The Finale Spike, updated

Revisiting this post from a couple weeks ago where I examined just how much bigger a show gets for a season or series finale as a tool for predicting such things in the future. I wanted to fill in a few shows whose finale numbers I didn't have at that time. I'll re-post all the shows but only have commentary on the ones with new numbers.

Grey's Anatomy (ABC)
2007-08: penultimate episode 6.1, finale 7.0, spike of 15%
2008-09: penultimate ep 5.4, finale 5.9, spike of 9%
2009-10: penultimate ep 3.9, finale 5.4, spike of 38%

Desperate Housewives (ABC)
2007-08: penultimate ep 5.6, finale 6.2, spike of 11%*
2008-09: penultimate ep 4.1, finale 4.7, spike of 15%*
2009-10: penultimate ep 3.8, finale 4.0, spike of 5%

Lost (ABC)
2007-08: penultimate ep 4.9, finale 4.9, spike of 0%*
2008-09: penultimate ep 4.0, finale 4.4, spike of 10%
2009-10: penultimate ep 4.3, finale 5.8, spike of 35%

It was about half of the finale spike that ER saw. My prediction for the ep was a 6.6, which would've been a little over 50%, but some people were going much, much higher. I thought the ER percentage was the absolute ceiling. For a show whose audience has really been drilled down to the hardcore fans, though, not a bad number.

Dancing with the Stars (ABC) (not sure if it's better to compare to the previous results show in the same timeslot or to the competition ep from the night before, so I'm including both, and in parentheses are the respective spikes)
Fall 2007: penultimate results show 5.0, final competition show 6.0, finale 6.4 (+28%, +7%)
Spring 2008: penultimate results show 4.1, final competition show 4.4, finale 5.4 (+32%, +23%)
Fall 2008: penultimate results show 4.0, final competition show 4.7, finale 5.1 (+28%, +9%)
Spring 2009: penultimate results show 3.1, final competition show 5.6, finale 5.2 (+68%, -7%)
Fall 2009: penultimate results show 3.2, final competition show 4.0, finale 4.3 (+34%, +8%)
Spring 2010: penultimate results show 2.5, final competition show 4.0, finale 3.9 (+56%, -3%)

Dancing behaved much as it did last spring; until the finale, the spring is really not good to the results shows on Tuesday. That means there's a big spike from the last results show, but even on a finale night it couldn't even build up to the ratings from the previous night's performance show.

Survivor (CBS)
Fall 2007: penultimate ep 4.6, finale 5.2, spike of 13%
Spring 2008: penultimate ep 4.1, finale 4.4, spike of 7%
Fall 2008: penultimate ep 4.0, finale 4.4, spike of 10%
Spring 2009: penultimate ep 3.8, finale 4.2, spike of 11%
Fall 2009: penultimate ep 3.9, finale 4.4, spike of 13%
Spring 2010: penultimate ep 4.1, finale 4.2, spike of 2%

The Biggest Loser (NBC)
Fall 2007: penultimate ep 3.6, finale 4.6, spike of 28%
Spring 2008: penultimate ep 3.7, finale 4.4, spike of 19%
Fall 2008: penultimate ep 3.6, finale 4.6, spike of 28%
Spring 2009: penultimate ep 3.6, finale 4.7, spike of 31%
Fall 2009: penultimate ep 3.7, finale 5.0, spike of 35%
Spring 2010: penultimate ep 3.1, finale 3.9, spike of 26%

The Biggest Loser had a below-average cycle, as the raw numbers for both the penultimate and the finale are significantly lower than the previous ones. However, it remains a show that can bring 'em back for the finale moreso than most shows.

ER (NBC)
2007-08: penultimate ep 2.9, finale 3.3, spike of 14%
2008-09: penultimate ep 3.5, finale 6.0, spike of 71%

American Idol (Fox)
2008: penultimate results show 9.2, final competition show 10.1, finale 11.4 (+24%, +13%)
2009: penultimate results show 8.7, final competition show 8.6, finale 10.0 (+15%, +16%)
2010: penultimate results show 6.1, final competition show 6.7, finale 8.2 (+34%, +22%)

A bigger spike than usual for Idol this season, but it just goes to show how far down the show really was by the end that an 18% drop from last year's 10.0 finale is one of the better results of late. The penultimate results show was down 30% from the 2009 edition. Ouch.

24 (Fox)
2009: penultimate ep 3.2, finale 3.1, spike of -3%
2010: penultimate ep 2.9, finale 2.8, spike of -3%

I was wrong on this one. I thought a series finale for 24 would draw back some of the old fans, but it didn't happen at all. The show behaved exactly as it did last year, dropping a bit for the finale because it lost the support of House. Perhaps what I missed here is that 24 was never really a gigantic hit to begin with. Yes, it's been bigger than the middle and upper 2's it's gotten this year, but the pool of former fans is not as large.

House (Fox)
2007-08: penultimate ep 5.9, finale 5.8, spike of -2%
2008-09: penultimate ep 4.6, finale 4.7, spike of 2%
2009-10: penultimate ep 3.6, finale 4.3, spike of 19%

America's Next Top Model (CW)
Fall 2007: penultimate ep 2.3, finale 2.6, spike of 13%
Spring 2008: penultimate ep 2.0, finale 2.2, spike of 10%
Fall 2008: penultimate ep 1.9, finale 2.3, spike of 21%
Spring 2009: penultimate ep 1.7, finale 1.9, spike of 12%
Fall 2009: penultimate ep 1.5, finale 1.8, spike of 20%
Spring 2010: penultimate ep 1.5, finale 1.7, spike of 13%