American Idol has been a very popular show for quite a long time. The Sheconomist is a regular viewer and so I’ve seen many many episodes of the show. This week was the first time in the season in which the viewers can vote on the contestants they’d like to keep on the show. If you aren’t familiar with the format, up until this point the judges have been the deciders. Now it’s the viewers who pick.  Every week one or more people will be removed until a winner is decided.

On Tuesday and Wednesday night all the contestants sang and were then offered feedback by the judges. Overwhelmingly on the first night there seemed to be lots of favoritism among the judges. Not to specific contestants, but to every contestant. No one could do anything wrong, despite a number of weak performances.

I then became curious about the motives behind the judges. The judges were all too nice, so were they trying to manipulate anything? The success of the show depends on the viewer becoming attached to a contestant. The buy-in is built with exposure, life stories, and talent. But since this is the first night of voting I thought the judges were trying to build of the fan base as much as possible by being nice to all the contestants. But the second night the judges acknowledged they have been a bit too nice and were closer to providing appropriate feedback on Wednesday night.

So it’s unlikely the judges had been instructed to be nice to the first group and not the second, but it is clear the judges feed off each other. The opinions are not likely to be unique or unbiased, but maybe that’s never the point.

And that doesn’t mean the show still isn’t rigged

Thankfully I’m not the only person that’s noticed the potential for the show to influence the outcome in ways other than judge feedback. And thankfully some of those people also have the research to test some hypotheses.

We’re trained from the very beginning of the show to be anchored to certain contestants more than others. Sometimes it’s by luck. For example, there is a contestant named Phillip Phillips. Even if my real name isn’t Philip, I call myself that on this blog and in official Weakonomics business so I see and say this name regularly. He is likely to be the candidate I remember the most because of his name. But the show will go a few steps further to make sure I get attached to him.

If the show favors Phillip, they will give him additional exposure. Put him in more cutaway scenes, edit his life story for more drama, and just all around show him on camera more. Not only will this prime me to remember him, it will give others the opportunity to latch on to him as well. Everyone remembers a star for a reason and the more exposure they get the more opportunities for viewers to find something to latch on to.

And as it turns out there are a few ways in which the show may be showing favoritism and influencing outcomes. During the first part of the season when contestants try out and compete for a spot in the final group they are among hundreds and thousands of people. With 24 finalists it’s difficult to give everyone equal time in the show leading up. And the more time on camera during this initial stage a contestant gets, the more likely they are to stick around through the episodes where viewers get to vote. It stands to reason that greater exposure after the voting episodes start continues to help that contestant.  Are producers picking favorites early on?

It doesn’t stop with the stories either. The order in which a contestant performs also shows benefits. If you’re playing a statistical game and the producers ask you when you want to go in the order of all contestants that night, you want to go last. Going in the first quarter of the show will more than double your chances of getting voted off compared to the last quarter of the show. This is called the recency effect and it’s alive and well in American Idol. You want to go last.

There’s also a strong correlation with judge feedback and contestant mortality. This could be without causation though as the audience may see the song the same way judges do without their input. It would be nice to test this, but for now it would be safe to bet on the contestants the judges fancy most.

I don’t vote on shows like this, but I do watch.  And if you happen to be the voting type, you might be primed now for Phillip Phillips.

This is just a talent show, and perhaps not worthy of such an academic study of influence and voter behavior. But given that this is an election year, the parallels are quite easy to make.

Read: Priming The Idol Agenda

categories: business, media