In statistics, there’s a concept called “regression toward the mean.” It dictates that results that lie far outside of the average range tend to be drawn back toward the average over time.
Understanding this concept can help you understand some human dynamics… and save you money too. We’ll start with an interesting – and familiar – human dynamic.
There’s a common belief held by many parents, coaches, bosses and teachers – people who seek to bring out the best in others – that when somebody underperforms, the best strategy is to chew them out in order to motivate them.
And this strategy does appear to work fairly often… just not for the reasons we think it does.
The Case for Effective Feedback
Psychologist Daniel Kahneman discovered this when he was teaching flight instructors in the Israeli Air Force many decades ago. He had been explaining how rewards for improved performance are more effective than punishments for poor performance or mistakes.
One of the instructors disagreed, however…
On many occasions I have praised flight cadets for clean execution of some aerobatic maneuver. The next time they try the same maneuver, they usually do worse. On the other hand, I have often screamed into a cadet’s earphone for bad execution, and in general he does better on his next try. So please don’t tell us that reward works and punishment does not, because the opposite is the case.
Kahneman acknowledged that the instructor was absolutely correct in his observation of what happened… but completely incorrect about the cause.
The cadets who were praised for exceptional performance did worse the next time, and the cadets who were scolded for poor performance did better the next time – but not because of the praise or scolding.
Both groups were likely to see their performance drawn back toward their mean performance – for better or for worse – due to random fluctuations.
A pilot who executes an exceptional technique is just not likely to do it as well the next time. And a pilot who makes a mistake is not likely to do quite as poorly the next time.
The instructor experienced what he thought was cause and effect. But statistical reasoning often doesn’t match what our experience tells us. As I’ve written before, our experience connects to our fast, intuitive thinking, whereas statistical reasoning requires our slow, more deliberate thinking.
So while punishing people into excellence may make intuitive sense, it doesn’t really work the way generations of well-intended scolders thought.
Indeed, “spare the rod, spoil the child” and other clichés that reflect a philosophy of punishment aren’t as effective as some may think.
This reminds me of the classic comedian Jack Benny and his approach to mentoring.
From time to time, a young comic would ask his opinion of their act. When the act was mediocre, he’d simply say, “That was good.” Pure praise, but no feedback beyond that. He didn’t think it was worth his time and effort.
When a young comic showed promise, though, he’d tear their act to shreds. He’d dissect everything that he saw, leaving them wounded and overwhelmed – but with detailed feedback on how to improve.
He could’ve bypassed the cruelty. The clear feedback from a respected mentor would’ve been at least as effective.
In fact, this is one reason free societies are better at innovation than authoritarian ones: You can scare people into doing rote behavior, but you cannot scare people into being curious, creative or growth-oriented.
Curiosity, creativity and growth require a part of our nervous system – our higher brain functions – that fear and anger severely restrict.
Understanding this concept is important for parents, coaches, teachers and mentors.
But it’s also critical for us as investors.
Don’t Scold Your Broker
Property values in my town grew tremendously over the last several years. And while they recently dropped a bit, they’re still above their multiyear averages.
This concept is just as true in investing as it was for Kahneman’s exceptional fighter pilots: An extraordinary performance will likely be followed by a less extraordinary performance – not bad, just not quite as good – purely on the basis of probability.
Regression toward the mean is not regression below the mean. It’s just that if an investment performs exceptionally well, there’s a high likelihood it will correct back in the direction of its average performance.
So if you find that one of your great investments draws back a bit, that doesn’t mean it’s time to panic or scold your broker. It’s more than likely a simple random fluctuation, a regression toward the mean.
P.S. Please visit my new blog! It’s available here.