Sunday, December 14, 2008

Cordle-Scott Band at Brendan's Pub

After studying "crime analyst" stuff all day 'til my eyes glazed over (I'm gonna incinerate that exam on Tuesday), I went to hang with Brad, Tim, Dean, Rick, and Robert for a while last night at Brendan's in the Orleans. One solid lineup. It was slow, though. Beyond the usual seasonality, it's a sign of the times, I guess.


We can also call this the "Cautionary Cold Water Report," LOL...

So, after posting the update about the Hot Club of Las Vegas's new CD sales site with CD Baby in the prior post, I looked at the CD Baby "About" page. Saw the following:
  • "261,691 artists sell their music at CD Baby"
  • "4,957,710 CDs sold online to customers"
  • "$95,753,753 paid directly to the artists"
OK, Big Numbers are indeed exhilarating. And, $95.7 million is a Big Number in my quasi-impoverished worldview. But, let's do just a tad of instructive arithmetic (taking, for the moment, their numbers at face value).

That's an average of 19 units per artist, and an average paid to each artist of $365.90.

Don't order that tour bus just yet. In Statistician-speak, this is known as your "expectation."

Then something else caught my eye. $95,753,753 (!) paid to artists divided by 4,957,710 CDs sold equals, on average, $19.31 paid to artists per CD. So, given, that I have no life (and was for a while today tired of cramming for the Crime Analyst gig), I did a quick informally-random sample of their catalog (n=50 CDs). Here's what I found, rank-ordered from the cheapest to most expensive, and drawn from a number of their "genre" categories:
  1. $7.99
  2. $9.99
  3. $9.99
  4. $9.99
  5. $9.99
  6. $10.00
  7. $10.00
  8. $10.00
  9. $10.97
  10. $10.99
  11. $11.00
  12. $11.95
  13. $11.97
  14. $11.99
  15. $11.99
  16. $12.00
  17. $12.97
  18. $12.97
  19. $12.97
  20. $12.97
  21. $12.97
  22. $12.98
  23. $12.99
  24. $13.50
  25. $13.97
  26. $13.99
  27. $14.00
  28. $14.00
  29. $14.00
  30. $14.00
  31. $14.00
  32. $14.00
  33. $14.97
  34. $14.97
  35. $14.97
  36. $14.99
  37. $15.00
  38. $15.00
  39. $15.00
  40. $15.00
  41. $15.97
  42. $16.00
  43. $16.95
  44. $17.97
  45. $18.00
  46. $20.00
  47. $20.00
  48. $20.00
  49. $34.95
  50. $50.00
OK, so I find an estimated mean (arithmetic average) gross sale price of $14.74 with a "median" (interpolated halfway between lowest and highest) at $13.98 (meaning the two really expensive ones at the bottom pulled the average up a bit; a bit of mild skewing).
  • Note for stats heads: the "Std Error of the Mean" (in light of a "sigma" of $6.51) here is only ~93 cents (corrected for sample size at n-1). Meaning, were you to randomly re-sample another 50 product prices, the average is not likely to move all that much. Maybe a buck or less either way "on average" -- in relative terms, about 6.3% expected mean price variability (and certainly nowhere close to an average of $19.31).
The question obtains: How can they assert to be paying an average $19.31 per CD to the artist when their own gross average sale price is only $14.74? Can downloadable mp3 file sales account for all of the difference? (To be fair, it must be noted that using averages and "expected values" can be misleading. Most peeps don't get this. That's, uh, why there are 2 million of us here in Vegas. Otherwise this place would be Barstow II.)

Just a question. Not imputing any bad motives. Just that the math doesn't appear to square. NOTE: a sum is exactly calculated mathematically by multiplying the average times the number of items -- in this instance, $14.74 avg price x 4,957,710 units, for a gross of only $73,059,789.19. Again, perhaps mp3 download sales account for the difference. Dunno. $22+ million in mp3 sales? (And, of course, begging the question of their putatively adequately profitable cut of any overall gross.)

Yeah...Don't lease that tour bus just yet, LOL.

No comments: