Monday, February 10, 2014

Analysis of commercials during NFL playoff

I don't know much about advertising. My work is not related to it at all but I still find this article very interesting (and the sarcasm makes it more fun to read) The strangest thing is: I got the link and summary of the article from the daily newsletter I receive from MLB! I guess the guys at MLB think it's a good idea to trash another league/sport. Basically, the article pointed out not only that the whole 3-hour broadcast of a NFL game actually contains 11 minutes of real action (from analysis by Wall Street Journal), but also a really high amount of commercials (112 on average and 60% of the broadcast) per game!

So, what are these commercials? Beer is probably the first thing came to my mind as beer commercials are very high profile and generated a lot of conversations. Well, it turns out there are not as many as I thought: only 8%. Car commercials double it at 16%. And there are even more tech product related commercials (18%. Verizon is the top company that pays for ~50 ad per game!)

Talking about beer commercials in NFL, I recently read an analysis from Charles Wheelan's book "Naked Statistics: Stripping the Dread from the Data" about the famous 1981 campaign in which one beer company (Schlitz) challenges supporters of its competitors to do a blind tasting test. One would think it's tough to produce an outcome that makes Schlitz looks good. However, if you understand the statistics behind it, it's really a gimmick. From the book:
Let’s assume that Schlitz would have been pleased if at least 40 of the 100 tasters picked Schlitz—an impressive number given that all of the men taking the live blind taste test had professed to be Michelob drinkers. An outcome at least that good was highly likely. If the taste test is really like a flip of a coin, then basic probability tells us that there was a 98 percent chance that at least 45 of the tasters would. In theory, this wasn’t a very risky gambit at all.
So what happened to Schlitz? At halftime of the 1981 Super Bowl, exactly 50 percent of the Michelob drinkers chose Schlitz in the blind taste test.
There are two important lessons here: probability is a remarkably powerful tool, and many leading beers in the 1980s were indistinguishable from one another.