Sport, the churn reducer

Sports broadcasting might in recent years have seemed to be completing a four-decade march to subscriptions, but subscriptions are right now starting to look like a phase rather than a destination. The platforms that gave us the streaming era have been reinserting ads for three years.

In April 2022, for the first time in a decade, Netflix suffered a quarterly decline in subscribers: some 200,000 dropped off from its then total of 221.84 million.

Co-CEO Reed Hastings, until then a champion of ad-free streaming, announced plans for a new, lower-priced tier that would carry ads. The reasons? “[O]ur relatively high household penetration – when including the large number of households sharing accounts – combined with competition, is creating revenue growth headwinds,” as the Q1 Shareholder Letter reads. In truth, though, the subscription model had hit its ceiling.

Disney had brought Disney+ to market in 2019, charging $6.99 per month and proclaiming the absence of ads. Three years later, in December 2022, it began to follow Netflix’s lead. Both companies were dealing not with consumer demand but with stalled growth in high-income markets.

And what of sport?

The first revenue model for sport was the ticket. The second was broadcasting rights sold to advertisers on free-to-air television – a model that, especially in the US, made sport the most commercially valuable of spectacles. The third was cable, which introduced subscription revenue, but the customer paid for it in bulk. The monthly cable bill had sport as part of the bundle, its cost invisible. The fourth model, proper streaming with direct, disaggregated subscriptions, has made the cost of sport visible and countable. And with this fourth model has come fragmentation, the bane of any fan trying to watch a full season of games.

Estimates for the cost of watching all NFL games for the 2025 season in the US run from $935 to $1,500: high enough for the US Federal Communications Commission (FCC) to open a formal inquiry into the fragmentation of sports rights. The commission’s request for comment on the matter lays out the history of sports broadcasting.

Sinclair Broadcast Group’s filing from March with the FCC reads in part as follows:

The increasing migration of live sports to streaming platforms has introduced significant fragmentation that frustrates consumers. One survey shows that 87 percent of sports fans find it at least somewhat frustrating to figure out where to watch the games they want to see, and almost 25 percent feel very frustrated. Nearly two-thirds of sports fans say it’s a “hassle” to use multiple services to watch games during a season, and half say that it has become harder to find the games they want to watch compared to a year ago.

This fragmentation has consequences for viewers. Almost two-thirds of fans say having games on different platforms makes it more difficult to check on other games being shown at the same time. Further, nearly half of sports fans report missing games because they do not subscribe to the correct service, and 44 percent say they already subscribe to too many platforms.

Structural differences

Subscription revenue is bounded by the number of households and the magnitude of their disposable income. Advertising revenue, by contrast, scales with corporate spending; its relation to population is indirect. In theory ad revenue is bound by the economy at large. A big enough audience ought to make advertising the more lucrative model, and subscriptions would seem to have served as the path to a new era of advertising. The platforms, after raising prices as far as the market could bear, have found a new layer of revenue.

Since Netflix introduced its ad tier (November 2022) ad-supported streaming has gone from experimental option to structural default. By Q1 2025 nearly half of all US premium SVOD subscriptions were carrying ads.

Amazon’s contribution, in January 2024, was an opt-out switch. This proved to be the inflection point for scale, establishing in no time the world’s largest audience for ad-supported streaming.

Then, in August 2025, the industry’s largest dedicated sports streamer, ESPN, went DTC without an ad-free option, having presumably concluded that no ad-free sports tier could survive. Dedicated fans who wish to avoid ads on any major US sports streaming platform must already pay a premium, but on ESPN they must bear with ads like everybody else.

According to eMarketer, ad-supported subscriptions accounted for 57 percent of gross subscriber additions in premium SVOD for Q1 2025, and 86 percent of subscribers offered an ad-supported option have accepted it, while 75 percent of all streaming users have tried an ad-supported plan in the past four years.

But there’s a brake even on ad-revenue, and it too is applied by customers. At some point of ad saturation, presumably, either customers will walk away or streamers will end up opening the market to other, different players.

Another alternative, but a revenue layer?

ESPN+, which went live in April 2018, streamed live sports from the start, but the first exclusive national-league package on a streaming service was almost coeval with advertising on SVOD. It was Amazon Prime Video’s Thursday Night Football, which dates to September 2022. There was even a preseason game the month before. Netflix entered the picture in November 2023.

Sports streams generate revenue, of course. For one thing, they drive subscriptions. For another, they drive subscriber retention, because fans want to watch their team all season long. But sport is not like movies or TV shows.

Rights fees are any broadcaster’s single largest expense on content. The NFL alone commands about $10 billion per year from its various broadcasters and streamers in the US. By the 2021 estimate of CNBC, which cited “people familiar with the matter,” Amazon is paying about $1 billion per year to the NFL. In 2018, according to the New York Times, Netflix agreed to pay $100 million to stream the sitcom Friends in the US for a year. On the face of it, then, sports are an order of magnitude more expensive than TV. But they’re in fact more expensive than that.

The Amazon deal does not cover the NFL’s more than two hundred annual games. It covers 16 games in the regular season and one in the preseason. The playoffs and the Superbowl are separate. Friends includes 236 episodes over ten seasons.

Then there’s rewatchability. Highlights aside, live games and matches are dead content on the morrow, but people are slow to tire of Chandler and will get their children to watch – on the multiple, separate screens that children now have.

Rough reckoning yields this:

NFL rights per broadcast hour:

• Amazon TNF: about $1 billion ÷ 16 games × about 3 hours/game = roughly $21 million per broadcast hour
• ESPN/ABC MNF: about $2.7 billion ÷ about 18 games × about 3 hours = roughly $50 million per broadcast hour

Friends at peak licensing:

• $100 million ÷ 236 episodes × about 0.37 hours/episode = roughly $1.1 million per broadcast hour

NFL rights cost between 20 and 45 times more per hour of content than history’s most expensive licensing deal for entertainment. The ratio is greater per game/episode:

• Amazon TNF: about $62 million per game
• Friends at peak: about $424,000 per episode

That’s about 150 to 1. It would be interesting to see the ratio per viewer, but no streamer reports clear figures on viewership. What is clear is that sports are not keeping streamers afloat. Entertainment catalogues are. Sports are instead helping to reduce what the industry calls churn: subscriber losses.