For more than a decade, Chairman of FX Entertainment, FX Content & FX Productions John Landgraf has made an annual presentation to members of the Television Critics Association (TCA) discussing the current state of television. He's one of the few network executives willing to make a presentation to critics each year, which has made him extremely popular with its membership.
One of his over-arching themes during the past few TCA gatherings is the idea of "Peak TV," a phrase he coined back in 2015. Each year he brings out data compiled by FX's research folks, which among other things tracks the number of new English-language originals released over the past year in the United States. And as the number rose each year, the trend sparked an entire cottage industry of think pieces that argued the television industry produces too many new shows and that a rapid contraction is inevitable.
The Peak TV number is almost irresistible to reporters, even though the number doesn't really represent a usable metric for the health or even the growth of the television and streaming TV business. Yes, tracking the number of new English-language scripted originals released in the U.S. is interesting.
But it doesn't actually provide any accurate sense of the health of the TV industry. Any more than you can judge the health of Quaker Oats by tracking how many boxes of Cap'n Crunch it sold over the past decade.
And simply tracking the number doesn't reflect the number being spent each year on scripted content in total. Which industry is stronger? One that produces 50 Andor's or one that produces 150 BET+ originals?
What the phrase "Peak TV" is more than anything else is a sly bit of marketing. It combines a desire by Landgraf to craft a clever message with a tendency of a lot of reporters to use a number even though they don't quite understand what it means or how the numbers were assembled.
In today's presentation, Landgraf revealed that FX Research had determined there had been 516 adult scripted original series released in the U.S. in 2023, which marked a 14% drop year-to-year drop from 2022. He declared the "Peak TV era is behind us," and while he admitted that the combined Hollywood strikes might have had an impact, he believes the drop would have happened regardless.
Here is the chart he used in the presentation:
But even if you believe the number of scripted original is a true representation of the American scripted television series business, other people have attempted to estimate the 2023 number. And none of the estimates were remotely close to each other, or to the one provided today by FX.
In January, I estimated the number and it was substantially lower - nearly a hundred titles lower. And when Ampere Research released their number later than month, their number was also substantially lower:
When Ampere Research released its data, I spoke with the company's principal analyst, who laid out several of the many reasons why the estimates from the respective companies might not match up as closely as you might expect:
- The FX methodology recorded shows releasing new episodes in a given year. This means that there are elements of double-counting of single seasons- for instance if a show launches in October and ends in March the following year, it will be counted in both years. Our methodology instead follows a model of season premieres, which removes this double counting but also can lead to disparity in numbers
- Our data includes Animated and Kids scripted series- certainly the FX data I believe used to exclude these types of content, although I’m not sure if you included this in your research
- The FX data previously included all English-language premieres in the US- so that would include Canadian, British, Australian imports, and even shows that had eg. premiered in the UK in 2021, but in the US in 2022. Our methodology instead only examines content produced in the US- while it may not make too much of a difference to the audience, part of our role as an industry analysis firm is to assess the strength of the TV production industry in different markets, so in a single-market study like this the value of the research is diluted if British or Canadian productions are counted alongside American.
The problem that I encountered when I made my estimate that because of the complicated way so many shows are made these days, including multi-country co-productions, deciding when a show was actually an originals versus one that was licensed three months after it premiered in a different country turned out to sometimes be a judgement call. I purposely erred on the side of caution, but I have no doubt I underestimated the number.
And there is no way to reconcile the estimates. Ampere considers the list of shows it included in its estimate as proprietary data. And FX has been less been responsive to my requests for a bit of insight, instead referring me to the footnotes on their estimate slide, arguing that explanation should be enough.
All of which is to say that when you read a story that quotes the Ampere Research estimate or anyone else's, you should keep in mind that the number is interesting, but not one that is written in stone. It's meant to provide context, not a final judgement on the state of the industry.
Which didn't stop some of the trades today from going all in on the "Peak TV Is Over" narrative.
Here is Variety's take:
And this piece from The Hollywood Reporter treats the data from FX Research as the apparent final arbiter of the state of American television:
Of course, Deadline is also Team Peak TV, although in its defense the piece also has the upside of being extremely bare-bones:
And I know what you're thinking. "Rick, The Wrap pitches itself as the alternative to the Penske Media trades. I bet they took a more nuanced approach to Peak TV:"
All of this is not meant to take a shot at FX Research and/or John Landgraf. It's a reminder that the FX "Peak TV" number is just an interesting talking point. It's not an accurate measure of the strength of the industry and it's only the very roughest reflection of the pace of scripted production.
But what I do hate is how easily otherwise smart journalists report the number without understanding the underlying data gathering methods. They just see a number and think "oh, this is an easy story."
