UPDATE: After talking to the authors of the study, I found the disconnect. The original story linked below was incorrect. The share of TiVo owners doing the time-shifting was closer to 1/3 not over 2/3. Here are the numbers from the study courtesy of Starcom:

That makes much more sense. I can believe that TiVo owners would be 50% more likely to use their machines than the average DVR owner, but not 3x as much.

When I saw this story today claiming that more than 2/3 of TiVo owners time-shifted the last Super Bowl, I had to check our Nielsen data.

Here is Nielsen’s data for the DVR time-shifting of Super Bowl 41 (XLI for you Romans).

Nielsen TV Ratings Data For Super Bowl XLI, February 4, 2007:

Those ~1.9 million total time-shifted viewers are about 20% of the ~9 million DVR households that Nielsen estimated for last season. There is a significant disconnect between the claimed 66% and that 20%.

I just can’t believe that TiVo users behavior is so different than the rest of DVR owners, but the only other thing I can guess (other than simple error on somebody’s part) is that their definition of “time-shifted” is very, very different than the one Nielsen itself uses [and TVbytheNumbers uses as well].

Unfortunately, my schedule today prevented me from contacting the author. I will follow up and see what I can find out.

Nielsen Ratings Data: ©2008 Nielsen Media Research, Inc. All Rights Reserved.