As I mentioned a couple of weeks ago, I was going to go back and release my updated How Old Is Thing Song study. (Thanks again to Landon Cox of Burleson, Texas for sending these numbers in.) After several weaks of looking at the numbers, poking them, tweaking them a bit, what have we found? As far as younger songs doing better than older songs in Idol competition, the numbers are looking even better than they did three years ago. As for post-Idol success, that’s harder to say: although to be fair, there are a lot of other factors at work there. So, let’s look at Seasons 8 to 10. Next week (fingers crossed),
The methodology is still the same as I outlined in How Old Is Thing Song 2.0, with one change: coronation and reprise songs will not be counted in the averages and median anymore. Most of the time, they only serve to artificially lower the number of the top two placers.

Season Eight: The Surprise Finale That Really Wasn’t

Take a good look at the difference between Kris Allen’s average song age and Adam Lambert’s. On average, Adam was singing songs almost a decade older than Kris. Then again, on average, Adam was the graybeard when it came to song choice: he was singing much older songs than the rest of the field. Lambert was not only the most controversial contestant of the season, he was (with the exception of the robbed Alexis Grace) the oldest artistically. Add to the many reasons Kris won and Adam didn’t song age.

I wish I could say that the young song picks of Season Eight made for an enjoyable season – but I’d be lying if I did. It was torture. No other season has inspired me to write a three part editorial calling out the Idol PTB. It was that bad. There is nothing that causes more anger in the Idolsphere than selective judging – whether it’s for someone or against someone.

This season’s low song age was achieved in a surprisingly easy way: theme selection. By Idol standards, there were few graybeard themes: sure, you had Motown and Rat Pack songs, but beyond that there really weren’t any atrocities. On the other hand, you had themes like Year You Were Born (when, by far, the contestants did the theme properly and didn’t cheat), Top iTunes downloads, and songs of the Opry.

However, song age can really only do so much. Young songs done poorly will still stink. This chart was done by my friends at What Not to Sing, and shows the WNTS scores of the first seven eliminated contestants in the finals. Scores over 50 have been highlighted. It speaks for itself.

Spare a thought for one contestant in particular. Alexis Grace proved again that singing older than average in a given year is dangerous for frontrunners. At the time of her elimination, her mean/median score was 32.67/35. Only Adam Lambert had an older score in both categories at that time. It was simply her bad luck to be in Season Eight and not Season Nine, as you will soon see.

Season Nine: Guys versus Girls, Contestants versus Producers

From the point of view of song age, Season Nine got off to a good start. The top 12 finalists, through three rounds of semifinals, had an average song age of 19.17 and a median of 11.5. You couldn’t complain about songs being old. And then the producers came in.

The average song age during the finals was more than a decade older at 32.13. The median was an appallingly high 39. Stop for a moment to ponder these statistics: half of the songs in the finals of the search for the next great singer in America was four decades old. The first five finals episodes had an average song age of 37.76 and a median song age of 41. What. The. Hell. Was. 19E. Thinking?

Some of the patterns we’ve seen before are, again, present in this year. Didi Benami should have lasted longer – but at the time of her ouster both her average and median song ages were above 30, where it starts to become a risk for contestants. Siobhan was in a similar situation, and also went out earlier than many pundits thought she should. On the flipside, consider the significant gender gap: there’s no guy with an average song age in the 30s; conversely no girl (save the hapless Lacey Brown) with a similar score in the 20s. However, correlation does not equal causation: in quite a few of those cases (most notably, Katie Stevens) the girls were perfectly capable of choosing young material. It just happens to be that they were eliminated before they got a chance to do that in the finals.

Season Ten: Casey Abrams, King… Graybeard?

This season was a powerful demonstration of the power of song age. First, however, I’ll dispel one possible case. Looking at Pia’s numbers, one might think that high song age was the ultimate culprit for her exit. It wasn’t. The table below show’s everyone numbers when Pia was eliminated:

If anything, Pia was singing on the younger side of the top 9! Pia left with three straight old themes on her Idol resume. Season Ten, like Season Nine, didn’t really get young songs until five weeks into the finals. That was after Pia had already left, so she’s saddled with a number that reflects the geriatric nature of the early S10 weeks. There were many factors behind her early exit, but song age… was not one of them.

A classic example of song age kicking somebody out was Casey. At the time of his Judges’ Save, his average and median song age were both in the 40s (40.5/43, to be exact). That’s essentially unheard of in Idol. Anything above 30 is a warning sign, let alone 40. He left with an amazingly high median song age of 41. If there was ever a king of the graybeard songs, Casey was it. The queen, however, was, surprisingly, Thia Megia. In fact, that number seemed so wacky, that I’ve already done some digging into some… other aspects of that anomaly. Maybe that’ll come out in another couple of weeks, or it may go nowhere if the numbers don’t pan out.

Looking at the top three, the most remarkable number there has to be Lauren’s. Consciously or not, not only was she borrowing Carrie Underwood songs, she also borrowed heavily from the latter’s Idol strategy. Given the… changing conditions of song choice since Season Four (read: iTunes), getting a median below 20 and lasting as long as Lauren did was a great achievement. It was generally assumed at the time that Lauren was vote-splitting with Scotty (hence, few believed both country contestants would reach the finale), but in hindsight thay may well not have been the case. It may well have been the what she had was the teen/tween vote and not country voters. Had Lauren been picking more “average” songs age-wise, she would probably not have made it into the finale.

Haley’s numbers show what is just about the only way to get out of a song age hole. Pick a really current song and sing the crap out of it. She did it twice – Rolling In The Deep and You and I. It’s a high-risk strategy, but then again if you need to get out of a song age hole you don’t have a choice. Of course, if you’re in a song age hole because old songs are your thing – Casey’s a good example, Haley to some degree as well – then you have to realize that that is something which makes advancing week to week that much harder.

Whatever the case, Season 10 was a season of extremes. You had Lauren Alaina doing as good a job as anyone ever has of selecting young material for her stay. On the opposite end, Casey was picking from the older half of the songbook.

In Closing

The past three Idol seasons have not taught us anything we couldn’t infer original How Old Is This Song study years ago. The lessons of the past are still unchanged: on American Idol, you’re better off doing new songs rather than old ones, all other things being equal.