True Strength: The Issues Ahead with Competition

As we've established across the last slew of posts, there's a relatively reliable correlation between ratings levels and viewing levels for the most part. We've looked at a few exceptions to that rule already, but competition is one of the biggest exceptions. There are times when viewing levels rise not due to a general "tendency to watch TV" but because there is actually something specific on the TV that is bringing in new viewers. And that thing on the TV actually drags down the other things on the TV.

Broadcast Competition vs. Cable Competition

As we've looked at in some other posts, the totality of broadcast programming is much more variable than the totality of cable programming. Broadcast is a small collection of networks that runs big hits sometimes and comparatively big flops sometimes; cable is a much larger collection of networks which individually don't get big ratings, so there's less variance. When looking at how TV events drive viewership away from other programs, there just aren't many programs on cable that are big enough to meaningfully make that happen. The cable competition just doesn't fluctuate enough to make it worth throwing that whole extra wrench in. There are a few times where I may regret this, but the only ones I've really come across are the BCS Championship Game (recently moved to ESPN) and perhaps to some extent ESPN's Monday Night Football. So we're just going to look at "competition" as broadcast competition.

Comparing Competition - Raw Drops vs. Percentage Drops

I've been posting the "competition" stat in my Spotted Ratings tables for months, and explaining it is extremely simple: it's just the sum of all the ratings of the broadcast shows airing against a show. But it's a pretty tough number to get my head around in terms of actually using it for comparisons. Here's why.

1) Different shows, even in the same timeslot, have different levels of competition. It would seem that for Community, The Big Bang Theory and The Vampire Diaries, the injection of American Idol into the timeslot would basically be the same "size" event in all three cases. But in terms of increasing that competition number, it does it by pretty different percentages. Since Big Bang is a much bigger show (whose bigger numbers are reflected in the competition of the other two), adding in Idol has a much bigger effect percentage-wise on Big Bang. So it's better to look at changes in competition with raw number changes.

2) However, when looking at the changes in ratings for the shows, the correlation is closer with the percentage changes. An event isn't gonna take the same raw number out of shows of very different sizes.

So... I think we have to look at raw numbers for competition changes and percentages for ratings changes. Kinda awkward, but it's the best way I have right now of wrapping my head around it.

Baselines

Given the above, we have to come up with two key "baselines" before we can incorporate competition into the True Strength formula.

1) How much % change is expected from a one-point change in competition?

2) What is the "expected competition" in a given timeslot? In other words, what number do we use to figure if a show is facing heavy or light competition?

Those will be our next two posts!

Exceptions

Even after doing those things, there are a few situations where the raw competition number does not tell a particularly good story. One of those I've actually addressed many months ago: the 10:00 hour (where competition is inherently lower because there's no Fox). The other is the "type" of competition. The one I'm really thinking of here is sports. Lots of shows thrive in the fall against Monday and Sunday Night Football, maybe even do better than in the spring. So what do we do with the inflated competition numbers caused by sporting events that largely consist of people who aren't "up for grabs" by other entertainment programming?

I had some illusions pretty recently that competition would be a breeze and we could move right into lead-ins. But it looks like we're probably dealing with several more posts on this one... we'll get into the meat of it all next week.