« The Mecca of Dubai | Main | Hollywood / Bollywood Doppelganger »

TiVo Predicts 'American Idol' (and Flatters Me.)

So, I was amused to see this press release from TiVo today. For those of who haven’t seen it, the release is titled “TiVo Predicts ‘American Idol’ Results Based on Highest and Lowest Viewership, With This Week’s Viewership Saying Syesha Mercado Will be Sent Packing Thursday Night .

Here’s a snippet from the press release that’s particularly interesting:

For past four weeks, since American Idol has picked its final 12 contestants this season, TiVo has been using its second-by-second audience measurement to determine which performers had the highest and lowest viewership. It turns out, this information, especially among viewers who are watching on a timeshifted basis, is predictive of which contestant will be voted off. Viewers tend to rewind and watch their favorite performances multiple times, while fast-forwarding through the ones they don’t like.

Almost two years ago, I wrote an article on my Brutal Clarity blog, titled “TiVo and American Idol Predictions”. Here’s a snippet from my own ramble:

In the heat of the continuing American Idol juggernaut, I had another idea. I caught myself doing something interesting during the performance last night. I wanted to skip the commercials, so I watch the show an hour after it has aired. I noticed that I would skip the performances I didn’t like, and I would re-watch the ones I loved. Often, my distaste for a particular performance was so instant, I would start skipping within the first 10 seconds.

Looking at my TiVo privacy policy, I saw that they allow for the aggregation of anonymous content to help enhance the service. If that’s so, why couldn’t they track the skipping and rewatching habits of 5,000 users who watch American Idol post airing, and use the aggregate data around each contestant to predict their potential position in the voting lineup?

Well, at the very least, I’m glad to know that my late night rambling a couple of years ago ended up being statistically accurate. I have no idea if anyone at TiVo actually read the blog article and then decided to run the test, but if they did, (and you’re reading this,) drop me an email, wouldja?

I promise I won’t tell. 

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (4)

This was a really admirable topic! I was luck to come cross your website in digg Also I get much in your theme really thank your article very much i will come later.replica dresses

November 26, 2010 | Unregistered CommenterDelores R. Ellie

I like your replica Swatch jewellery article.It is great and useful for me !

December 16, 2010 | Unregistered Commenterroboter

Pretty good article. I just came across your site and wanted to say that I have really enjoyed reading your blog posts. Any way I’ll be coming back and I hope you post again soon.thanks for sharing.

Replica Audemars Piguet

July 18, 2011 | Unregistered Commenterabbyabbie

This page sorted out all of Prada 2010 Collection my issues, it truly is very helpful. It's really a resource for that replica designer handbags learner. Thank you giving out information

October 12, 2011 | Unregistered CommenterPrada 2010 Collection

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>