Top 11 forecast (final)

Updated 2:52PM on 3/15 with new VF numbers. Popularity consideration now 80% Votefair, 20% TVLines as Votefair’s total number of votes is almost 200 now. Quentin’s raw probability (without normalization) is now slightly below the cutoff. No change in rankings. Daniel’s probability falls to 50% if he was in the Bottom 3 this week (but there’s no way to know).

Green = predicted safe, Yellow = too close to call, Red = bottom group. Probabilities assume a bottom 3. Note that the show did not reveal the bottom 3 on Thursday. This is crappy and the show should start revealing that.

Name WNTS Rating Popularity Order Safe Probability
Tyanna 78 18.8 8 > 98.0%
Clark 64 20.9 4 > 98.0%
Nick 50 11.0 11 > 98.0%
Joey 88 15.5 3 > 98.0%
Quentin 44 6.5 10 98.0%
JAX 35 12.9 5 88.2%
Daniel 19 1.7 9 82.4%
Adanna 58 3.7 7 69.5%
Qaasim 21 2.9 6 69.3%
Maddie 46 4.2 2 56.3%
Rayvon 31 1.4 1 45.0%

I posted the methodology here. Probability is calculated by considering order of song, WhatNotToSing rating, “Popularity”, gender, and whether or not the person has been in the bottom 3 before. The only people currently recorded as having been in the bottom 3 are Quentin and Adanna, who were Wild Card picks. This is because the producers did not tell us who was in the bottom 3 on Thursday.

Note that if a person was chosen to be in the bottom 3 at random, the chance of being safe would be 73%, quite high.

Daniel falls about 6 points since the initial post. If you are confident that Daniel was in the bottom 3, his probability falls to around 68% 50%, but still can’t be assigned in the bottom 3. The model is not confident enough to assign him to either category of people. Rayvon is the most likely, since he sang at the beginning of the show, was the least popular, and had a low WhatNotToSing rating. By contrast, Daniel sang near the end of the show, and is a bit more popular. Song order is quite significant. Two singers are now below the cutoff: Maddie and Rayvon. Maddie barely there.

Jax rises 2 places, as her numbers are far higher than they were on Thursday in popularity, while Adanna falls out of the safe zone, and Qaasim also falls in ranking.

As the season goes on, I assume Votefair is going to start getting more voters. To help with the noise, I’m taking a weighted average of online polls, but Votefair is counted more than the others. Once Votefair gets to > 200 voters, I will just count Votefair by itself as popularity. Right now, with only 177 voters (excluding people who voted for Sarina-Joi, who isn’t even in the competition anymore), I’m not sure that it can yet be trusted.

Top 3 Tracker for the Top 11

The top 3 tracker is a look ahead to the end of the contest, assigning the chance that somebody will be in the Top 3 based on how they are doing now.

As you can see, there is a high amount of uncertainty at this stage. Even someone predicted super-safe like Clark has only a 40% chance of being in the Top 3. (Think Pia Toscano). Below is the historical information:

Hover your mouse to see who each dot corresponds to. Blue dots are contestants in previous years. Green dots that lie on the theoretical curve are this year’s contestants. Nobody making the Top 3 would be shocking, including Daniel Seavey (Haley Reinhart was considered even less likely than him at this point).

Singing order

In the methodology post, I mentioned that contestant Order was an important variable. In fact, it you correct for the other variables, you are significantly better off being near the end of the show versus the beginning of the show:

SongOrderVsSafenessPoints at the bottom are people who were not “safe” (either eliminated, bottom 3, or saved) in Final rounds with no-multiple performances. Points at the top are people who were Safe. The two curves are the probability of an average contestant (both male and female) being safe. All other things being equal, a man goes from a 65% chance singing first to a 95% chance of being safe at the end of the show. A woman goes from a 50% chance to about 85% if she is in the pimp spot.

I’ve idly speculated about why this was in the past. It could be that the producers put the better contestants at the end of the show for ratings purposes, and it could be that most people vote at the end and only remember the recent singers. Most likely it’s a combination of both of these things. The forecast model from previous years didn’t include this explicitly as a variable because it was no better than Dialidol at accounting for singing order. But now that Dialidol is kaput, it’s worth to consider explicitly.

Sarina-Joi’s elimination was not shocking (Finals model methodology)

I’m seeing a lot of surprise among the Idol blogosphere about Sarina-Joi’s elimination, and while I kind of understand it, I disagree. If you read the Top 12 forecast, you saw that I had Sarina-Joi second most likely to be eliminated, within a hair’s breadth (1 percentage point) of Daniel Seavey. After the jump, I’ll say how I get that.

Continue reading