Top 8 prediction (final)

Name Song WNTS DialIdol VoteFair Not-safe Probability
Devin Velez The Tracks of My Tears 45 0.302 2 0.602
Burnell Taylor My Cherie Amour 45 0 4 0.567
Lazaro Arbos For Once In My Life 26 2.813 4 0.536
Janelle Arthur You Keep Me Hangin’ On 80 1.677 10 0.380
Amber Holcomb Lately 65 4.789 13 0.302
Kree Harrison Don’t Play That Song 73 1.407 21 0.235
Candice Glover I Heard It Through the Grapevine 76 5.808 17 0.228
Angie Miller Shop Around 38 2.580 31 0.150

(Compiled with preliminary Dialidol and Votefair numbers at 12:24 AM EST. I will update in the morning.)

Updated 8:49 AM EST. As in previous weeks, Dialidol seems to have stopped updating after midnight on the east coast. Votefair is extremely noisy, so Burnell and Devin switched places. But those two are so close that I wouldn’t read anything into it. It’s essentially tied between Burnell and Devin. Janelle Arthur was moved into safe territory.

Updated 1:40 PM EST. Final update. Devin is now back as the top pick, as Votefair’s statistics has gotten better. I could keep updating all day, but I think I’m just going to freeze the prediction here.


The methodology for the finals model is described here. 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.

Original post:

Names in green are predicted safe. Names in red are the most likely bottom 3 (though the model is not necessarily sure of them). Names in yellow are undecided. Any person not in green is not considered safe by the model. The most probable bottom 3 is Devin, Burnell, and Lazaro. However, anybody on the list being in the bottom 3 would not be shocking.

There still hasn’t been any real movement in the Top 3, so far as the model is concerned. While Angie turned in a (perceived) sub-par performance, she is still popular. Candice and Kree follow, a bit closer than they have been.

The model is on the fence about Janelle and Lazaro. If I weren’t purposefully declaring a Bottom 3, the model would have both their names in yellow, meaning it is not sure. It is reasonably sure about Burnell and Devin. For Burnell, it is confident at the 87% level, but it is confident of Devin at the 95% level (meaning his being safe would be historically quite unusual).

Janelle received quite a large boost on her WNTS approval rating; however, her numbers on Dialidol are still not great. As the year goes on, Dialidol becomes a better predictor than WNTS and Votefair, and as such those scores are being given less weight week by week.

I honestly thought Lazaro did better than his numbers reflect, but I’m obviously not representative of the voting public (I thought Angie was really good). Devin and Lazaro have already been in the bottom 3, which would make one suspect it could easily happen again. Burnell, though, has consistently outperformed his numbers, so I’m personally a bit skeptical of the model’s call.

Bookmark the permalink.

Comments are closed.

Comments are closed