Friends, I have taken a dive into the ocean of numbers, and in these numbers I have found the face of God.
On Thursday morning, I asked you to take a 31-item quiz in which you were given nothing but a minor-league player's name, and asked to guess whether that player was a pitcher or a position player. It's a binary test, so one would think that random guessing -- in other words, deciding your answers with coin flips -- would eventually shake out to an average of 15 or 16 correct answers per test, right?
I accepted the scores of 93 people. In reality, more than 93 people took the test, but I didn't want to rely on the data accumulated by Sporcle.com's system, since it would count the inevitable dawdler who was either completely screwing around or didn't put much thought into his or her answers. I wanted the scores of people who gave the quiz enough thought to tweet me their answers.
One might say, "well, of course you didn't get as many reports of low scores. People were too ashamed, or not proud enough, to send in their scores!" That doesn't appear to be the case. Below is a screenshot of the Sporcle data, which, while not as detailed, is more or less reflective.
In total, 58.06% of the questions answered were correct (in other words, like a coin coming up heads 58.06% of the time). That may not seem like much of a statistical deviation... unless it's 2,883 coin flips we're talking about. Indeed, 93 completed quizzes times 31 answers equals 2,883 total answers.
Now, it is possible that some quiz-takers were expertly familiar with the Rockies' minor league system (where these players play), but I received a few scores from folks who said, "I knew a couple of these names already." I was forced to discard these scores, and I sincerely thank these folks for being honest. That, for the most part, takes care of the "familiar names" issue to my satisfaction.
Well, enough of that. What have we learned here? What is the significance of these numbers? My sense is that if 73.2% of us are better-than-average guessers, and if we can guess correctly 58% of the time within a huge sample size if we're thoughtful bout it, there is something here. There is, to some extent, a "pitcher name."
Why? What does it mean? How could this possibly be? How could a baseball player's birth name have anything to do with whether he becomes a pitcher?
We aren't done with this. I feel as though we have just started. Thanks to you, we have found something genuinely confusing and wonderful.