home runs or dumb luck? The kind of production the Rockies got from Jhoulys de la Chatwood this season seems, at my own first glance, to show that pitchers make outs all the time that aren't either strikeouts or wild flailing luck.
Mini-Purple-Hazed-Idea: Some advanced stats need verbs, as in "he really got BABIPed on that one," or "this pitcher is totally FIPing out right now." But when a guy swings and misses badly at two pitches to get himself into an 0-2 hole and then proceeds to pop out to the catcher, I have a hard time believing that the pitcher just got lucky because he didn't strike the guy out.
Is giving points for the number of times players went first-to-third by a team who plays on a field with more wide open spaces than a Dixie Chicks song really measuring the value of good base running?
Does WAR throw itself into shadowy ambiguity by not taking into account any specific situations? Is Wilton Lopez's Purple Row's ranking of players based on WAR telling us something more about how limited the stat can be?
What about Eric Young Jr. being nominated for a Gold Glove at least partially based on the stats. And come to think of it, why was defense the first area of the game to incorporate these stats into their awarding process?
I don't have the answers to any of these questions yet, but I think it's important that we are all on the same page when it comes to understanding what these statistics tell us. Every time I hear Joe Morgan at the end of Moneyball, laughing in the face of people who try to win "using a computer," I roll my eyes so hard I can check myself for the aneurysm his comments may or may not have induced.
The same thing happens when I hear someone like Hawk Harrelson defend the pitcher Win stat or dismiss huge amounts of hard evidence and information a lot of people took a lot of time to discover and decode in favor of his own gut feeling on who has TWTW, "the will to win."
Conversely, I have roughly the same reaction when an uber-stat head treats those who still value RBI or anyone who is reticent to accept all of the assumptions of stats like BABIP and FIP as though they are stupid or living in a cave.
Sabermetrics are here to stay, it's true, but they are still in their infancy and those who can see the game only through their lens are likely missing just as much as those who pay them no heed at all.
I will test this theory in the coming weeks and am excited to see which stats hold up the best to someone who is both skeptical and highly interested. I will ask the poll question again at the very end to see if there have been any changes after the investigation.