Thursday, June 27, 2013

Accountability Week, First Two Weeks

This is part two of a look-back on all the predictions made over the course of the 2012-13 season. I will save the planned third part, looking at the best and worst The Question lines, for the end of the summer and include the summer lines in that. (Alternate explanation: I ran out of time in my pre-vacation work-ahead. You decide!) This time, we'll look at my predictions on the posts detailing the First Two Weeks of new shows.

I'm not gonna discuss the actual lineups and the "system" this time, because I just don't care that much anymore about trying to develop a truly systematic way to do this. Here were my prognoses for each show based on the "eye test" from two weeks of data:

The New Normal - CANCEL - CORRECT
The Mob Doctor - CANCEL - CORRECT
Revolution - RENEW - CORRECT
The Mindy Project - RENEW - CORRECT
Ben and Kate - CANCEL - CORRECT
Animal Practice - CANCEL - CORRECT
Guys with Kids - defer -> CANCEL - CORRECT
The Neighbors - CANCEL - INCORRECT
Last Resort - defer -> CANCEL - CORRECT
Elementary - RENEW - CORRECT
Made in Jersey - CANCEL - CORRECT
666 Park Ave - CANCEL - CORRECT
Nashville - defer -> RENEW - CORRECT
Beauty and the Beast - CANCEL - INCORRECT
Emily Owens M.D. - CANCEL - CORRECT 
Malibu Country - defer -> CANCEL - CORRECT
Deception - CANCEL - CORRECT
1600 Penn - CANCEL - CORRECT
The Carrie Diaries - defer -> CANCEL - INCORRECT
The Following - RENEW - CORRECT
Hannibal - RENEW - CORRECT
Family Tools - CANCEL - CORRECT 

FINAL RECORD: 27-6 (82%)

82% is my best percentage yet, though it's not that different from the 80% last year or the 75% a couple years ago. I went 4-1 (80%) on the week three deferred shows, so adding that option didn't really help percentage-wise, although the only deferral miss (The Carrie Diaries) was one of the biggest surprises.

The original goal of the First Two Weeks posts, one of the longest-running features on the site, was to figure out how much of the verdict is in based on a show's very early ratings. Because everyone wants to sound smart by saying "the true test is in the next few weeks," I've believed the conventional wisdom now undervalues the importance of the opening couple weeks. To paraphrase a sports saying, you can't necessarily win a renewal in the first two weeks, but you can sure lose one if you're dead on arrival, and that matters. Statistically, it's a hell of a lot better to start off well.

At this point, the verdict seems to be in about how much of the verdict is in; I can correctly predict renewal/cancellation for a new show about 80% of the time based on two weeks.

But that 80% number is not the real metric of importance here. What's important is how much worse 80% is than waiting till the renewal decision time and predicting based on the full body of work. So the ones that really interest me are the shows for whom the picture changes to a great degree after the first two weeks.

There was only one show for which the first two weeks prediction was clearly wrong by upfront time: NBC's Chicago Fire. I don't really regret dooming ChiFi based on the first two. In the end, week two of Chicago Fire just so happened to be the weakest data point in the history of the show. That doesn't happen very often, but every so often those bad breaks will happen.

The second category consists of three misses that weren't totally clinched by season's end, but personally I had changed my mind on them: Go On, The Neighbors and Beauty and the Beast.

In the final Power Rankings, Go On was at a level I thought was clearly deserving of cancellation, but I expected it would scrape out a season two as a sort of face-saving measure.

I would have renewed The Neighbors in the end, but by no means did that seem to be a sure thing down the stretch. I thought there was a good chance it'd be squeezed out by How to Live.

And here was my line on Beauty and the Beast: "So I'm saying Cancel on sheer merit and hoping everything sorts itself out eventually." Well, it didn't. It looked pretty likely by the late part of the season that the CDub didn't have enough depth to do away with this show. This was this year's version of last year's Terra Nova pick, where I basically called out exactly why it should've gone one way, then went the other and was wrong.

By the way, there were a couple correct picks that fall in this category too: Malibu Country and Hannibal. I would have renewed Malibu in the end (though I did have it much lower on my personal totem pole than many people), and I thought Hannibal was doomed when it dropped into the 1.0 vicnity. My guess is co-production Hannibal didn't get season two because of acceptable "normal" ratings but because it was judged more like a summer show (and one with huge acclaim).

And the final category: two shows on which I was wrong initially and wrong based on the final data: The Carrie Diaries and How to Live. I'm still not totally sure what to make of those final decisions, so I'm not gonna lose a lot of sleep over the first two weeks calls. Maybe Carrie means the CW still puts huge stock in young female demos/buzz. Maybe How to Live was just the one time ABC decided to go retentionista. Either way, the "mistake" here had nothing to do with the first two weeks not telling the tale, as my view on both ended up the same.

So to sum up, there was only one of the worst kind of miss (clearly wrong by the end), Chicago Fire, but I'm not too bummed about that one because that second week just happened to be the worst point of the season. Beauty and the Beast was probably the "mistake" of the year, since it was already clear by then that the CW depth was not there, but then it also held up better over the course of the season than did last year's The Secret Circle.

If I did these again at the end of the season, I'd have probably wrongly changed Malibu Country and Hannibal and rightly changed Chicago Fire and Beauty and the Beast, while I'd have been about 50/50 on The Neighbors. So my record probably would not have been more than one show better. Moral of the story: the first two weeks tell you a lot!!!

No comments:

Post a Comment

© 2009-2022. All Rights Reserved.