Bad arguments drive me crazy…

One thing that causes me persistent heartburn are those statistically invalid arguments, often made by those who either do know better or should know better.

It is one thing to, say, pass a statistics class, or even properly use statistics to analyze data. It is a different thing to make statistical thinking as part of one’s every day thought process.

Here are some examples:

1. “X is 60 years old and ran a 3:10 marathon. So what is your excuse?”
The fallacy: X is almost certainly an outlier. Example: a 61 year old lady ran a 3:12 at this year’s Chicago marathon. It was Joan Benoit, who has an Olympic gold medal and has a PR of 2:22.
My response: “Y is your age and ran a 2:05. What is your excuse?”

2. “It is “racist” or “classist” to use the ACT for college admissions (or placement). The HS GPA is a better predictor.
The fallacy: just because one measure *might* be a better predictor doesn’t mean that several measures, taken together, aren’t better predictors than any single measure taken alone. After all, NFL teams measure several things (speed, strength, speed through a cone course, etc.)

3. “X achieved great things and there were naysayers saying that they couldn’t.”
Fallacy: most of the time, the naysayers are right. Most do not achieve elite results, by definition. Related: “hey they laughed at Einstein” (well, not really, he got 4 papers published in a top notch physics journal while still a graduate student that no one ever heard of..I don’t think that is getting “laughed at” but whatever.

4. “The deficit/debt is X dollars..that is more than it has ever been, etc.” Duh. We are a growing country. That is why debt and deficits are measured in terms of GDP. Newsflash: a million dollar deficit might be a lot for a small town, but it is nothing for an entire country, due to the relative size of the economies.

5. “Polls..ha, ha, ha, ha…they predicted Hillary would surely win!” Uh, no. Yes, a couple of models based on polls had her as a heavy favorite. BUT, not all of them, and these were MODELS. Nate Silver game Trump a 34 percent chance (roughly the probability of a below average NBA player missing a free shot). And the predictions were pretty spot on in terms of national vote; they were all just a bit off in a few key states..and off in the same direction.

My own “lower confidence interval” had Trump with 266 EV, 4 short of what he would have needed.

6. “Hey, these athletes get paid a huge amount of money to play a game whereas people who do more important work are paid far less..”
Fallacy: entertainers (athletes, actors, musicians) tend to be either “strike it HUGE” or “amateur” with not much inbetween. Put another way: being 95’th percentile in math probably means that you can earn a nice living as an engineer. Being 95’th percentile at sports means..well, maybe getting playing time at a D-3 program..if you are lucky. Only the most outrageous outliers make it as a professional athlete.

October 20, 2018 - Posted by | social/political, statistics |

No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: