Misleading Charts

One of my pet peeves is media publishing misleading charts. I saw a great example of this in the Dec. 2011 Ski magazine:



Visually it’s a striking chart. Look at those peaks and valleys--almost like mountains themselves! And look how much sales fell! For skis, in blue, it fell to half before rebounding. And for snowboards, it has fallen like a rock and is now at half the level it was just three years ago. But wait a second. The scale starts at 300,000, not zero. And another thing: the swoopy valley minimums have nothing to do with the data—they’re just there so the shapes look like mountains. One thing this chart did right was publish the actual data underneath, so I took it and re-graphed it in a more conventional format, below.

Reads differently now, huh? Now I see that ski sales fell during the recession and have returned to 2007/08 levels, whereas snowboard sales appear to be in a steady decline (sorry, knuckle draggers). But the decline is less than 20% over three years, or about 6%/year. But the minimum point for skis was not half the max, but down about 15%.

Now I admit that my chart isn’t as pretty as Ski’s and certainly not worthy of publication, but shouldn’t the point of a chart be to educate rather than confuse the reader?

Oh, one last thing. I know these are “sales,” but do they mean dollars, thousands of dollars, units? One great thing I picked up from McKinsey consultants was a strict method for labeling charts, including the units, and this is missing from the Ski graphic. But what Ski did right was to list the source, so I poked around the Snowsports Industries America website. At their media fact sheet they say that 2010/11 Equipment sales were nearly $653M, which is much lower than the 926,479 somethings from adding the ski + snowboard numbers from the chart together. After wasting considerable time on the website looking for the answer, I wrote to the PR Manager, who failed to respond. I guess we’ll never know. What a useless chart!

After all the talk about Groupon this year, many readers have been encouraging me to comment on Facebook’s IPO. At this point, I’m still on the fence. On the one hand, with 850M users and a market valuation of, say, $85B (solidly in the $75-100B range I hear discussed) that means FB has to extract $100 in value out of each of us. They do know an awful lot about us, but $100 is big money, not so much for wealthy American consumers, but substantially more for a subscriber in a developing country. Given that valuation is as much art as science, I think now is time to retreat into art, specifically poetry. And even more specifically, limerick. Consider this from the clever site Limericks Économiques:

   Facebook is valued at plenty
   By Wall Street’s high-tech
      cognoscenti,
   Based on 1 billion friends
   Times $5 each, then
   Times the IPO multiple, 20.

One of their readers responded:

   Facebook steals data from you
   And sells it marketers too!
   Mark wanted more dough
   So he went IPO
   Now it’s ads that you’ll have to
      wade through.