Rating The Conventions

On my showbiz podcast BOX OFFICE WEEKLY I covered the TV ratings for both conventions, and they were historically high. Obama held the record for most-watched acceptance speech, that is until McCain took the record by another 500,000 viewers. Palin was pretty close too. Biden, not so much.

But there's all kinds of fascinating qualifiers for these ratings, and they add up to nothing. For example, Obama's speech wound up in the top ten for the week while McCain's did not. It's a weird fluke of statistics - the ABC feed made the numbers, but the speech was carried live on 11 networks. A few more viewers on CNN or Fox would have meant less on ABC. McCain's speech was carried on 8 networks (BET, Telemundo and Network One declined on McCain but carried Obama) but still got larger overall numbers, but none in the top ten. I think he was edged out by sports on Sunday.

McCain may have also benefitted from a couple of good lead-ins - there was a football game on just before his speech. Had it run long or McCain started earlier, it probably would have hurt him but as it is, people are already in front of the TV anyway, right? Also Bill O'Reilly sported the first night of a multi-night interview with Obama, which garnered O'Reilly his second-best ratings ever. This preceeded the McCain speech, so arguably you couldn't ask for a more appropriate prelude. On the other hand, Obama had the advantage of being first, so he had to do less battle with convention fatigue.

And interestingly, the Neilsen company picked the month of September to add about 100,000 more people to its sampling pool; so they were sampling differently between the Democratic and Republican conventions. Does this skew the numbers any? Not really. More people in the sample presumably means a more accurate sample because there is more data, but it doesn't fundamentally alter the conclusions that Nielsen reaches.