Professor Ed Kaplan Explains How 'American Idol Voting Is Broken' in Op-ed

'Idol' Voting Is Broken; Phone System Limitations Skew Tally Of Contestants' Popularity
by Edward H. Kaplan
Published in the Hartford Courant on April 25, 2007

Now that "American Idol" is headed toward the finish line, our nation faces the weekly challenge of deciding who will be voted off the show.

Thirty-eight million votes were tallied before Sanjaya Malakar was sent home last week. While 38 million sure sounds like a lot, remarkable data from DialIdol.com reveal that well over half of all phone vote attempts met a busy signal as the poll began, while throughout the voting hours, more often than not the chance of encountering a busy signal was above 15 percent. And therein lays a problem: the votes "American Idol" counts do not reflect true voting preferences. But this is an easily fixed problem.

First, we must understand what busy signals really tell us. DialIdol.com is premised on the idea that more popular contestants generate more phone calls and consequently more busy signals.

So, tracking busy signals should reveal who the leading (and lagging) contestants are (the DialIdol data are generated from freely provided speed dial software that reports both attempted calls and busy signals back to the site). But the votes that count are those that get through, not those that are attempted. When so many votes are lost (which is what happens with each busy signal), the tallied vote depends much more on the number of calls that actually get through on the show's phone lines than on voter preferences.

Recall that each "American Idol" contestant receives his or her own dedicated phone number. To illustrate the problem, suppose hypothetically that each such number is capable of processing 4 million calls over the duration of voting. Now consider a random contestant, say Beat Box Blake.

If 6 million votes were attempted on Blake's behalf, only 4 million could get through, while one-third of the Beat Box Boosters would get busy signals.

Now consider Melinda Doolittle and her 10 million attempted votes. While 60 percent of Doolittle fans can do little but grouse over busy signals, about 4 million votes are registered - rather similar to Blake's count in spite of Melinda's much greater support among voters. As this example shows, busy signals and lost calls can turn lopsided preferences into a coin toss.

Considering past results - Reuben Studdard's victory margin over Clay Aiken was reportedly less than 1 percent in spite of tens of millions of votes tallied - along with currently observable busy signals, results to date could have been dictated by phone capacity constraints rather than true fan appeal.

There is a simple solution to this problem that does not require "American Idol" to install additional phone capacity. Instead of assigning each contestant a personal phone number, use a single number for all voting, and have voters select their favorite by pushing a button after the call has gone through. This simple fix would equalize the chance of encountering a busy signal for all callers.

Many votes would still be lost. In fact, if the total phone capacity was left unchanged, more calls would be lost due to the increase in processing time per call necessitated by button-pushing. But, since all calls would have the same chance of getting through, the total votes received would be a representative sample of votes cast.

The fix proposed is not perfect - it does not address voting by text — messaging (which is only allowed via Cingular, a show sponsor) — but it would correct a huge bias currently present in phone voting. "American Idol" should not be distorted by American Busy!

Edward H. Kaplan is the William N. and Marie A. Beach Professor of Management Sciences at the Yale School of Management.