After every elimination episode, you can now predict the elimination order of the remaining contestants.  At the bottom of the Rate Performances page, there is a form where you can order the contestants from first to last.

Here’s an updated list of our Accumulated Performance Order metric.  To refresh your memory, the higher the score, the more often the contestant has performed towards the end of the show.

1. Megan (Joy Corkrey) – 24 points
2. Matt Giraud – 24
3. Adam Lambert – 24
4. Anoop Desai – 24
5. Danny Gokey – 20
6. Allison Iraheta – 19
7. Lil Rounds – 12
8. Scott Macintyre – 11
9. Michael Sarver – 11
10. Kris Allen – 11

Poor Kris Allen always performs near the beginning of the show, at which point we’ve almost already forgotten his performance.  Question: do you think APO or performance order matters?

As of this writing, here are the top rated performances of the week.  All ratings out of 10.  Let your voice be heard by putting in your votes here!

1. Danny Gokey – 7.4
2. Anoop Desai – 7
3. Matt Giraud – 7
4. Kris Allen – 6.1
5. Allison Iraheta – 5.7
6. Adam Lambert – 5.2
7. Lil Rounds – 5.1
8. Alexis Grace (eliminated) – 5
9. Megan Joy (Corkrey) – 4.4
10. Scott Macintyre – 4.1
11. Michael Sarver – 3.6

Up-to-date ratings can be found here.

One metric I want to track is something I call “Accumulated Performance Order” (APO).  Basically, this sums the performance order of the contestants over the weeks.  The higher the number, the more often you’ve appeared in the back-half of the performance order, and vice-versa.

After 2 rounds of the Finals, here are the APOs (out of a possible 24 points):

1. Matt Giraud – 23
2. Megan Joy (Corkrey) – 20
3. Alexis Grace – 20
4. Anoop Desai – 17
5. Adam Lambert – 16
6. Danny Gokey – 11
7. Kris Allen – 9
8. Allison Iraheta – 9
9. Scott Macintyre – 8
10. Michael Sarver – 5
11. Lil Rounds – 5

Matt Giraud got the last spot this week, and the second-to-last spot last week.  I don’t think it’s a stretch to say that being in the last spot is very advantageous, since you can remember the performance very clearly as soon as voting begins.

Worst of all is the first spot of the night, whose performance you can barely remember (unless it was amazing).  Is it possible we say goodbye to Michael Sarver, given how early he’s been appearing on the show?

• Matt Giraud, very impressive!
• Kris Allen, very nice as well.
• What happened to Megan’s Corkrey?  Also, did they alter her voice, b/c she sounded different tonight.

Two growing trends that I did NOT like:

1. Judges suggesting what songs the contestants SHOULD have sung.  I’m sure that choosing a song is a very difficult process, given a) they can only sing songs with permission from the artist/label b) they’re fighting with 10 other people for the same set of songs, and c) maybe they don’t know many country or (this week’s theme)-type of songs.  Give them a break, let them sing what they want!
2. Contestants talking back to judges.  I don’t like hearing the justifications of why they chose certain songs, because I think the performance should speak for itself.  Only when the judges are being ridiculously stupid should they at least defend themselves.

I personally like this season.  There are some clear favorites, but there are some dark horses that SHOULD get much more attention than they’re currently getting (Matt Giraud and Kris Allen).

The country-themed week is usually the week of train wrecks. I can’t wait! Remember to rate and predict performances after the show.

If Michael Sarver is such a blue-collar guy, then why does he own an iPhone? He seems technologically hip, and he’d fit right in here in San Francisco.

I stumbled upon this thread out on the American Idol forums.  It’s really interesting to see who people thought would advance, based entirely on what the American Idol producers wanted us to see.

Predictions at the American Idol forums

I was impressed by Michael Sarver tonight.  But looking at the bigger picture, I believe Michael proved something: that the wild-card round sucks.  The general consenses prior to tonight was that Michael had no business in the Top 13, especially over Anoop Desai (see this post at idolblog).

Based on tonight’s performances, you could argue that America, and also the wild-card round, got it right.  They chose the better singer (Michael) and didn’t choose the weaker one (Anoop).  While this is all true, my problem with this is we only got to see each contestant perform one full-length song.  In the statistical world, each contestant had a sample size of exactly 1.  It’s pretty ridiculous to judge someone based on one performance.

And this is where editors’ choice enters the picture.  If everyone is given one song, then those who have been shown on air more than others (Danny Gokey, Lil Rounds) have an advantage over those who didn’t get much at all (Felicia Barton, Ricky Braddy, Ju’Not Joyner).

I feel last season’s format was MUCH better.  For all but 4 of the semi-finalists, we saw each perform more than 1 song, giving each person a chance to redeem themselves after a poor performance or song choice.  This year, over half of the semi-finalists were cut after 1 song.  As Michael (and also Matt Giraud) proved, maybe you just need a second song to show how good you really are.

There didn’t seem to be much bias in which contestants were chosen to go first, and which were chosen to go last.  Two early favorites started 1 and 2 (Lil Rounds and Danny Gokey), while two others went somewhere between the middle and end (Anoop Desai and Adam Lambert).

I don’t know if it’s my growth into adulthood, but my memory sure ain’t what it used to be.  I can’t remember exactly how good or bad the first 4 or 5 performances were.  I thought Lil was pretty good, but not great, I thought Danny was great, too.  But if I’m comparing against Adam or Alexis Grace, can I accurately say that I liked Lil more than I liked Adam?  Fact is, I remember Alexis because she went last.  Later on, I want to do a more formal analysis on performance order, because with the level of talent fairly high this season, I think smaller factors like performance order and air time will play a larger role.