The Twitter profile for The Cancel Bear.
I never had a huge personal following on Twitter. I topped out somewhere around 10,000 followers before I left post-Musk enshittification. The emphasis there is on personal — because for about three years starting in 2015, I was the person behind an (in)famous account in TV Twitter — one that had been name-checked by award-winning actors and referred to on screen and in TV episode titles, just about always in a negative way.
For those three years, I was the Cancel Bear.
Now, I should say that the Bear’s notoriety was not in any way due to me. The account’s beefs with TV stars, producers and some reporters and critics mostly happened well before I took over the account and the site it was affiliated with, TV by the Numbers. But I carried on the account and TVBTN’s most notable feature, the Renew/Cancel Index, the entire time I was there.
TV by the Numbers carved out a pretty sizable niche on the 2010s internet, providing a service — a heavy focus on TV ratings and what they meant for shows’ chances of survival — that wasn’t readily available from the Hollywood trades or more general-interest TV sites. A lot of places had at least some ratings coverage and had “bubble watch” features, but TVBTN became a one-stop shop for daily and weekly numbers for broadcast and cable shows, and it flourished.
By the time I left the site in June 2018, though, the obession with ratings — both among readers and, especially, within the industry itself, had cooled way down. Tribune Media, which owned the site, never paid it much attention and shut it down about a year and a half after I left. I don’t know that TV by the Numbers would thrive today the way it did a decade ago, and I for sure don’t think a feature (or gimmick, if you want to call it that) would work now.
On the other hand, I might not be writing this newsletter if I hadn’t put in the time there and started really digging into the data behind the TV business. Ratings are much less likely to be the sole arbiter of a show’s lifespan now than they used to be, but knowing what people watch is still foundational to the TV business and worthwhile for the rest of us to understand too.
Over the next couple weeks, I’ll delve into the rise, peak and eventual decline of the Cancel Bear and TV by the Numbers, because it’s as good an illustration of how the TV business (and social media, for that matter) have changed over the past 15 years.
TV by the Numbers was born in 2007, created by two guys named Robert Seidman and Bill Gorman. They recognized that while a lot of places dabbled in TV ratings coverage, no site — or at least, no site with any sizable footprint — put ratings front and center. And at the time they started the site, ratings were still the lifeblood of the television industry.
To set the stage: In the 2007-08 TV season, about 45 network shows averaged more than 10 million viewers per episode, and a dozen of those were above 15 million, led by the peak era of American Idol on Fox. Sunday Night Football, which in the present day is perennially the No. 1 primetime show on TV, finished 11th that year.
Netflix had launched its first streaming video service in January 2007, but it only worked on Internet Explorer browsers and offered just 1,000 movie titles; the company was still mostly a DVD-by-mail business. Cable networks were elbowing into the original series market, but “TV” still mostly meant the broadcast networks.
For as seemingly straightforward as they were, ratings and their interpretation were also an extremely heated topic of discussion in comment sections and other forums. I was the primary ratings person at the site I worked for at the time (the late, frequently great Zap2it), and there and elsewhere people would argue every day about why Show A deserved another season over Show B even though B had a much bigger audience and A was very much on the bubble. Those arguments would also drive traffic and discussion at TV by the Numbers and helped lead to the Cancel Bear’s creation.
Gorman developed the Renew/Cancel Index during the 2007-08 TV season after experimenting with a handful of ways to predict shows’ futures earlier that season (the original explanation for the index is here.) He compared a show’s rating among adults 18-49 — the key ad sales demographic for network TV — to the network average; generally speaking, above-average shows were good bets for renewal, those at or just below the average were tossups, and shows that were more than 10 percentage points below average were likely doomed. For its first couple years, TVBTN also applied the index to unscripted shows before focusing just on scripted ones.
There were a few exceptions to the rule: If a show was in its fourth full season, it was a candidate to be sold into syndcation and got an extra boost in the index. Shows that aired on Fridays had a lower threshold for success since ratings were generally lower that night than on other nights.
The index worked pretty well, as 18-49 ratings (without the addition of DVR numbers) correlated pretty well to which shows survived and which didn’t. While fans of marginal shows would argue vociferously in the site’s comment sections for why their favorites deserved another season and/or why the index was unfair, it was a fairly straightforward calculation.
Then the Bear showed up. As best I can tell, the Cancellation Bear, aka Cancel Bear, began stalking the pages of TVBTN at the start of the 2010-11 season (a post from October 2010, a few weeks into the season, references the Bear, and so do some commenters on other renew/cancel predictions). The Bear’s zombie Twitter account — the source of the image at the top — says it was created in September 2012; I think (but can’t prove) it existed before then.
With the Cancellation Bear persona, TVBTN leaned into a trollish aspect of the ratings discourse. The Bear never wavered in its belief that the Renew/Cancel Index was right, and while the account was careful never to say much about the quality of a show on the verge of cancellation, it would mock fans arguing for their very special show to be saved — at one point, TV by the Numbers even created printable “Fan Excuse Bingo” cards.
People in the industry started to notice — including Kathy Bates, who was then the star of an NBC show called Harry’s Law. It premiered in January 2011 and had a decent-sized total audience, but one that leaned toward older viewers. The Bear gave a grudging thumbs-up to the show in its first season but came after it in season two, when it had NBC’s lowest 18-49 ratings but still a decent-sized total viewer count. “Some of these people are just so stupid,” Bates said in an interview with EW. “I don’t even get it…. All [they] talk about is the blessed [18-49] demo this, demo that, and how the cancellation bear is gonna eat us and all that stuff.”
A couple years later, ABC aired a show called Galavant, a musical comedy from future This Is Us creator Dan Fogelman. It didn’t do that well in its first run, and the Bear predicted a cancellation — along with a bunch of other outlets, because it was one of ABC’s lower-rated shows. The network renewed it, however, and the season two premiere was titled, no kidding, “A New Season aka Suck It Cancellation Bear.” It was canceled after the second season.
In 2010, Tribune Media made a deal with Gorman and Seidman to make TV by the Numbers a partner site of Zap2it — which didn’t really change anything either site did. Zap2it’s listings started appearing on TVBTN, and ad sales covered both sites.
Tribune bought TVBTN outright in 2014, but it continued to operate as usual. Other than the Renew/Cancel posts, Gorman and Seidman had mostly stepped away by then, with other writers handling most of the day-to-day work. In the fall of 2015, Tribune decided to bring the whole thing in house. I had left Zap2it in May of that year when my family moved from L.A. to Austin, but a few months later, they asked me if I’d like to take over the site.
Soon enough, I would become the Cancel Bear. More about that next weekend.
Thank you for reading The Data Stream. If you’re enjoying it, please consider telling your friends about it and upgrading to a paid subscription: For $7 a month or $75 $60 annually (the discount runs through June 30), you’ll get access to every free and paywalled newsletter, commenting privileges on the website and future benefits (once I decide on them). Your subscription helps cover hosting and distribution costs with the Beehiiv platform and could lead to me going fully independent.