American Idol - A Technical Critique of the Voting Results

Parent Cafe
1

<p>Last night, as Elliot was sent home and Katherine and Taylor moved on the the finals, the results were touted as the closest ever, in a narrow range from something like 33.06% to 33.68%. There were also a record number of votes: over 50 Million. How is that possible? How could the USA be so evenly divided among the last three contestants.</p>

<p>Actually, the explanation is quite simple. With the use of autodialers from computers (get yours today at dialidol dot com, which also has correctly predicted the loser almost every week, based on the percentage busy signals they receive during voting), and the popularity of the show, it’s easy to predict that the usage of the voting lines is very close to 100% of capacity. Thus, if the three contestants have equal capacity on their phone lines, the votes almost HAVE to be equal (within a small statistical variation).</p>

<p>Problem is, when the usage is 100% on the lines, the vote then becomes meaningless.</p>

<p>Assume that a “line” can support one vote per second and each contestant has the equivalent of 2400 “lines” of capacity. If that capacity were used at 100%, that would mean 2400 votes per second per contestant, or 8.6 million votes per hour (3600 secs/hr * 2400 votes/sec). The voting is open for 2 hours, so that translates to about 17 million votes per contestant for the night. For the three contestants, that translates to a total of a little over 50 million votes total.</p>

2

<p>i vehemently think that only one vote should be allowed per phone number. why should i, as someone who doesnt have time to sit for 2 hours pressing redial, be counted less than someone who votes 100 times? you cant vote for politicans more than once (thank goodness) and i think the results would be much more accurate that way instead of penalizing someone for not calling 100s of times.</p>

3

<p>I think it also reflects the lack of a true standout in the bunch. But your technical points seem valid assuming the lines are always full for all. That probably would not happen if one truly had a bad day.</p>

4

<p>Don’t forget that one can also vote via cell phone text messaging. That’s another part of the puzzle that can’t be overlooked.</p>

5

<p>And they will never limit that because they get too much revenue from the messaging (see the small print on the screen? “Standard messaging rates apply”).</p>

6

<p>I don’t think Einstein had anything on digmedia :D.</p>

<p>And I’m not so sure relativity is all that important compared to integrity of American Idol voting, either. :p</p>

7

<p>Does this mean the texters are actually making the decision? (I am still waiting for Taylor to sing “Touch of Gray”.)</p>

8

<p>

</p>

<p>In part, yes. But, that is just part of the fact that the contestant with broader, deeper, more energized support is still going to win. The saturated phone voting system just means that the contest will appear “close”.</p>

<p>Let me give an example:</p>

<p>Suppose for the sake of argument that Taylor and Kat both registered 15 million votes on the saturated system the other night.</p>

<p>DialIdol’s sample of 800,000 calls showed that Taylor’s lines were 80% busy nationally meaning that only 20% of the dialed calls got through.</p>

<p>The sample showed that Kat’s lines were 70% busy, meaning that 30% of her calls got through.</p>

<p>For Taylor, 15 million divided by .20 success rate means that 75 million calls were dialed for him.</p>

<p>For Kat, 15 million divided by .30 success rate means that 45 million calls were dialed for her.</p>

<p>That is a huge disparity in support – support that also translates into willingness to pay 10 cents a vote for text messaging, organized get out the vote campaigns, etc. </p>

<p>So, while the announced results will always be close, there is no way that Kat could have generate the additional 30 million dialed calls necessary to tip the saturated stats just a few tenths of a point.</p>

<p>BTW, the east coast is horribly saturated. Dial Idol on my computer dialed 748 calls in the two hour period. It got busy signals on all but 5 of those.</p>

<p>The volume of calls is such that there is nothing Idol can do except extend voting hours. That will play into the hands of the contestant with broader, deeper support.</p>

<p>I think this one is foregone conclusion before the final, just like it was last year. In both cases, one contestant has dominated the voting over the course of the entire season. It is in Idol’s best interest to saturate the phone system and make the vote look artificially close. However, it is not in their interest to saturate it to the point that a full season of support is invalidated. After all, they know the numbers, week after week after week. There’s nothing in it for them to have a less popular winner.</p>

9

<p>ID: Good analysis; that’s why the results are close, even if many more calls are dialed for one of the contestants and why the results seem consistent from week to week (Taylor always safe). Interesting to see the busy percentages and how the numbers run.</p>

10

<p>You can look up the sampled busy percentatges for each week at the <a href=“http://www.dialidol.com%5B/url%5D”>www.dialidol.com</a> website.</p>

<p>Taylor was 1st for two of the three weeks when there were just boys. The other week, he was second with huge gap back to third.</p>

<p>He was first at Dial Idol every week after that, although he was tied with Chris one week. That was the week that Kat jumped up and knocked Taylor down to third – although I think the top three were all essentially tied.</p>

<p>Most of those weeks, Taylor had a significant edge in measured busy signals. For example, the other night he had a 10% advantage over Kat or Elliot.</p>

<p>The statistically close results from Tuesday night will ensure no complacency in his get out the vote effort.</p>

11

<p>I HATE American Idol with a passion. What a complete waste of air time. A bunch of people being made to sing all styles of music regardless of whether their talent is suited to those styles, only to be judged by three hacks. Who cares what Paula Abdul thinks? I mean why don’t they just require all the contestants to only sing Phish songs, that would take some talent.</p>

12

<p>They aren’t judged by three hacks. They are judged by millions of hacks – the American public.</p>

<p>What’s been fascinating is that the winner selected by the American public almost never fits the mold of a contemporary top-40 “pop star”. I mean…this year’s winner sang two Sam Cooke songs from the 1950s on the show. That’s pretty cool. </p>

<p>Part of it is that the demographics of the audience skew older – just as the demographics of the CD buying consumers are skewing older.</p>

<p>Part of it is the lack of viable well-defined “top-40” product at the moment. </p>

<p>I think it’s actually a fun show. Cuts across generations. Fairly innocuous family-oriented entertainment (no machine gun killings). Generates a lot of water-cooler interest. Gives viewers a taste of a wide range of musical styles. I can think of worse shows.</p>

13

<p>The week that Kellie Pickler got voted off, my 80-year-old father could get no more than about 12 votes through over the entire 2-hour voting period due to busy signals. He had been averaging over 90 calls per Tuesday night throughout the earlier parts of the competition. After talking to other Pickler fans (she’s from our hometown), he heard the same story. It was like there was something vastly different about dialing for everyone that night. It seemed as if the lines for her were “blocked.” </p>

<p>Of course, she never looked at any of it as “fishy” or controversial, but admitted that it was “her time to go.” We did not think she had the most outstanding voice at all, but we surely did enjoy the way our small town “pulled together” every Tuesday night. It was a boom for the local businesses, and it really unified the entire county.</p>

14

<p>Huh???</p>

<p>Why should I watch that show when I can waste my time with this:</p>

<p><a href=“http://talk.collegeconfidential.com/showthread.php?t=187878[/url]”>http://talk.collegeconfidential.com/showthread.php?t=187878&lt;/a&gt;&lt;/p&gt;

15

<p>The week Pickler got voted off, her “busy percentage” in Dial Idol’s national sample of 340,000 calls dropped to 48% from the mid-60% range for the two weeks prior.</p>

<p>Your grandpa’s busy signals were localized. The bottle necks can occur at the local level, regional level, or national level. It helps to have broad national voting support rather than having all of the support concentrated in one region – especially an Eastern time zone region.</p>

<p>Pickler got hit with a double-whammy. The “Ellie Mae Clampett by the cement pond” act she had been milking wore thin. And, she demonstrated several weeks in a row that she really wasn’t a particularly good singer outside of a fairly narrow “rock/country” vein. Her version of the Patsy Cline classic, “Walkin’ After Midnight”, was dreadful.</p>

16

<p>I live in rural NC. Kellie was a huge deal to our town.</p>

<p>You have to understand that our small town held huge block parties in the downtown area every Tuesday night while Kellie was on AI. There were gigantic TVs set up outside, a local millionaire served FREE food, all of the restaurants featured specials like free “Kellie’s Calamari” and fried pickles. The local bakery baked decorated cookies that said “Pick Pickler.” T-shirts saying “Naughty Little Minx” and other slogans for her were sold AND worn. The local phone company even provided a large area of phones for the public to use to vote after the show. </p>

<p>Then, when she had her homecoming parade, 12,000 people showed up in our little town from all over the country!</p>

<p>No, I did not attend the block parties nor her parade, but I surely enjoyed the excitement over Kellie each week with my co-workers and friends.</p>

17

<p>I believe it. I think the hometown (or home state) support for the contestants is fun and positive for the communities. The daughter of the Boston TV news anchor and State Representative got a lot of media attention here in Massachusetts, but didn’t make it far enough for things to go nuts.</p>

<p>Judging from the video posted all day on the Fox affiliate there, Birmingham went absolutely nuts last Friday for Taylor Hicks’ visit. Scary nuts. Easy to see why “he is the odds-on favorite” nuts.</p>

<p>Seems like good, clean fun to me. I can see why someone wouldn’t be interested, but it’s hardly the “worst” thing to come out of network TV these days.</p>