Thursday, August 14, 2008

On Predictions

Now is the time of year when the internets are awash with predictions regarding the upcoming college football season. Everything from how many wins each team will have, to who will win which conferences, to who will play for the national championship is being predicted by sportswriters and bloggers alike. But there's a problem.

Gunther's Prediction Paradox states that in their natural form, correct predictions always grant the predictor exactly zero bragging rights. This is because there is a direct relationship between how brag-worthy a correct prediction is and how much luck is assumed to be involved in said prediction.

For examples we'll look at two extremes. Predicting that "the sun will come up tomorrow" relies on no luck at all, and when it happens, bragging about your prediction being correct will make people think you're a total moron. At the other end though, a prediction of "tomorrow's lottery numbers will be 5,15,34,38,35,&41" is based completely on luck, and if those numbers actually come up, bragging about being correct will make people think you're a total dick. (But you'd be rich so they'd still be nice to your face.)

The factor of skill would seem to negate Gunther's Prediction Paradox, since not all correct predictions are based on luck alone. But skill has an interesting relationship within Gunther's Prediction Paradox: those with the most confidence and skill at predicting outcomes are the least likely to brag about it. Think about it - when was the last time you saw a website by a successful, professional sports gambler that bragged about their methods or how spot-on their picks were? Right. You haven't. That's because they keep their methods and success to themselves. Did Biff Tannen go around flashing his Grays Sports Almanac from the future? No, he kept it to himself. (Yes, Biff was both a moron and a dick, but that had nothing to do with his predictions. Don't try to change the subject.) The websites that do brag about their success at predicting? They're trying to make money by selling their secrets, which, if the secrets were actually that good and profittable, they wouldn't want to sell. The pros are willing to put their money on the line, and that does plenty of talking.

So, am I saying that people shouldn't make predictions, or that if they do they should never brag about them when they're right, or that the only people with skills are the professionals who make a living at it? Of course not. Part of the fun of college football is trying to guess what's going to happen each week. And we all know that college football fans like to brag about their team, their conference, their coach, etc. And there are some fans out there whose knowledge and expertise rivals anyone in the business.

But the question I'm getting to here is, are there ways to brag about correct predictions without people thinking you're either a total moron or a total dick (or both)? Yes there are, and as usual, the answers are found somewhere in the middle.

Rule #1: Find the Right Balance - One of the keys to making solid predictions is knowing how assertive to make your statement. For instance, your predictions can't be too easy, general, or popular. Predictions like "USC will win the Pac 10" or "Duke will win less than 5 games this year" are wimpy and nothing to brag about. At the same time, your prediction can't be too tough, improbable, or against the grain, mainly because people won't believe that it wasn't just luck. But you can shade to the assertive side and go out on more limbs if you...

Rule #2: Reveal Some of Your Methods - This is important because it's the main weapon you have to prove that you're not just relying on luck. Showing your mental path and how you reached your prediction will help prove that you actually thought it out and are basing it on information, evidence, and your own skills. It's not that you have to give everything away or reveal your Uber-Secret Predicting Formula to the world, but you really have to have some sort of proof that your guess is based on more than just luck.

Rule #3: Admit Your Mistakes - This rule is all about not being seen as a dick. Really, how many sportswriters and broadcasters do you know who come up with wild predictions, and then when they're completely off-base never man up to being wrong? But then when they do get something right the I-told-you-so's start flying. It happens all the time. I'm not saying that you have to acknowledge every time you screw up, but a simple "Yeah, I was wrong on that one" every now and then can do wonders for your likability. Not owning up to your mistakes (or making excuses for them) is the mark of a true dick.

Following these three rules will help you become more respected and well-liked in your predicting community.

No comments: