Monday, November 12, 2012

Droning on


I’ve taken a lot of classes as a college student. As a result, I’ve come up with more than a few ways of wasting time during lecture. One of my personal favorites is FanGraphs – an analytically inclined baseball blog. The irony of reading up on new hitting metrics during my stats lectures was delicious and I like baseball, so it’s a pretty good marriage. One of their tricks is to post two stat lines without identifying the player. If done right it illustrates for readers how their preconceived notions of player value may have clouded their judgment when evaluating talent. For example, which of these players would you rather have:
Player A – batting avg.: 0.302; on base %: 0.360; slugging %: 0.522*
            Player B – batting avg.: 0.251; on base %: 0.332, slugging %: 0.475*

*Stats per Baseball Reference, contract info (below) via Cot’s Contracts.

Player A is clearly better right!? Well player A is Aaron Hill, a second baseman for the Arizona Diamondbacks who pulls down $5 million a year and didn’t even make the all star team this year. Player B is Mark Teixeira; he of the $180 million contract and New York Yankee fame. Unless they had already done their homework, few baseball fans would guess that Aaron Hill had a better 2012 than Mark freakin’ Teixeira! Anyways, if you’re into this sort of thing it can be quite fun, and since the presidential election was just a short while ago it might be a cool trick to apply to politics. So, try this one on for size:
President A – drone assassinations: 278; enemy combatants detained: 2,400; awesome quote: “I’ve now been in fifty-seven states; I think one left to go.”
President B – drone assassinations: 52; enemy combatants detained: 1,400; awesome quote: “I know the human being and fish can coexist peacefully." 

President A is the newly reelected Barack Obama. President B is George W. Bush. No, it’s not what I expected either – kind of a sharp contrast huh? Let’s go through and break down each category.

Drone Assassinations
These numbers are per this article from USA Today. Even if we get shifty and say the US’s drone capabilities weren’t humming at full speed until 2008, GW averaged 52 drone-targeted assassinations a year. Obama put up just under 70 annually.* Moreover, the democrat managed to pile up those numbers while (ostensibly) drawing down the wars in Afghanistan and Iraq. Impressive.

*The LA Times implies it’s only about 42 that came during Bush’s final year of office.

Enemy Combatants
For someone whose administration basically invented the term “enemy combatant,” I’ll bet it’s pretty embarrassing for Bush to loose this category.* Even if we assume the Drone numbers are affected by advances in technology that make them more readily available to POTUS 43, our ability to capture people and throw them in jail remains unchanged across administrations: Obama had to do some real work to take home this category. Here we’ll give Bush credit for the entire pool of prisoners shuttled through Guantanamo Bay as well as one-fifth of the denizens of Bagram Air Base in Afghanistan (per CBS), bringing Bush's total to 1,400 over his eight years of presidency. To his credit Obama mostly kinda sorta kept a campaign promise to close down the Cuban prison camp**, but he also is responsible for the remaining four-fifths of Bagram’s population. That’s a full thousand more prisoners than Bush – Obama wins again! Yikes.

**He did largely reduce the prison population there, but Obama also signed a bill restricting transfer of prisoners to other facilities in the US or abroad. As a result, over 150 prisoners remain in GitMo. In the end, the fact that it remains open is largely semantics – I can’t imagine there’s much difference between being an enemy combatant incarcerated in Cuba or in Afghanistan.

Quotes
We all say dumb stuff every once in a while. It’s no crime. Here’s Obama and here’s Bush.

So what does this all mean? Well since it is a comparison between Obama and Bush it doesn’t really tell us anything about who should have won the election on November 6th. It doesn’t tell us much of anything about who was better at his job either. Like a good statistic should, all these numbers do is make apparent to us things that we would normally miss. We’re conditioned to think a certain way about democrats and republicans just like we’re conditioned to think a certain way about Mark Teixeira and Aaron Hill. It’s easy to get caught up in rooting for a certain team, but often times it can be quite instructive to take a step back and ask, “are we really winning?”

1 comment:

  1. As much of a fan I am of quantifying parameters, I feel the analogy used (and extended) here of taking two players' stats at a particular moment serves as any basis of comparison.

    Now I don't know no nothing about no baseball but right away the numbers seem suspect. For each of these percentages, what was the n used to arrive at them? How does this n compare to the n of previous seasons? For this averages, what are they like normalized to the league, the player's best season, the player's average, to the team average? Assuming there have been multiple seasons for each player, what does it look like to plot each player's seasonal averages as a function of time? Is there an observable trend? How well can a simple fit be made and do the current values reside within it?

    You could even say that a player does not a team make and see how the stats of each player's teams have changed since the players joined them. Or from a team owner's perspective, how much has revenue on a variety of fronts (television, merchandise, ticket sales, etc) changed with each player?

    These are all details that matter. Without context, facts mislead and statistics obfuscate. Of course I don't expect you to write a dissertation on this blog about the comparison of baseball player stats and presidential candidates, but only rant to suggest that we all understand that our interpretations of situations is limited by the context in which we can find and work with facts.

    ReplyDelete

Keep it civil.