Last-Minute True Strength Tweaks

The daily Spotted Ratings posts will go back into "Full Table" mode starting in a few hours. In creating the spreadsheet for the 2011-12 season and incorporating True Strength, I've run into a few snags/issues that it's reasonably possible to address immediately, though several still remain at the bottom of the summary of the process:

• I've made yet another tweak to the Old Methodology estimate. I broke it down by hour when I first came up with the current decreases, mostly to increase the sample size, but I think doing that is undercounting the PUT particularly in the 8:00 half-hour, where the New Methodology change is much smaller than at 8:30. So I'm now breaking it down by half-hour. More details on the Methodology Adjustment page.

• I really wanted to just have one local programming lead-in number because I was operating under the assumption that it'd have about the same rating the whole week. But really, a lot of the local programming is syndicated Monday to Friday stuff and then something different on Saturday and Sunday. So I ran the Saturday 8:00 and 8:30 episodes of Cops (which IMO should have about the same "True Strength") through the True Strength wringer and found that assigning the 8:00 episodes a lead-in of about a 1.00 nearly equalized them. This is a bit of a conservative estimate, and it's sorta variable because a lot of times there's afternoon sports stuff on the weekend that affects local programming, but we'll say that Saturday and Sunday local programming lead-ins are equal to 1.00.

• NFL days are a particular challenge on a couple levels:
• As I said earlier, the "final" True Strength calculations will be based on half-hour breakdowns through the whole of primetime rather than full program ratings. This will be a problem for NBC's football games because I don't think I'm going to have the half-hourlies beyond 11:00 (and even if I do, I wouldn't have them for all the programming on the other nets beyond 11:00 to calculate competition). So the "final" True Strength for NFL games that run past or before primetime will be based on the primetime portion only, and the primetime portion often has a different (if not very different) rating than the full program. So what I'll do on the SpotVault is list the full program A18-49, a True Strength for the primetime portion, and a Sitch that compares the primetime portion A18-49 with the primetime portion True Strength. Sort of complicated, but I'll write this on the sports program SpotVault pages too.

• The other issue is the entertainment programs that get pushed by overruns. Since the half-hour breakdowns are attached to time periods, not to programs, I won't be able to get half-hour breakdowns of overrun-affected programs. So here's the plan: rather than try to estimate half-hour program breakdowns, True Strength will count the program as the same rating throughout (same way I've done it so far and will for prelims going forward) and weigh the competition/PUT in each half-hour based on the size of the overrun. For example, for a The Good Wife episode starting at 9:20, the Persons Using TV will be equal to 10 times the PUT from 9:00 to 9:30 + 30 times the PUT from 9:30 to 10:00 + 20 times the PUT from 10:00 to 10:30. Then I divide by 60. In other words, ten minutes of 9:00, thirty minutes of 9:30, twenty minutes of 10:00, mirroring how much the show aired in each of those half-hours. The shows that run past primetime will basically have their last half-hour assigned the same competition and PUT as is calculated at 10:30. This still won't be completely accurate, but it'll be a much better estimation than the preliminaries give.