It’s rare that the aftermath of a major sporting event can provoke more interest and intrigue than the event itself. But given the bar that was set by the thoroughly uncompetitive exhibition, “sanctioned” or not, between Jake Paul and Mike Tyson which Netflix streamed to the world Friday night, there was more than a puncher’s chance that could happen–ironic considering not a whole lot of punching occurred during the sixteen minutes of actual “combat” people struggled to see.
As so many apparently experienced first hand Friday night, the mere act of airing the event proved to be more of a challenge for Netflix than a 58-year-old version of Tyson provided for the far younger albeit relatively inexperienced Paul. THE ATHLETIC’s prolific Richard Deitsch summed it up as coldly and plainly as the current climate in his Canadian home in his recap that dropped Sunday:
The Paul-Tyson bout was horrible, and so was the streaming experience for many viewers. As my colleague Tess DeMeyer chronicled, viewers were plagued by frequent bouts of buffering and freezing. There were technical issues in the broadcast, with Evander Holyfield’s earpiece and Jerry Jones’ microphone malfunctioning during separate interviews.
Streaming issues of course vary depending on multiple factors, including internet connectivity. But there were loads of viewers who experienced problems Friday night…Netflix has over 280 million subscribers in more than 190 countries…The website Down Detector noted nearly 85,000 viewers logged problems with outages or streaming leading up to the fight.
And that provided a jumping-off point for self-described Streaming Media Expert: Industry Analyst, Writer and Consultant. Chairman of the NAB Show Streaming Summit Dan Rayburn to go off on an epic rant that dominated most of my LinkedIn feed yesterday, and given the dozens of comments and reposts it drew I suspect some of yours as well.
Facts matter, except apparently in the aftermath of Netflix’s boxing event. This was Netflix’s 8th live event (open to correction), not their first. Everyone is an expert in CDN and can “fix” Netflix’s live streaming, even though they don’t know the problems. Multiple vendors claim their cloud/edge/P2P/”insert anything here” solution could have solved the problem when some vendors are in beta with zero deployments. Many people share and compare numbers that are completely made up but state them as fact. Having a fruitful conversation on the topic is hard since many refuse to take the time to educate themselves on the real numbers of past events. The volume of inaccurate posts I’ve seen on LinkedIn wastes people’s time.
Rayburn then rattled off an airing of grievances that would have made Frank Costanza proud. Some of them are more than valid; a few are downright niggly. And I’m anything but a technical expert myself, so I fully identify with his observation about not knowing enough about the streaming problem to fix it. But when it comes to the alleged overstatement of audience estimates, I can certainly offer more than a few detailed insights of my own.
And it happened to be an issue that was amplified by one of the few remaining true research experts capable of generating his own press, as NBC SPORTS’ website detailed yesterday:
After Netflix issued a press release claiming that 60 million households tuned in for the main event on Friday night, Fox Sports president of insights and analytics Mike Mulvihill had this to stay, via Sports Business Journal: “Nielsen is fully capable of producing a U.S.-only viewership number for Netflix on a next day basis, but they can’t do it if Netflix doesn’t ask for it. And why would [Netflix] ask for it when the entire Internet runs with their worldwide number no questions asked?”
We couched our item on the numbers as a Netflix claim, adding that they didn’t use the traditional average-per-minute audience figure. That’s the traditional way of measuring viewership. Mulvihill goes a step farther, not-so-subtly suggesting that the numbers are fugazi.
Well, they may very well be. And there’s certainly precedence for the media being hoodwinked by blurred lines of spin coming from streaming companies crowing about big numbers. This was how THE WRAP’s Tony Maglio reported on the NFL’s first-ever streaming-exclusive broadcast the morning after it happened back in October 2015:
The NFL is a hit — even when two weak teams play American football online in a foreign country.
The Buffalo Bills versus Jacksonville Jaguars Sunday game in London scored 15.2 million unique viewers on Yahoo. The online network actually expects that number to grow by about one million viewers once all buckets are counted, including local TV markets here, in the U.K., and digital China figures.
The game generated 33.6 million video streams. About one-third of that big audience was international, the digital company touted, with 185 countries having had access. All-told, the U.K. match saw more than 460 million minutes consumed.
But a mere few hours later CNN’s Brian Stelter, after having consulted with his own in-house research guru and his team who had already smelled a rat in a similar manner that Mulvihill did on Friday, dropped what can only be described as a mansplain for those otherwise too uninformed or ignorant to know the difference:
“Over 460 million total minutes of video were consumed” during the 195 minute game, the partners said, which implies an average viewership per minute of 2.36 million. An NFL spokesman confirmed the figure. Afternoon and evening NFL games on TV average 10 to 20 million viewers per minute.
The Netflix reach number for global households can be compared to the 15.2 million unique viewers which Yahoo! touted. And depending upon how many viewers per household were watching the fight (if you’re to believe the many complaints from folks throwing watch parties that contributed to Rayburn’s comments, it may have been way higher than the usually assumed 2.5 standard), the Paul-Tyson number was likely into nine figures.
That said, Statista notes that as of last quarter Netflix had 84.8 million paid domestic subscribers–roughly 30 percent of the 280 million number Deitsch referenced. And as NBC SPORTS accurately notes, it’s entirely possible Netflix may have employed further goosing: Did, for example, Netflix treat each effort to reload the stream as a separate view? It’s almost a certainty that there were a lot more than 85,000 streams that were impacted in such a way.
Mulvihill’s obvious frustration is amplified by the reality check that several Nielsen executives confirmed to me that even if he had wanted to pay for a data run of a measured competitor–which many of us have done at times (a list that included at one point his one-time corporate colleagues who worked for Roger Ailes’ division so as to get numbers on CNBC’s daytime business news when FOX Business News began), the limitations of Nielsen’s streaming meters can only produce a “destination” number for the platform’s entire viewership, as opposed to a program-only number for the fight. In order to get an actual number for the fight itself, it would have had to have been “watermarked” so that it could have been uniquely identified. In theory, a stakeholder could have done just that.
But in this case there were no advertisers, and the sponsors involved were mostly attached to Paul’s digital world where they accept YouTube views as a viable metric. And as we’ve documented many times, Nielsen’s panel footprint still only reports “connected screen” consumption, which is even more of an issue for younger viewers who watch on a device, an even greater likelihood given the age bracket and technology leanings of Paul’s fan base.
As Mulvihill asserts, Nielsen was capable of producing an apples-to-apples number, and perhaps if someone like him had leaned on them in advance to do so, perhaps even offering to pay for the watermarking as well as the data, they might have been open to doing so. It’s entirely possible that conversation was had. But from Nielsen’s standpoint, they do have a deal in place with Netflix to measure their upcoming Christmas Day NFL games, augmenting their panel data with first-party proprietary consumption from Netflix akin to the way they are providing Prime Video with their Thursday Night Football figures. No chance Nielsen would want to jeopardize that incremental revenue, especially at a time when Paramount Global isn’t a paying client.
I know Mulvihill and FOX do care about how accurately those games are being measured and reported on, as they stand to lose street cred and, down the road, possibly some games if Netflix’s first foray into the NFL camp is perceived as more of a success than it may actually be.
So maybe start the process of ensuring less controversy then than now by demanding, or even paying to order, a cutback number from Nielsen that shows only the Netflix viewership from the current Nielsen panel? Maybe work with a couple of advertisers to put on more of a full-court press? Maybe start with some of the data points above, plus a few I wasn’t able to think of after just waking up, to at least call into further question the fact that on a level playing field what remains of the press–especially those read by Wall Street types–might just be giving Netflix way more credit for record-breaking than they actually achieved?
Possibly add to that some sort of comparison of Netflix’s overall Friday night viewing last week versus that of recent weeks’ to at least get some sort of a gauge of measurable lift–allowing for that fact that the majority of their catalogue is consistent and while that growth can’t directly be attributed to the fight it can at least be reasonably argued that a great deal of it was because of it? It’s an admittedly flimsy argument, but look what the press has already bought into.
And add to that the kind of chiding that Rayburn’s rant concluded with:
For those making up numbers and publishing them without sources, and you have the title of “Research Director” or “Analyst,” shame on you. Playing fast and loose with numbers is not acceptable.
Maybe Tyson wasn’t able to put arrogant Gen Zs and Gen As in their places. But we still can.
Until next time…