My vision for the 2014-15 SpotVault going into this summer was to do what I've done for the War of 18-49 posts: have one table with purely the official, "raw" numbers and one with the metrics I've devised that are of debatable usefulness. Along those lines, I've been reworking the True formula to a point where I can somewhat clearly communicate how much each factor (viewing, competition, lead-in) impacts the number.
I'm still working toward that vision, but it's not there just yet. There are some radical changes to True, and I want to observe the new formula in action for a couple regular season weeks before I start breaking it apart in a public setting. There may be some further adjustments needed.
So for now, what you will see in the Vault is the following:
| Info | Ratings | Breakdowns | Metrics | |||||||||||
| # | Day | Date | Time | Length | Viewers | A18-49 | Share | Lead | 1st | 2nd | 3rd | 4th | True | Sitch |
| 1 | Su | 9/7/2014 | 8:00 | 2:00 | 4.63 | 2.0 | 6 | 4.9 | 2.6 | 1.9 | 1.7 | 1.6 | 1.69 | +19% |
| 2 | Tu | 9/9/2014 | 8:00 | 1:00 | 2.48 | 0.9 | 3 | n/a | 1.0 | 0.9 | 1.03 | -12% | ||
| 3 | Tu | 9/16/2014 | 8:00 | 1:00 | 2.42 | 1.0 | 3 | |||||||
This new table introduces the most important thing I want to put in the "raw" tables: half-hour breakdowns. Now you can see how a show progressed over the course of the telecast (and compared to its lead-in). I'm not sure there's much of a need for a record of breakdowns for every single episode, but we'll see how useful people find them this year. They're one of the very few things I have access to that doesn't get widely posted elsewhere, and maybe some intrepid readers can make more of them than I can, so I think it's worth a shot. I'll be posting some observations/explanations I've gleaned from looking at half-hour breakdowns as an intro to this feature sometime shortly.
Additionally, this table will contain the new True numbers, even though I'm not currently getting into an explanation of what's changed from last year. Please don't beat up on me too much for these numbers at this point, as they may get tweaked further. But it's close enough to the old formula that it may still have some usefulness.
So what's to come?
Phase II - The next upgrade will be an enhancement of the "averages to date" usually seen in the Vault. Namely, A18-49+ will be incorporated, as I've noticed there seems to be a decent amount of demand for how things stand in Plus as the season goes on. I can't introduce this now because there literally is no 2014-15 league average at this point, but it may show up as early as the Vault update next week (Wed 10/1).
Phase III - Later, a realization of the initial vision explained above: one table of "raw" data, one table that breaks down how much of an impact each aspect of the timeslot has according to the True formula. This will be down the road, but hopefully not too far.
12 comments:
I'm actually interested in seeing the half-hour numbers for a lot of shows so I'm glad you're doing this.
I'll noticed that sometimes CSI grows in the half hour after losing a nice chunk of the Criminal Minds lead in. Is there a way to account for that in True terms?
All three phases sound like awesome updates and clearly stuff I've been looking forward! Thank you a lot Spot! I am especially happy with the half hour breakdowns as it's been something I've been looking for and never got! The PLUS number will also be extremely helpful!
And of course, I am going to be eagerly waiting for the updated explanation on the true formula! (And yeah, I know I am one of the ones always asking "why did true result in X or Y" ahah, so the factors breakdown should definitely help).
By the way, a couple of suggestions that maybe you could incorporate in a subsequent revision of the vault:
- Year to year trends for each episode, both for each single episode and to all episodes up to that point. I find that very useful especially when predicting ratings the year afterwards and it would definitely be very telling in cases like OUAT or Castle this past season for example
- This one I have no clue how much work it would involve or even if you have access to it, but I'll just throw it out there that it would be very nice if at some point you could also add further genre breakdowns, particularly A18-34.
Thanks for everything, as usual!
Kudos. I see you burdened yourself with a lot of work, and thus not really asking for suggestions here... However, knowing you as a person open to suggestions:
When you're already kinda overhauling it, raw ratings with 2 decimal places (like you already have it for cable shows) would be much appreciated from my side. That is, if for broadcast shows those are available to you at all. I know those are meaningless for many shows, but for The CW shows, and for (ever increasing number) of big 4 shows that are nearing CW zone, those 2 decimal places ratings are very interesting.
I had the exact same idea on y2y by episode. I was preliminarily planning to put that where the True numbers are right now. Unless I think of something better, that's likely what will be there by Phase III.
I wanted to do A18-34 and/or gender skew, but I got those from Son of the Bronx and I'm not sure he's posting them anymore. If he does, I will strongly consider that.
I was planning on doing this but I don't have them anymore. I got them from Travis, like you. If they pop up again somewhere, I will incorporate it.
That's even more awesome, thanks a lot! This website really is, BY FAR, the best ratings website there is. Major kuddos to you.
Nice point. I also have them in my spreadsheet from last year like that but they are apparently gone now. For the CW they truly made a difference.
As always, Phase IV is when the ants take over.
I don't always get A18-34 data but will post online whenever I do.
Thanks!
http://blog.sitcomsonline.com/2014/09/week-1-tv-ratings-and-analysis-how.html
Those guys posted 2 decimal places A18-49 ratings for period Sept. 22-26, Sun through Thu. But they don't have it for Fri/Sat, it's obviously because at time of their post (28th) any kind of final ratings for weekend weren't available yet.
Post a Comment