Regression to the mean
Aug 17
A manager I once knew would always use the same technique to motivate his team irrespective of the way they performed. He had come to this conclusion after many years of management.
When his production staff had a disastrous run – he would use harsh words to try and get them ‘fired’ up again. He would then see an improvement in performance in the person. However, when people in his team were doing well, he would use the same harsh technique to try and keep performance going at a high level. Not that performance did stay high – it didn’t. But he had noticed that when he used to praise team members, performance would drop. He had come to the conclusion that performance seemed to drop less when he used the same consistent critical method.
Like my blog on causation v correlation, what this production manager had done was forgotten about the concept of regression to the mean. It is a common problem. It explains stock market returns on an annual basis, why some people think homeopathy works (it doesn’t!) and why sports people who are labeled as ‘stars’ tend to have a poor subsequent season.
Anything that can be measured will have an average. The average will be that for a reason and over the long term – performance will vary around that average, unless there is an exceptional and permanent change (new technology etc).
So there is a strong likelihood that if you have one or two exceptional years or months of performance, that future performance will revert to the mean. Of course this does not mean that permanent improvements are not possible, but there have to be some underlying reasons for this. It is not possible to consistently outperform an average.
Why is this important as a concept for entrepreneurs and for business angels?
Many entrepreneurs when writing their business plans extrapolate from data that suits them over a relatively short period of time. They may assume that a new trend marks a permanent change in the landscape. It may – but I would want to know how the business would cope with a regression to the mean. That is a basically a way of saying – a return to the average. Can their breakeven points, and profitability points still be hit when the fast growth inevitably returns to an average?
Business angels will want to know the answer to that question.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=87a82fc3-4fdb-4e40-8df3-9d0b670fb504)
Hi and welcome to my blog. 