The Meaning of Statistics

(inspired by my favorite blog- joeposnanski.com. If you haven't read it I suggest you do, he is bar none the BEST sportswriter in America and one of the best writers period. Man I wish I could write like him.)


The one thing that baseball best represents for me is the evolution of ones tastes. Through ever age I've found a new way to appreciate the game. When I was 5, I loved baseball because it was what my father liked and played- it was simply my way of emulating him.
As I progressed through elementary school, I started to collect baseball cards and while I loved the numbers on the back- the conventional statistics, batting average, home runs, RBI, I loved the mythology even more and more importantly I accepted it. I accepted that heart was what truly mattered more than anything else, that there were players who were just naturally clutch. Like Mark Lemke- a guy whose Triple Crown stats were absolutely abhorrent but who saved his best for when the game was on the line. Looking back the mythology was a metaphor for life, at a time when you are still trying to understand just how the world works, and you desperately want to find out what makes you special.
I was always labeled smart, but more than anything else I wanted to be talented- I didn't want to be able to do a math problem quickly, I wanted to hit a baseball far or be on Star Search. If I could hustle, then maybe, just maybe, I could be just as good as the people with more natural ability.
Over time, that worldview comes into question. As a child, you start to understand that while heart and other intangibles do play a part in accomplishments, the field is winnowed down by talent. And if you're not up to the challenge in that regard, your heart could be the size of the Grinch's after he gave back Christmas and it wouldn't matter. Around high school, when I realized a) that growth spurt thing they talked some much about in health class wasn't happening and b) that I just wasn't that good, I decided to approach the game from a different perspective.

I started reading the work of Bill James, Rob Neyer, tangotiger, and other sabermatricians at first, because I wanted to learn about ways to quantify the effects of defense on a baseball team. Like most young baseball fans, I believed that I was throughly versed in how offensive stats and pitching stats correlated to winning games. I had always been a fan of watching excellent defensive players, but I didn't know exactly how good defense effected the fortunes of a baseball team. It was almost like great defensive plays and great defensive players were window dressing, they looked nice on sportscenter but they were not particularly integral to the overall fortunes of the team. Most balls in play are of the routine variety, the kind every major league caliber fielder can get to, and while it is quite easy to tell the difference between Ozzie Smith and Hanley Ramirez, most fielders do not exhibit characteristics of that extreme. The only statistic to really go on from mainstream sites was fielding percentage, the ability to not make an error on balls that you reach. Even when I was younger though, I saw that fielding percentage had many flaws, most importantly, that it only counted balls that you could reach. I thought it was unfair that a player who was slow he couldn't even touch the ball was penalized, while the player who had the ability to get to it would actually look worse at the end of the season. What reading the sabermatricians did for me was help to see defense in a new light. Things like Win Shares (which is only pretty good), Revised Zone Rating, John Dewan's +/- system, defensive efficiency ratio, and others helped to show just how much good defense could help the fortunes of a team. Dewan's +/- system, for instance, slices the field into zones which a particular fielder is responsible for. It then documents every ball hit in said players zone and then calculates the percentage of the time that a fielder makes that play. If a fielder is successful on a play that is typically made 86% of the time, then he gets a +14%. If he fails to make a play that is made 36% of the time, he gets a -36%. Defensive efficiency ratio is much easier, it just documents how often a team turns balls in play, (everything except what a pitcher can control, homers, strikeouts, and walks) into outs. A team that can cover more ground can obviously turn more balls in play into outs (it's one of the major reasons that Tampa Bay is doing so well with a mediocre offense and the Yankees are doing not so well with a much better offense) Many of the systems are highly advanced, taking into account the speed of the ball, the spin, the positioning of the players next to the fielder. They are all quite intuitive and they have greatly enhanced my understanding of defense. There is great work being done on the effects of baserunning, the reasons why certain pitches are effective, and one of my favorite stats, Win Probability Added, which states how much a certain event, such as a single in the top of the fifth with a man on second, changes the probability that a team will win. Hopefully someday these things will go into the mainstream of sports, but as of now, many sportswriters still disparage the study of the game in this manner.
One thing that I've noticed is that the political bent of the sportswriters who lampoon the use of statistics as the primary measure of identifying the MVP, Hall of Famers, and All-Stars is usually quite conservative. Not necessarily the bible-thumping, holy-rollin, anti-abortion crowd, but rather the misogynistic, homophobic, beat on their chest with false-patriotism type. The ones who have pretty skewed views on masculinity, maybe they're ex-jocks. Anyway, their argument against statistics is that the game is played by humans who cannot be reduced to simple numbers, which apparently is what us stats minded folks do. They see people who use stats as nerds who watch Star Trek, live in dark basements, can't play sports, and hate apple pie and puppies.
What this is actually about is insecurity; the insecurity of knowing that their jobs of analysing baseball, about knowing what it takes to win baseball games, is actually done better by an amateur who watches the game and has an understanding of the stats that document what has happened. Because many of these writers do not understand what the statistics entail, how to formulate them, or what their connection to the game is, they automatically diminsh their worth.
Most bad baseball writers have a set number of narratives that they forcibly and retroactively try and match to the scenario which has taken place. If a bad player makes a good play in a big moment, that player instantly becomes clutch. If a good/great player does badly in a given situation, he is a choke artist. Once that narrative is formed it is impossible to break out of. Nobody cares if Alex Rodriguez has been an absolute monster in many playoffs series with both Seattle or with the Yankees. They need to find a reason for why such a great player has not won a championship. Consequently, his infield partner, Derek Jeter, because he has made big plays in big moments on a few occasions is seen as the consummate clutch player, even when it is not exactly true. Even when he hit batted .148/.179/.259 (average, on base percentage, slugging percentage) in the 2001 World Series where he was dubbed Mr. November. The truth is that the amateur assault on the way we report news and the way we analyze information is happening in every category and hopefully it will keep its vibrancy when it comes to the mainstream.