Analysis after the jump.
Above you can see the change from last week to this week. Angie fell a nominal 3 points, while Candice and Kree made big jumps. Of course, just making it through to another round improves your chances of making the Top 3—even by random chance you would have improved from 0.333 to 0.375. The flip side of this is that your not-safe probability also rises for the same reason, since by random chance you had a 0.33 chance to be in the bottom 3 instead of 0.300. The fact that Kree and Candice didn’t increase much shows that they have real staying power. Both are now more likely than not to make the Top 3.
Amber declined somewhat, which is obviously rational since she was in the bottom 3. Janelle showed some improvement from her poor standings before, and is now roughly in line with the remaining men. Though the model was wrong about Burnell’s position this week, it is still justifiable to be skeptical that he will make the Top 3. That will change if he continues to be safe.
The Top 3 Tracker in this round (nominally round 5, so that it matches up with previous Top 8 years) makes use of an averaging scheme to provide a less noisy picture of the contest. For historical context, here are the average round 4 and round 5 not-safe probabilities plotted against whether or not the person made the Top 3 (if the dot is at the top (1) then he or she did make the Top 3, and if it’s near the bottom (0) he or she did not. Some vertical jitter has been introduced to make the dots visible.).
Hover your cursor over the points to see which past contestants they correspond to. The only people ever who did not advance with an average value of less than 0.3 were Siobahn Magnus, James Durbin, and Casey Abrams. The other 12 people below this threshold all made the Top 3. The only person to ever make it with an average above 0.4 was Syesha Mercado. Since Burnell is above this value (0.401), historically speaking he is highly unlikely to make it. The 18 other people above 0.4 did not make the Top 3.
The blue line is a regression line indicating the probability of a certain unsafeness resulting in a person making the Top 3. This is where the Top 3 Tracker’s numbers at the top of this post come from.
**Correction: this post originally showed the Round 6 graph by mistake.