Why You Should Leave Statistics to the Statisticians

This post is over a month late. It’s been bugging me all that time. I still can’t figure out how to write it. Over the last year I have come to agree more and more with this guy; The Sports Guy sucks. This whole column was about the dumbest thing I’ve ever read in my life. From top to bottom it’s an exercise in deliberate ignorance.

This part was a whole new level of stupid though. My jaw dropped that any one would ever expose their idiocy in a public forum like this.

Put it this way: The Colts weren’t exactly on fire. Admittedly, I am terrified of Manning and have written as much. But Indy had already started and completed two long touchdown drives in the fourth quarter against a good defense. Had the Patriots punted, Indy would have had to pull off a third long touchdown drive to win the game. I asked Peter Newmann to research the number of times a team started and completed three touchdown drives in the fourth quarter to erase a double-digit deficit and win an NFL game since 2005. Here’s how the list looked before that fourth-and-2 call.

2005: 1
2006: 2
2007: 0
2008: 1
2009: 0

In 78 weeks of football dating back to 2005, it happened a whopping four times. Four! If you’re playing the statistics card, why not play that one? By punting, the Patriots would have been asking Peyton Manning to pull off something THAT DOESN’T HAPPEN EVEN ONCE EVERY EFFING SEASON. You’re damned right I just went all caps. Hold on, I have to repeatedly bang my head against my desk again.

(Ow.)

(Damn!)

This is so fundamentally stupid it’s hard to even know what to say. I’ll start with an analogy.

The Sports Gal is playing roulette and red has come up the last nine times in a row. She’s deciding what to do for the 10th roll. She decides to bet on red. In busts The Sports Guy. Are you crazy!? I had my stats people look this up – do you know how often there are 10 reds in a row? It’s less than one-tenth of one percent – what are you doing? Bet on black!!

Needless to say, this is stupid logic. It’s strongly related to The Gamblers Fallacy, but it’s basically confusing probability of an event with probability of an event given that most of that event has already happened. In the above example, the probability of red coming up is the same as always. It doesn’t matter what has come before. And that stupid forced statistic is completely different when you look at it after most of it has already happened then if you had asked the odds of The Colts doing it before the game started.

Maybe you don’t follow what I’m saying. I’m too annoyed to explain it well. But that’s okay. You don’t have to get it. All you have to do is not be the most popular sport columnist in America using your column to trumpet the exact opposite of the actual truth.

I can’t even write this. It’s one thing to be ignorant of statistics. It’s another to be willfully ignorant. To ignore statistical arguments from people who know what they are talking about and try to make up your own. To misuse numbers so badly that all you do is demonstrate your own stupidity.

4 thoughts on “Why You Should Leave Statistics to the Statisticians”

  1. Ole – This isn’t just any beat writer. He’s just the opposite. What’s both good and bad about him is that he writes more from the perspective of a fan than a normal writer “insider”.

    He’s literally the most popular sports writer in America. His twitter feed had over 1 million people in it’s first day. He most recent book was #1 on the NYT book list.

  2. You crack me up Muttrox. The funny thing is that I know you are really annoyed by this too.

    I don’t believe you need to be a statistician to know that calculation is flawed. I only took the introductory stats 101 class in undergrad and a half semester stats class in grad school, and even I knew that his calculations were incorrect.

    However, that writer got you Muttrox. I’m sure he could care less that he used the wrong calculations to support his thesis. The fact is that he got a rise out of a reader who created a conversation out of it and got people talking about his article. That’s more important than being right for a beat writer.

  3. Granted, it’s not flipping a coin. But if he denies the analogy, it’s on him to show why it’s not valid.

    You know what – I can’t even talk about this. There’s so much stupidity in his column I can’t discuss it.

  4. Yes, his stats are bad. But playing football is not flipping a coin, which you allude to in your analysis.

    The Pats defense had rolled over the last two times that Indy had the ball. It was highly LIKELY that they would do it a third time. I agree with the coach’s call.

Leave a Reply

Your email address will not be published. Required fields are marked *