"Objection sustained! The jury is instructed to ignore these remarks." The judge's stern warning made the defense lawyer shut up and sit down, but as he did he smiled with satisfaction. It is a familiar trick by lawyers in an effort to bias the jury's opinion; the inevitable reaction by the judge that the jury should ignore an unacceptable claim is ineffective. An impression has been created and for all practical purposes cannot be erased from the jurors' memory. No matter how much they try to forget what they heard, something will remain at least in their subconscious mind.
In a similar situation astute businesspersons think little of the favorite practice among academic forecasters namely to ignore recent-history data and try to predict them with their models. Pragmatic men and women in the marketplace know that once you have seen the outcome it is difficult to forget it. Forecasts can be biased, if unwittingly, in a number of subtle ways. For example, if you already know that a particular product has become a bestseller, you will not persist working with a forecasting model that from early-history sales predicts an imminent decline for the product in question.
The only real proof of success in forecasting involves the test of time. "Make your forecasts and then sit and wait until the time comes," I was told. "Only that way you'll be able to convince people of an enhanced ability to make predictions."
In 1992 Simon & Schuster published my first book Predictions – Society's Telltale Signature Reveals the Past and Forecasts the Future. The book broke new ground in understanding society and ourselves by applying fundamental scientific concepts to predicting social phenomena. Twenty years later the challenge arises to confront the predictions made then with the way reality turned out to be. This is something forecasters generally refrain from doing.