Google Inc. reads, indexes and searches through billions of Web pages. On any given day, eBay Inc.'s more than 100 million members are listing millions of items for sale. Yet when it comes to the law of large numbers, executives at both tech powerhouses have committed statistical misdemeanors.The law of large numbers says that the more times you measure something, the truer your results, and less they are affected by random variation. Flip a coin a million times, and you're likely to get heads half of the time (or very close to it). But flip that coin five times, and you might get heads four times, twice, or never.
But in corporate-speak, the "law of large numbers" has been misused as a catch-all explanation for slowing growth as companies mature: It's harder to maintain high growth rates from a larger base. Last week, Google finance chief George Reyes told a Merrill Lynch & Co. investor conference that the company is "getting to a point where the law of large numbers starts to take root." (The Online Journal's MarketBeat column pointed out that Mr. Reyes meant the law of diminishing returns.) And in a January appearance on CNBC, eBay chief executive Meg Whitman said, "Now, our businesses are getting larger and we will obviously face the law of large numbers, but we have actually changed the trajectory of the growth curve in our two largest businesses over the last three quarters." (She used the term correctly in a 1999 appearance on CNBC.) Several other executives have misused the term.
Misstating a statistics law isn't a federal offense, and lots of analysts make the same mistake when posing questions about disappointing earnings, in effect providing companies with an excuse for slow growth (reporters, including those at The Wall Street Journal, have also used the term incorrectly).
An eBay spokesman told me the term is "common corporate vernacular." Google declined to comment.
I found this part of his weekly column worth repeating, so I am.
The column, by the way, is only available via an on-line subscription and the column is not printed in the paper version of the Wall Street Journal. Carl Bialik's (the author) comment are always on point, well written and provide ever more evidence of how we, (the royal "we") often misuse numbers in our daily lives.
No comments:
Post a Comment