This Spring: Worse than the Other Springs?

Is the typical 35 or 36-week broadcast season just too long nowadays when the longest-running seasons have about 24 weeks of originals? That was my question when I examined what's still by far the most-used piece of "filler": the rerun.

What gets the media really asking this question, though, is the ratings declines in the spring. With all the data in the books, now I can really examine that side of the coin: are the ratings declines in the late spring creating enough of an advertising incentive to pile all the episodes of the big-money shows in the highest-viewed parts of the year?

It's not a question I can answer directly, because I don't know where the threshold is. I don't know at what point the late spring ratings get so untenable that there'd actually be a change. All we really know is that the September to May season has worked in the past, even though ratings always decline in the late spring.

So what I can do is look at the trajectory; if the attitude is going to change, it has to be because the late spring is getting even worse than it was before. I can at least see if it's heading in that direction. I'm going to look at this using a few numbers that should, in theory, remain about the same throughout the season. If these are getting worse, maybe there really is a problem.

My first approach is the most macro of all: the average of all big four original entertainment programs that I've used in all the A18-49+ posts. This chart looks at the season-to-date league average and compares it to the previous season's season-to-date league average at the end of each week of the season.

The white background part is essentially the spring, starting at approximately the end of February sweeps. What sticks out is that 2011-12 was the first season in which the year-to-year entertainment average significantly changed during that period, in either direction. And the direction was down, from about -3.5% at the end of February to about -7% by season's end. None of the other seasons saw a change of more than one percentage point or so during that same period; that seems natural, since it gets harder to move the average with five months already in the books. Looking at each week individually, the entertainment average was down double digits year-to-year in 12 of the last 13 weeks in 2011-12. Just two of the first 23 weeks were down double digits, and both of those were during the holidays (when the sample size is very small).

It seems pretty undeniable that this spring was the first spring in the last several years that was significantly worse year-to-year than everything that came before in that season. The problem is that using such an extremely general perspective means there are several other factors. As you can see, the 2011-12 season was much stronger year-to-year than any of the previous four seasons for most of the way. The average was rounded to -0% as late as mid-November and was down an almost unfathomably good 1% at the end of 2011. So maybe the fall created an unrealistic expectation and then the spring drops were just evening it out to some degree.

A big reason for the great fall was the fact that Fox was significantly up across a large portion of their fall real estate with The X Factor (which was often compared with stuff like Human Target). Then, in the winter/spring, they were wayyy down in those same exact slots as American Idol had a very bad season.

In other words, that one big number may well be largely driven by just a few shows. Idol and X Factor take up a lot of space, and since they're so high-rated they also contribute a disproportionate amount to the averages.

So my other way of examining this is to look at individual shows, particularly ones that were on the schedule for pretty much the whole season. Did the year-to-year trends tend to get worse late in the season for individual shows?

What I did was look at 42 returning scripted shows (after trying to weed out shows that made big midseason moves that may have skewed their numbers) and compared their premieres year-to-year with their finales year-to-year in 2011-12. 24 of the 42 shows (57%) had a better premiere year-to-year than finale year-to-year. Granted, several of the remaining 18 were either second-year shows (that may have had inflated series premieres) or announced final seasons (that may have ended on a better note as people came back). Still, while 57% might suggest that shows were doing slightly better at the beginning than at the end, it's not exactly a smoking gun.

One last observation: my overall viewing estimates from year to year:

So while there were a couple strangely low-viewed weeks (the last week of 2011 and the Super Bowl week), the only time viewing was consistently down by more than 2% was in the last couple months of the season. Of course, broadcast viewing and overall viewing affect one another to some extent, but this suggests it might have been something beyond the broadcast shows sucking. Maybe the weather was just phenomenally good this spring?

Conclusion: Spring 2012 was definitely significantly worse than at least the last four springs before it. Does that mean it's time to wrap up every season by March? I'm not ready to go that far, because I think this season as a whole was more about what happened in the fall. Fall 2011 was one of the best in recent memory on several levels. There was the Two and a Half Men explosion, there was The X Factor, and there was an extraordinary volume of new show success (followed by virtually no promising debuts in midseason). And it's not like the spring has been gradually getting worse over the last several years. For now, it looks like just a one year blip. I'm gonna hold off till we have more seasons like this one. Sadly, we may be setting up for another artificially strong fall, since NBC will have The Voice in the fall.

But hey, maybe spring 2013 will have terrible weather! We can only hope, right?!?!