Why do tennis players, doctors, and band members make better business predictions?
They All Know That It Starts With A Baseline
Every day business men and women the world over making a staggering number of business predictions. Sadly, most are horribly wrong. Truly, humans are really bad at making predictions! We aren’t very good at applying general statistics and we give more power to our inside view of things than the more statistically sound outside view that is supported by better data and objectivity.
Before Making A Judgment, Remember You Are On The Court
Predictions are judgments. Judges preside over courts. Any player of a court sport like tennis or volleyball know that the game begins at the baseline. Business predictions should start here, too. Unfortunately, while we all learned that failing to pay attention to the baseline in tennis results in a fault, few in the business world realize that similar ignorance in forecasting will result in faulty forecasts.
Baselines are built on an outside view. They have little to do with your current situation and everything to do with the type of scenario you are predicting. “4 out of 5 businesses fail…” is both a well-known fact and a baseline. If you are trying to predict the chances of a particular start-up failing, the answer starts with 80%. The particulars of your particular business are not likely to matter, not for the baseline. In certain cases, you may be able to refine your baseline when a large amount of data is available for a more specific baseline. For example, “9 out of 10 restaurants fail” would be a better baseline if your start-up was a restaurant.
Now Remember, The Bassline Sets The Rhythm
Your baseline sets the starting point for your prediction, but it is not the whole ensemble. You are free to make adjustments, but you if don’t follow the bass, it is just going to be a bunch of noise and nonsense.
Perhaps you believe that your start-up idea is better than others? Perhaps you have a better management team, more capital, or a strong competitive advantage? All of this can modify your prediction, but it still must do so using the baseline as its starting point.
There is an entire body of statistics dedicated to properly measuring how we should change our probabilities based on new information. Bayesian statistics are excellent for this sort of work. More here. Unfortunately, Bayes himself was not that great at applying them and you aren’t likely to be the exception. But if you can acquire the discipline to use this tool, the result will be much better predictions and outcomes.
More Information Should Better Inform the Prognosis
Many doctors collect a baseline measurement before begin treatment on a patient. Whether that is a fetal heart rate, blood pressure, or a starting brain scan — physicians utilize these baselines to determine the impact of their treatment strategies. As a forecaster, you should, too.
The original baseline serves as a benchmark for measurement, while the new measurements should continue to correct the original forecast. In this way you are able to collect progress from your initial starting point, while developing a better understanding of your final destination as well. Once again you must continue to be disciplined about the true impact of new information, stay clinical.
The understanding to start with a solid baseline, to adjust it only through the use of solid statistical frameworks, and the discipline to monitor and inform your predictions over time will vastly change the reliability of your business planning. Organizations that effectively embrace this process become learning organization. Organization that learn have a huge competitive advantage.
Ken Fisher is a highly successful money manager who details a similar forecasting process applied to investments in his book:
