Notes from the Vault, 9/30/11

This will probably not be something I regularly schedule around here, just something I'll do whenever I notice some stuff. But as I've started working with the True Strength numbers in more of a real-time environment (as opposed to last summer when I was looking back on the whole of the season), I've started noticing some things that I feel I need to mention but that don't seem to fit too organically in the other posts I do. Some of this stuff is about the True Strength metric itself and some is ratings stuff it illuminates. All of it may be incomprehensible, but I'm mostly just taking notes for myself, and if anyone else finds it interesting, neat! Here goes:

Preliminary to Final True Strength Adjustments

As I noted when I first talked about how I'd use TRUE on this blog, I calculate a "preliminary" True Strength (based on just the final program ratings) which goes into the daily Spotted Ratings posts, and then a "final" True Strength the next week when I see half-hour breakdowns that allow me to get more accurate viewing levels and competition. That goes into the SpotVault. I only did limited playing around with the "finals" from the 2010-11 season (since I didn't have the breakdowns for most of the season), and from what little I did, I basically just estimated that across the entire landscape the average change from preliminary to final was about ±5%.

But beyond that, some trends are sticking out that I thought it might be worth sharing to give a sense of where the TRUE numbers on the daily posts will "end up." 8:00 shows (especially half-hour shows just at 8:00) are consistently adjusted up in the finals, usually by about 5-10%. I knew that the competition and the PUT were both being "undercounted" at 8:00, but my hope was those would about cancel out. Instead, the PUT difference at 8:00 is a much bigger deal. Here are just a few preliminary TRUE to final TRUE examples: How I Met Your Mother (8:00): 4.62 -> 4.95. NCIS 4.62 -> 5.02. Glee 4.28 -> 4.66. The Middle: 3.31 -> 3.55. Up All Night: 2.49 -> 2.72. Community: 1.79 -> 1.97. The Big Bang Theory (8:00): 5.84 -> 6.36. So when you see the True Strengths that I put up on the daily full tables, you can usually assume the 8:00 shows will end up 5-10% higher.

The other two hours? A little less consistent and also usually less drastic. 9:00 usually sees a very slight adjustment downward, but typically less than the 8:00 upward (more like 0-5%), and it's sometimes a little up and sometimes a lot down. And 10:00 is similarly slightly down on average and a little all over the map. My instinct is that on a week-to-week basis, the 10:00 True Strengths will be particularly volatile because there are fewer (and generally lower-rated) shows from which to calculate viewing levels, meaning the margin of error is higher.

Viewing Levels

The thing I was most "worried" about going into this season of True Strength was how the viewing levels would look, since this is the first time we're in the fall since Nielsen made its methodology change. I didn't really know what would happen, and the best I could do was make an estimate based on how viewing levels morphed right around the methodology change. I'm going to have to take a much more formal look at this at some point down the line, but my instinct is that they are often even going beyond the adjustments I came up with. The most obvious offender is Sunday, and particularly Sunday at 9/8c where all programs in that hour had a hugely positive Sitch. Most egregious was The Good Wife, which got adjusted from a 2.2 A18-49 to just a 1.47 True Strength! My estimate was that 43%+ of adults 18-49 were watching TV in that hour! As always, some of that may be a rounding error, but it sure does seem high.

Strange Yet TRUE Comparisons

Despite a whooping 0.9 difference in A18-49, new NBC comedies Up All Night (2.4 A18-49) and Whitney (3.3 A18-49) actually had the exact same TRUE during premiere week (2.72). Considering it was week two of Up All Night and the premiere for Whitney, that would seem to bode much better for the former. (They're about even in the preliminary TRUE for this week, but I'm guessing adjustments will take Up All Night ahead.)

As is true in raw A18-49, the only two big hits among new scripted shows are 2 Broke Girls (5.17) and New Girl (4.30). Then there's a huge gap to the next tier led by Pan Am (2.88), Unforgettable (2.72), the aforementioned NBC comedies (2.72), Revenge (2.65), Charlie's Angels  (2.49) and Person of Interest (2.46). Kinda funny that Angels actually graded a little higher than Person even though they were a full point separate in 18-49, and the show was also much more competitive with ABC's other new dramas than it appeared based on raw numbers. But that will be a short-lived victory for the Angels based on the week 2 Thursday numbers. Then another big drop to the flop tier: A Gifted Man (1.77), The Playboy Club  (1.52), Prime Suspect (1.41) and Free Agents (1.14).

The CW was led in premiere week by The Vampire Diaries (1.45) and the season premiere of Supernatural (0.95), but almost everything else except for H8R (0.38) was fairly closely bunched. There was Ringer (0.84), The Secret Circle (0.82), America's Next Top Model (0.81), 90210 (0.76) and Nikita (0.75).