Freakonomics: Sumo Slander (?)

Ah, yes: Freakonomics; they made it into a movie, you know. I just saw it. It was broken down into segments, many of them directed by Morgan Spurlock, the McDonalds marauder. His movie, 2004’s Super Size Me, is an absolute love letter to bullshit science: he proved that a vegan might feel like crap if they ate McDonald’s food for a month. A vegan would feel like crap for a month if they ate a single hamburger, so that proves nothing! Additionally, had he done it for a second month, I’d be willing to bet he would have begun feeling normal again, as it takes about that long for the human body to adjust to a major life change. (Ask any drug addict how long it took them to feel well after quitting drugs; probably they’ll say: “about a month”.)

So my guess is, he had an agenda to attack McDonald’s (and get famous), seeing as he’s a hippie, so he did an extremely dubious hatchet job on them, and pretended it was real science (a scientific study with a sample size of 1!). What a load of crap! And he’s the guy writing and directing the stuff in the Freakonomics movie? Really?? Awesome! So you know it’s going to be absolute hack science with no real evidence to back anything up (at least in the movie)! My mouth is watering already!

Okay, so they get to a segment on Sumo wrestling in Japan, and they say that they must all be cheating because the numbers look like it. Um, looking at numbers doesn’t prove anything if you don’t know what the data is about. How much do they know about Sumo wrestling? I’d say probably not much. But that won’t stop them from calling them all cheaters, I’ll bet! (Luckily for them, they live in the United States, far from the Sumo wrestlers!)

Okay, so here’s what they say: when a Sumo wrestler wins 8 matches, they move up to the top tier; and then they found that when a Sumo wrestler with 8 wins wrestles another one with 7 wins, the one with 7 wins will prevail 75% of the time, so they must be cheating, right? Um, not so fast there, number-crunch cowboys! A guy with 8 wins doesn’t need to win as much as the guy with 7 wins, so he may not try as hard, because he’s not as desperate, and he might not want to hurt himself for future matches. And that may not even be a conscious thing, either. If you’re going to call them cheaters, you’d better have some real evidence, and they don’t. But still, they’re all cheaters!

Let’s look at another sport: Soccer. In the group stage at the World Cup, when a team wins what they need to move to the next round, they usually don’t try as hard in the last game. That’s to prevent injuries and stuff, since they’re already through. That’s not considered cheating, and I’ll bet it’s the same with Sumo wrestling. And guess what else: almost every sport that has a group stage type of tournament acts the same way! Are they all cheating? If they said so, they might get beaten up, so let’s just accuse the sport from the faraway country where they’ll never run into them! (That sounds the safest!)

Hmm, you don’t think it’s possible that the number nerds are trying to get revenge on the rich jocks, do you? I’ll bet that’s what’s going on here! They’re mad that the wrestling team always gave them wedgies and purple nurples, so they’re getting revenge the only way they can: through slanderous suggestions of corruption and cheating! Well, they might be right, but I don’t know. It seems kind of fishy to me… (Yes, admittedly they did have a famous retired Sumo guy at the end of this segment come out and admit there was match-fixing, confirming their suspicions. But how do I know it wasn't just one of these statisticians in a Sumo wrestler costume impersonating the real Sumo guy just to try to make it look like they were right? You know that's what they did! I heard they do that kind of stuff all the time! In fact, when you look at their purported record of being correct about this stuff, I'd say the numbers prove that they're cheating!)


(Actually, the Freakonomics stuff is pretty interesting, and I'm just funning them.)