A friend recently sent me Malcolm Gladwell’s Outliers, which is currently the #1 New York Times nonfiction bestseller. A few chapters into the book I came across some surprising statistics that aroused the mathematically inclined skeptic in me.

In chapter 1, Gladwell argues that the January 1st age group cutoff in Canadian youth hockey leagues favors athletes with birthdays at the beginning of the year. He writes of a psychologist who noticed a pattern in the birth months of young, elite hockey players:

More players were born in January than in any other month, and by an overwhelming margin. The second most frequent birth month? February. The third? March.

The author argues that this advantage persists as athletes age, and in fact never goes away:

He looked at the composition of the National Hockey League. Same story. The more he looked, the more Barnsley came to believe that what he was seeing was not a chance occurrence but an iron law of Canadian hockey: in any elite group of hockey players – the very best of the best – 40 percent of the players will have been born between January and March, 30 percent between April and June, 20 percent between July and September, and 10 percent between October and December.

If only I had access to an easy-to-use database with which I could check this. Oh wait, I do. I went to Yahoo Sports and copied the rosters of all NHL teams into a spreadsheet, then loaded the data – 636 professional hockey players – into TrackVia. (You can check my data here.) Then I used TrackVia’s text formulas to split out the month of birth. Finally, I created a filtered view that only included Canadian players, and displayed that view in statistics mode. Here’s the result (try moving your mouse over the different sectors):

The results are nothing like what Gladwell described! The distribution looks pretty even. There are certainly some fluctuations – April and May are big, November is small – but the trend I expected is absent. In fact January and February are both below average, and October and December are right on par. If there were no bias, the average month would have 27 players, with a standard deviation (plus or minus) of 5 – pretty much what we see. A birth-month bias may well exist in Canadian youth hockey, but it appears to average out by the time players have reached the professional level.

What’s my point? I don’t really have one. Except that this is why I’m usually skeptical of simple explanations of complex phenomena. And any excuse to load data into TrackVia to slice and dice, and I’m all over it!