To some extent, this grim reality hurts the integrity of historical ratings. TVByTheNumbers has this thing they call the "Gunsmoke Rule" which essentially says that any ratings beyond the previous season are meaningless for comparison purposes. Sometimes I feel like they just wave that around to save themselves the trouble of looking something up, but the general principle is certainly correct; as ratings decline, so too do standards for renewal and cancellation and pull-me-from-the-schedule-right-now, and this is drastic enough that the raw number standards for network decision-making are completely different within a few short years. Just one example: CBS pulled fall 2006 newbie Smith from the schedule after it got a 2.8 demo in its third episode. Five years later, the occupant of the same timeslot, Unforgettable, premiered to a 2.9, has averaged about a 2.1 for most of 2012, and will air (at least) a full season.
All this collective declining is meaningful and needs to be reported. At this rate, eventually ratings are going to get low enough to seriously challenge the broadcast model. No denying that here. But we're not there yet, and sometimes I think all the talk about series lows and the collective decline misses the trees for the forest. For now, broadcast TV ratings are still a system that largely operates by the same set of rules. It's just that the standards for "hit" and "flop" shift to match the collective declines. There are other things happening in this system beyond "everything's down."
Since it's basically the same system, that means there should be a way to put those old, much higher ratings on a level playing field with those of today. Enter what I'm calling the A18-49+.
That name borrows from the world of baseball, where sabremetrics has numbers that adjust for the run production of the league as a whole. For example, ERA+ compares a pitcher's earned run average with the league's average ERA of that season. That allows for a relatively good comparison of pitchers across eras in which the collective offensive production was hugely different.
That's what A18-49+ does too. It sets the larger trend off to the side and creates a world in which every season always averages out to 100. Over the last few weeks, I've compiled all the broadcast A18-49 numbers over the last six regular seasons, and I've used those to come up with the TV ratings version of "league average": the average of all original entertainment programming across the big four networks during the regular broadcast season. It's a little unfortunate that I only have enough numbers to come up with a good "league average" for six seasons out of the history of TV. Still, the landscape here is changing a lot more quickly than in baseball, so it seems potentially pretty useful.
The formula for A18-49+ is simple: 100 * show 18-49 average / league 18-49 average
This will output a number similar to ERA+. 100 = league average, and the bigger the better.
This is not the kind of number that can get thrown organically into daily ratings conversation the way I do with my timeslot adjustments. It's more of a big-picture thing. But there's a lot of stuff that the A18-49+ can help illustrate, and I'll be exploring much of it each Friday over the next several weeks. Some of these will be more about scheduling than ratings, but combining it all, I hope it'll make for one of the most comprehensive looks at the shifting TV landscape over the last few years.
For now, I leave you with this introductory look. There are a few fairly distinct types of shows that emerge from looking at this. Not every show fits neatly into one of these three categories, particularly across the whole six-year period, but quite a few do.
First, there are the shows that "swim upstream." In other words, they hold up from a raw numbers standpoint over time and, as a result, grind out an increasing amount of value relative to the ever-declining league average. A few shows, like the two below, fall into the sweet spot where they actually decline a bit in raw numbers but gain in A18-49+. Of course, the shows that actually grow in raw numbers (NCIS, Modern Family, The Big Bang Theory) have risen even more precipitously.
Then there are those reliable staples of primetime that never seem to get any real attention and amazingly decline at almost exactly the rate of primetime as a whole every single year. These shows "ride the current." Many of these are the kinds of shows you'd expect:
Perhaps the most amazing member of this "ride the current" group is the biggest show on TV. American Idol has been down in raw numbers every year that A18-49+ data is available, but its complete dominance of the relative landscape in the second half of the aughts has remained almost exactly the same...
...or at least it has until this season. More on Idol in future posts!
Sometime in the near future (probably this weekend), I'm gonna begin installing the A18-49+ numbers by season on the War of 18-49 pages, so stay tuned. Next Friday, we'll dive back into this by trying to define commonly bandied-about words like "hit" and "flop" from an A18-49+ standpoint.