A Netflix Analysis: The Demography of Cancellations
Updated: Feb 4, 2021
A few weeks ago I read a headline that said something like "Netflix cancels GLOW and renews Space Force." First I was peeved because I love GLOW and was looking forward to the next season. Next I was flabbergasted because Space Force was pretty disappointing (I'm being kind there - it sucked). I wondered what would compel this decision. And I wondered if this was just a weird one-off or if there was any kind of pattern to recent cancellations and renewals of Netflix Originals. I decided to find out what I could.
The reason I felt the need to do this analysis is pretty simple. Netflix is still the biggest player in the streaming space. If you can afford only one streaming subscription, odds are you are going to have Netflix (this may not last for much longer, but it is today's reality). This means they have pretty enormous influence over what content people are consuming. The decisions they make about what is or isn't made can mean that someone sees one less diverse cast or one less strong female lead. In the entertainment industry, there is (and has always been) a delicate balance between the business imperatives of what makes economic sense to produce and what is good. It's kind of a chicken or the egg type of conundrum. Is a song popular and thus it is played constantly on the radio? Or is it popular because it is played constantly on the radio? Are women's sports teams inherently less entertaining? Or do they get significantly less marketing investment? There is a lot of data to suggest that more often than not, the issue is not with the content quality itself, but with the marketing investment in the product. This means that if a Netflix decides not to market a show with a diverse cast or made by a female showrunner as hard as something else, it might underperform even if it is actually a better show. Then when they cancel it, they have deniability because of performance metrics, but those metrics were, at least in part, created by Netflix and not by the show.
Netflix does not publicize most of their data so we can only speculate on whether any decision they make is really due to the performance of a show or whether it is simply some decision-maker's taste or bias. And even if we had that data, we wouldn't be able to know whether a show performed well because it was boosted by the platform or whether it was truly a great show. And conversely, whether an underperforming show did so because it was lacking in some way or because it wasn't properly promoted.
Given all that, I simply collected what information I could find and drew some basic conclusions.
I want to preface this by saying there are some very serious limitations to my data:
There are only 70 shows from 2020/21 that I could find definitive information on whether they were renewed or canceled. I did not include any shows that had not been publicly decided on yet (Sources: Newseek 1 & Newsweek 2)
For most of my data I used IMDB, which is definitely not infallible
For cast diversity, it was entirely a subjective measure of my interpretation of the race/ethnicity of the cast list on IMDB - there are things I can't see, like whether someone is queer, so I did not include that, and also I am sure to have made incorrect assessments of people's racial and ethnic makeup. Please take this particular part of the data set with a huge grain of salt.
For the gender of the lead, I went based on the highest billed actor on IMDB - since I haven't watched all of these shows, I have no idea if that is accurate to the true lead of the show
OK so now that I've basically told you all of the data is seriously flawed, let's see if there IS anything useful or interesting that we can conclude.
Out of the 70 Netflix Originals I looked at, 51 were renewed, making the overall renewal rate 73%. The overall average IMDB rating for all shows was 7.4.
The average IMDB rating of the canceled shows was 7.2/10 and the average rating of the renewed shows was 7.4
The Mode rating of the canceled shows was 7.6 and the renewed shows was 7.4
The Median rating of the canceled shows was 7.5 and the renewed shows was 7.6
The lowest rating overall was a renewed show that rated 4.4/10 (The Unremarkable Juanquini)
The highest rating overall was 8.7 (both Stranger Things & The Crown) and the highest rated cancelled show was The Dark Crystal: Age of Resistance with 8.5
My Conclusion: basically rating (at least IMDB rating) has little relationship with whether something gets canceled or not. This possibly confirms what I was saying about show quality having little to do with whether something gets renewed or not.
Only 27% of the shows had what I would describe as high cast diversity, 54% had either high or moderate levels of diversity (again based on my eyeballing of casts)
The renewal rate of highly or moderately diverse casts was 68%, lower than the average renewal rate. The low diversity casts had a renewal rate of 78%, higher than the overall rate
The ratings of highly diverse casts averaged 7.1/10 on IMBD, with moderately diverse casts 7.4, and with low diversity 7.7
My conclusion: Assuming that my basic assessment of the diversity in the first place is somewhat accurate, unfortunately cast diversity has an inverse relationship with renewal. Not only are there simply fewer shows even made with more diverse casts, but the ones that are are being renewed at a lower than average rate. The IMDB ratings might provide some insight into why things were getting canceled, but even the moderately diverse casts had a higher than average IMDB rating, but were still being renewed at a below average rate.
Overall there were more shows with female leads-- 54% compared to 44% with male leads
The renewal rate for male leads was 81%, significantly higher than average, while the renewal rate for female leads was 66%, significantly below average
Women led shows averaged 7.2 on IMDB and male led shows averaged 7.6/10.
The one show led by a non-binary actor (Elliot Page in Umbrella Academy) was renewed and it was rated 8.0/10 on IMDB*
My Conclusion: Female led shows just didn't come close to getting renewed as much as male led shows. They were admittedly also rated lower on average, however the difference in ratings between male and female-led shows doesn't appear to be nearly as large as the difference in renewal rates and thus can't account for the entire difference, if it can account for any of it.
51% of the shows were created by white men, 34% had at least one woman showrunner, and 41% had at least one BIPOC creator (note that there was overlap on some shows and that is why 34% and 41% add up to more than 49%)
The renewal rate for shows with no diversity in their showrunner team was 72% and the renewal rate for the shows with diversity in their leadership team was 74%
Shows with white male creators and shows with diverse leadership both averaged 7.4/10 on IMDB
My Conclusion: This part of the data was much more encouraging than the rest. Almost half of all shows had at least one woman or BIPOC creator and those shows had an above-average renewal rate, compared with the below average renewal rate of shows created by white men.
OK so what does all of this mean? Maybe nothing. But it does, at least to some degree, seem to confirm my suspicion that shows with diverse casts are getting short shrift at Netflix. And I wish it didn't. Because, as I said before, Netflix, and all of the big platforms and networks, have a responsibility to ensure representation. They cannot simply shrug their shoulders and say "well they aren't getting as good of ratings so they go" because firstly, the ratings may be a result of sub-par marketing, and secondly because people's tastes shift and change based on culture. Netflix can't just say they are responding to cultural trends. They are creating cultural trends and they need to recognize and wield that power responsibly.
*Elliot Page's character is gendered female in Umbrella Academy. If one counted the gender of the portrayed character, rather than that of the actor, then the proportion of Female leads is 55% and the renewal rate for them is 67%.