Measuring progress for technology adoption in freight may seem like an easy thing.
All you need do is look at the amount of dollars spent, projected stock valuations of companies, market projections by expert investment firms, projections by researchers at non-profit groups, media coverage, government projections, other trusted estimates and many more supposedly easily measured metrics.
In our collective experience, every single one of these has always been spot on, accurate, correct and unbiased, right? We have grown so confident in the veracity of these projections that seldom do we actually go back and audit them to see if they actually were accurate. It’s almost as if we trust them as much as we trust daily weather reports or hurricane tracks for storms starting out somewhere in the mid-south Atlantic where old cartographers used to draw dragons at the edge of maps.
Sarcasm aside, a fundamental core principle of the stock market, of weather forecasting, of market forecasting – in fact anything to which the word “prediction” can be implied – is statistical modeling, or more exactly, educated guessing. The science of estimation is around trying to model what has happened in the past and creatively use it to predict the future. This works extremely well for flipping coins. It works relatively well for rolling dice at a casino. It works less well for political polling. It gets really hard for technologies.
The complexity of the modeling increases with the number of variables having to be modeled. A variable is something that is not controlled, hence the name variable. The real world has a lot of variables.
The underlying belief that the future is an accurate predictor based on the past is contracted in the fine print of every statement put out by companies governed by U.S. SEC rules, the infamous “safe harbor” rules about “forward looking statements.”
I found this example explanation on an SEC website:
