Randomness and our Feedburner stats
This is a guest post by Brian Lash who can be found at his entrepreneurial blog BrianLash.com (feed).
I looked at my Feedburner subscriber numbers today. Yesterday, too. And the day before.
If you maintain a blog I bet you did the same.
Then I remembered an interesting passage from a book I read a few months back titled Fooled by Randomness. In it, mathematician and stock broker Nassim Nicholas Taleb tells a surprising story of the role randomness plays in the stock market.
He writes, “A 15% return with a 10% volatility (or uncertainty) per annum translates into a 93% probability of success in any given year.”
He then shows these same numbers, but on a narrower time scale. Note that the following chart contains the same information as the previous statement (keep the 93% chance of success in the back of your minds):
Source: Taleb, Nassim N. Fooled by Randomness. New York: Random House, 2004. 65-67.
Remarkable. At the smallest increment (a second in this example) the probability of success that corresponds to a 93% chance of a 15% annual return is a mere .5002.
Half. Well, about that.
Let’s take a step up to the day increment. Using a 365 day calendar year, this implies an expected 197 pleasurable days relative to 168 unpleasurable ones.
Consider what this means for our subscriber numbers (which, like stock portfolios, are a function of human behavior. Do you get upset when your numbers are up one day, down the next? So do I. But this example illustrates that daily changes don’t mean much of anything in the grand scheme of things.
We can still have a 93% probability of, say, a 15% increase in subscribers by “succeeding” just a little more than half the days in a given year.
The lesson: Don’t worry so much about your subscriber numbers. They don’t mean nearly as much as we’ve assumed they do.
