Intro to Forecastinating

One aspect of my work as a transit planning consultant involves forecasting ridership on a rail or bus line 25 years into the future. You can imagine the inherent inaccuracy of this. Yet we still do it, in part because it is necessary to at least attempt to plan for the future, but mainly because getting paid to predict the future is pretty cool.

Many of us forecast futures of some kind – economists, marketing managers, even your doctor predicts the course your hacking cough will take in determining a treatment. And once weathermen showed us that accuracy is not a prerequisite for repeat business, it became a lot easier to establish oneself as an expert in some sort of forecasting (see: “What housing bubble?”).

With this in mind, I have determined that my transit ridership forecasting is far too timid – what with its foolish reliance on stacks of data and intelligent input and statistical significance. It is time to establish a bold new field that generates fantastic predictions based on nothing more than completely tangential and/or meaningless information. No, I’m not talking about 24-hour news analysis. I call it Forecastinating.

Forecastinating uses the appearance of a seeming past relationship as the argument for accepting a seeming foolproof future. It avoids complexities and mitigating factors at all costs. And because we’ll accompany it with a snappy graphic, you will believe. Of course, when it proves untrue sometime in the future, I will mumble something about standard deviations and sample size, point in the opposite direction, then run away to a new prediction.

Despite this you will continue believing, for a simple reason: You want to. We like to believe there is order in the universe, and that with enough poking around, we can figure out what that is. So read on – the future won’t forecast itself!

(or will it? Find out in a future post!)