Stats in the Wild

Perhaps Mac Engel is a brilliant satirist or perhaps he is just trying to get people to pay attention to him.

As a statistician, I’m going to make an assumption (that’s what us stats guys do), and I’ll start by assuming that Engel is indeed sincere in his belief that “Mathematics is not a friend of baseball” in his article of the same title. Assuming this is true, it seems clear to me that he doesn’t really have any idea what he is talking about.

Engel’s article opens with some jokey, over-the-top claims about math in sports (number-crunching will ruin your kid’s dodgeball game!). But his overall point appears to be in earnest. In his second paragraph (if you can call anything in this a paragraph; it’s more collection of individual sentences) he says (emphasis added):

Scores of math whizzes, nerds and live-in-their-parent’s-basement geeks are threatening to turn Royals at Rangers into a Bobby Fisher vs. Boris Spassky chess match, minus the intellect.

This sentence demonstrates clearly that Engel does not understand the nature of statistics today and how they are applied to baseball.

First, the phrase “live-in-their-parent’s-basement geeks” is being used derogatorily here to indicate that statisticians are losers unable to function in society, and are thus relegated to their parents’ basements because of their personal and professional failures. This was my first indication that Engel has no idea what he’s talking about. These “live-in-their-parent’s-basement” geeks surely exist , but in today’s world these geeks more often come in a different variety: the employed-in-a-recession-six-figure-income kind. (There’s also the rock-star-stats-geek like Nate Silver, who is more important to the New York Times’ website traffic than most of its regular journalists.) While journalism jobs are rapidly disappearing, jobs based on data are expanding so rapidly that there are not enough people with the skills to fill these positions.  Nearly all of the successful businesses in the world know the value of data (see: Book, Face) and you can argue that one of the big differences between Obama and Romney in the last election was how the two campaigns handled, managed, and took action based on data and analysis (Obama embraced it, Romney largely ignored it; how did that work out?). The point is, the best way to make decisions is when those decisions are based on data and appropriate analysis. Baseball is a game where many decisions must be made, and those decisions can best be made when they are aided by data.

Secondly, Engel seems to fail to grasp a major difference between chess and baseball: chess is deterministic; baseball is not. Every time a chess player moves Bg5, the bishop gets there with certainty, whereas in baseball, not every routine fly ball will get caught (and sometimes they bounce off your head and go over the fence). He correctly notes that this uncertainty is what makes baseball exciting. As he says: “Math is never wrong. Baseball very much is, which is why I love it.” I’m not sure exactly what he means by “math is never wrong,” but when he says “baseball very much is,” I assume he is implicitly referring to random event outcomes. And this is precisely why math, specifically statistics and probability, is useful in making decisions in baseball. Statistics and probability give us a principled way to quantify uncertainty about events that have occurred in the past and events that will happen in the future. Statistics can’t say what is going to happen on any given play with certainty (e.g., Miguel Cabrera will hit a home run in this at-bat), but it can help someone to make statements about how likely an event is to occur (e.g., Miguel Cabrera is more likely to hit a home run in an at-bat than Norichika Aoki). And just because we can’t say with certainty what will happen on any given trial doesn’t mean that it isn’t useful. There are entire companies that exist based on events that might only occur 1 in 1000 times (think click-throughs on an ad, or responses to fake Viagra emails).

So what numbers and maths are specifically ruining baseball in the opinion of Engel? Is it the often cited and complicated wins above replacement (WAR) statistic? Nope. Is it the opaque Ultimate Zone Rating (UZR)? No, not that either. The entire rest of the article is about how math has ruined the game and made us a “prisoner to all of these bleepin’ numbers,” and his example is: PITCH COUNTS!

Pitch counts are an example of math in baseball in the same way that page numbers are an example of math in literature. If you think numbers and math are ruining baseball, which I disagree with thoroughly, I’m sure you can make a reasonable argument, but you need to be more persuasive than talking about pitch counts. There is a reason the Oakland A’s won all those regular season games with their tiny payroll, and it wasn’t simply luck. They were making better decisions that everyone else at the time because they were using data in a better way than their opponents, and it gave them an advantage in the long run.

So what I’m trying to say is that math and statistics aren’t ruining baseball. Now if you’ll excuse me I’ve got to go do the dishes. I can hear my mother yelling at me from upstairs.

I’m still blocked from following @BillBarnwell on Twitter.  But what I don’t quite understand is what is the point of blocking anyone if your account is public?  I’m blocked from following him, but I can still read all his tweets because his account is public.  So, what do you gain by blocking someone?  There must be some benefit to it?  Is it just that I can’t direct message him?  

Cheers.

Deadspin answers the question of whether you live in IHOP America or Waffle House America.

I also found it particularly interested that at the very end of the Deadspin article they say:

Waffle House does so much business in the South that the impact of hurricanes on Waffle House service is actually an informal measure of hurricane intensity.

Cheers.

I graduated from UConn in 2011, and I gave a talk there this morning.  After my talk I met with some friends for lunch who are in the middle of their dissertations.  As it starts to get nice outside in Storrs, many of them are having trouble writing their dissertations.  As such, my former office mate (and future “doctor”) has been running “Dissertation Boot Camp”.  I’m told that this wasn’t her idea, but she has added her own touch to it.  Basically, student pay her $80 dollars and if they don’t commit too many infractions over the period of boot camp, they get all or part of their money back, hopefully with a completed dissertation.  Below are the rules and consequences that Val has set forth for boot camp.  If you want to use Val’s template for your own dissertation bootcamp, you get get it here.

The rules:

• Show up on time. You must sign in/out when you enter/leave the lab.
• Keep talking to a minimum. If asked to stop talking, you must do so immediately.
• Internet usage is for dissertation related business only
• You are allowed one hour off during bootcamp hours to be taken at whatever time (or times) you chose.
• Apply for a minimum of 3 jobs per week outside of dissertation bootcamp hours. Print confirmation and submit at end of week.

The consequences:

• First infraction: Verbal warning
• Second infraction: Buy Val coffee that morning
• Third infraction: Buy Val coffee for a week
• Fourth infraction: Val will take $10 from your dissertation fee
• Fifth infraction: Val will take $20 from your dissertation fee
• Sixth infraction: Val will take $50 from your dissertation fee
• Seventh infraction: Val will rate you on http://www.ratemyprofessor.com. It will be a one star rating.
• Eighth infraction: Val will send an anonymous email to your advisor. It will say “<Insert your name here> is not motivated or smart enough to earn a PhD”
• Ninth infraction: Val will send an anonymous email to your advisor. It will say, “<Insert your name here> thinks your research is not at all innovative or useful.”

I feel like tornadoes have been everyone recently.  I was driving home yesterday and I was listening to an NPR piece about the recent tornado in Oklahoma, and the broadcast was interrupted by a tornado warning for western MA/Northern CT/Eastern NY.  Even though this warning happened to be no where near me, I’m way more terrified of tornadoes than I ever used to be as I figured they couldn’t happen in Massachusetts.  But my false sense of security was destroyed almost 2 years ago when a tornado missed my house by about a quarter mile on the first official day of my post-doc appointment.  I certainly won’t forget my first day of work.

Anyway, here is a nice visualization of all tornadoes that have killed someone since 1950.  Click on the image to get to the interactive version of the map.  Stay safe.

Cheers.

1902 – Two deaf-mutes face each other for the first time when Dummy Hoy leads off for the Reds against Dummy Taylor of the Giants. The Reds win, 5 – 3, with a 5-run rally in the 9th. Hoy goes 2 for 4.

–baseball-reference.com

Wow.