Top 12 forecast and Twitter sentiment

Probabilities are based on a bottom 3. Multiply by 10/9 if there is only a bottom 2.

Name WNTS Rating Popularity Order Safe Probability
Clark 79 29.305 9 > 98%
Quentin 69 13.87 11 > 98%
JAX 75 11.37 7 97.7%
Joey 59 8.87 10 97.2%
Tyanna 75 10.01 5 91.3%
Rayvon 74 6.615 2 84.3%
Nick 48 5.4 6 83.8%
Adanna 43 2.665 12 78.1%
Qaasim 27 1.55 8 70.9%
Maddie 43 4.42 4 50.4%
Sarina 33 4.795 1 25.5%
Daniel 9 1.13 3 24.5%

Green means projected safe. Red means projected in the bottom 3. Yellow means too close to call.

How did the stars align to put Sarina-Joi projected to be in trouble? Her performance was poorly rated, she is female, she sang at the top of the show, and her numbers on internet polls has cratered. It’s a bit of whiplash, but I can definitely see it happening.

Quentin is again called super-safe, which historically would be right-on, but Quentin is a bit of an odd-duck. He was called super-safe in the last round, and in fact was not voted through. True, he may have been in the Top 6 (we weren’t told), but nevertheless that’s not very confidence inspiring. I’m leery of the call.

As for Twitter, they also hate Daniel:

TwitterSentiment2015-03-12Jax, Tyanna, and Nick hold the top spots, with Joey and Qaasim garnering quite a few negative comments.

Top 16 forecast (final)

Green means predicted safe (outside of the margin of error). Red means predicted eliminated. Yellow means too close to call. Methodology is here.

Update: Joey misses the cut-off, and is too close to call. Adanna gains and is predicted safe. Alexis is no longer called eliminated, she is too close to call.

Men:

Name Approval Rating Popularity Order Chance of advancement
Clark 58 24.5 5 > 98%
Quentin 66 19.6 8 > 98%
Qaasim 58 15.3 7 95.6%
Nick 57 14.7 6 94.4%
Rayvon 62 13.3 3 91.3%
Adam 32 3.2 4 44.6%
Daniel 17 6.5 1 43.2%
Mark 18 3.2 2 31.9%

Women:

Name Approval Rating Popularity Order Chance of advancement
Sarina 85 34.2 5 > 98%
Tyanna 64 18.6 1 > 98%
JAX 56 19.2 7 > 98%
Adanna 55 6.6 6 79.5%
Joey 44 8.7 4 78.0%
Maddie 42 6.3 3 64.9%
Loren 36 3.7 2 46.4%
Alexis 7 2 8 34.9%

On the men’s side, I’m not confident calling anybody gone for sure. Mark seems like the best bet: he was early in the show, lazy as hell, and doesn’t have a young girl fanbase. The front-runner men, Quentin, Clark, Qaasim, and Rayvon seem like locks.

As far as the women, It’s hard to see a world in which Sarina-Joi and Tyanna aren’t advanced, and Jax being eliminated would be a stunner. Joey is conceivable, though. Alexis is really far down, and I’m comfortable calling her gone for now.

Top 24 forecast (final)

NB: I forgot to say earlier, though regular readers know: green means predicted safe (outside of the margin of error). Red means predicted eliminated (outside the margin of error). Yellow means too close to call. I will clarify how this margin of error is computed in a later post.

Men first:

Name Approval Rating Popularity Order Chance of advancement
Quentin 84 22.5 10 98.2%
Clark 73 22.4 6 98.1%
Nick 75 15.5 11 98.3%
Qaasim 65 9.0 12 97.6%
Rayvon 62 7.0 7 80.8%
Daniel 30 5.4 8 63.3%
Riley 33 3.7 9 59.5%
Mark 41 3.2 4 43.9%
Savion 41 3.0 3 40.2%
Trevor 12 3.9 5 37.6%
Adam 33 2.6 1 29.1%
Michael 20 1.9 2 24.4%

Then women:

Name Approval Rating Popularity Order Chance of advancement
JAX 82 27.6 11 98.7%
Sarina 85 27.3 10 98.6%
Tyanna 66 14.6 12 98.6%
Joey 53 8.2 4 89.4%
Katherine 45 5.5 5 73.7%
Maddie 39 2.4 9 67.1%
Loren 43 1.6 7 56.4%
Shannon 14 3.9 6 50.3%
Shi 11 2.3 8 47.1%
Alexis 41 1.8 3 41.6%
Adanna 42 2.0 2 39.7%
Lovey 39 2.8 1 39.0%

We’re in uncharted territory, so I’m just making the best guess I can with the data I have to work with. Several facts to note, some of which I’ve said elsewhere:

WhatNotToSing has not published approval ratings for the Top 24 performances. I don’t know why. In place of those numbers, I’ve used numbers based on my sampling of 18 blogs for approval of performances. While it would be better to have a larger sample space, I have some confidence that these numbers shouldn’t change too much. The numbers did change by more than I thought, but it didn’t affect the rankings very much at all. Shannon moved one spot up, and Alexis dropped two places, which I am skeptical of. Mark and Riley switched places, and Adam fell one spot as well.

The number of respondents to the Votefair poll is pretty bad. In order to correct for this, I’ve averaged the results of three such internet polls, Votefair, TVLines, and MJsBigBlog polls. These together, in equal measure, account for the “Popularity” column. One complaint you might have is that this isn’t the right way to do this: the polls should be weighted according to the number of respondents. Well, I can’t. TVLine and MJs doesn’t seem to indicate how many people voted. So I’ve just made an assumption that they are all roughly the same.

The model here accounts for the order that the contestants sang in, which has been significant in the past. However, it’s not clear that the same significance is still there, given that voting is available as soon as the singing starts. Nevertheless, I’ve left it in, since to remove it is just as much an unfounded assumption as anything.

To be honest, nobody here would be a shocking inclusion. But there would be some shocking omissions. If we lose Jax, Sarina-Joi, or Tyanna, I would be very surprised, as well as Quentin, Clark, Nick, and Qaasim.

Note that Daniel F***ing Seavey is above 50%.

How to calculate these yourself:

The parameter estimates given the data linked to in this article lead to a probability for a given contestant

P =1/(1+EXP(-1*(-2.184291+0.1216*Order+0.017125*WNTS+0.184374*Popularity)))

You can use Excel to calculate this if you wish. The only adjustment made to these probabilities is that the sum of them must be 8, since 8 people will advance. Normally I would just take the sum of all probabilities and multiply each value by 8 divided by this sum, but there’s a problem: some of the contestants will then have a probability greater than 1, which is not allowed. Therefore, you must run a procedure (in whatever language you want)

p = p*8/sum(p);
while sum(p > .99) > 0
    p(p > .99) = p(p > .99) - .01;
    p = p*8/sum(p);
end

or similar. Doing this will get you the above probabilities. If you want to adjust the numbers, have at it.

Top 2 projection (updated)

Name Song WNTS MJs VF Not-safe Probability
Caleb Johnson Dream On / Maybe I’m Amazed / As Long As You Love Me 57.667 61.7 32 0.585
Jena Irene Dog Days Are Over / Can’t Help Falling in Love / We Are One 65.333 38.3 68 0.415

Important!

The methodology for the finals model is described here (though some modifications have been made to replace Dialidol with MJsBigBlog’s poll). The model is 87% accurate on ranking within a margin of error of +/- 3%. Probabilities being what they are, somebody with a not-safe probability of just 0.25 will be in the bottom 3 one out of four times. Please do not comment that the numbers are wrong. They are probabilities, not certainties or even claims. Do not gamble based on these numbers.

Name in green is most likely to be safe. Name in red is considered most at risk for being eliminated. The most probable elimination is Caleb. However, no result would be shocking.

The contest is far from a runaway for Jena Irene, but it certainly doesn’t look bad for her.

No matter what happens tomorrow, we will see some Idol history made. Continue reading

Top 3 projection (updated)

Name Song WNTS MJs VF Not-safe Probability
Caleb Johnson Never Tear Us Apart / Demons / Dazed and Confused 37.333 48.4 24 0.531
Alex Preston Pompeii / Stay / Story of My Life 57.667 39.9 33 0.349
Jena Irene Titanium / Heart Attack / Creep 53.667 11.7 44 0.120

Important!

The methodology for the finals model is described here (though some modifications have been made to replace Dialidol with MJsBigBlog’s poll). The model is 87% accurate on ranking within a margin of error of +/- 3%. Probabilities being what they are, somebody with a not-safe probability of just 0.25 will be in the bottom 3 one out of four times. Please do not comment that the numbers are wrong. They are probabilities, not certainties or even claims. Do not gamble based on these numbers.

Names in green are most likely to be safe. Names in red are considered most at risk for being eliminated. Names in yellow are undecided. The most probable elimination is Caleb. However, anybody being eliminated would not be shocking.

Updated 12:07 PM EST.

Caleb had a rough night, having contracted bronchitis, and resulting in a very low (for him) average performance approval rating, and also giving him the lowest approval rating for a song, a 12/100 for Demons (which the judges said was good???). As least popular on Votefair, and slightly favored on MjsBigBlog to be eliminated, he is the best bet.

Alex is ahead of Caleb by about 10 percentage points. He had the highest approval rating average of the night, and is slightly more popular on Votefair. Jena is running significantly ahead of both men on MJs and VF, and slightly trails Alex on WNTS, though she did have the highest rating of the night on her reprise of Creep.

I will update if the numbers change significantly.

The Top 3, historically, has not been very hard to call. Huge surprises, such as Melinda Doolittle’s elimination, have happened, but more often we see contestants like Nikki McKibbon, Jasmine Trias, and Syesha Mercado, who were overdue for elimination. There does not appear to be any reason to think there is a gender bias in this round, although the sample set is only 9 (four Top 3s were monogendered). I would say of all the years, this is one of the closest, with Elliott Yamin vs Katharine McPhee being the hardest of all to call.

Below is the updated performance for the season. All ranges are within tolerance, though 50-60% is barely so (it tends to understate how vulnerable those people are).

S13ProbsvsResults