Two new data points released in the past 30 hours poured more cold Gatorade on the excitement surrounding the shift from linear TV to streaming for live sports, adding to questions about the short-term financial outlook of the strategy.

First, Amazon released mixed Nielsen and internal viewership data on this season’s “Thursday Night Football” (TNF) broadcast, the first full slate of NFL games shown exclusively on a streaming platform.

Nielsen’s research pegged the average TNF audience at 9.58 million viewers over 15 games, while Amazon said its metrics showed an average audience of 11.3 million. Both numbers marked a significant drop from the 16.2 million people who tuned in, on average, to watch Thursday night games last season on Fox and the NFL Network. National broadcasts of Sunday NFL games often draw 20 million-plus viewers.

Amazon officials on Wednesday did tout the broadcast’s more advertiser-friendly audience this year, which skewed younger and more affluent than last season. They also noted that Amazon viewers stayed glued to the telecast slightly longer. Still, Insider reported Wednesday that viewership fell 25% short of company expectations, forcing Amazon to compensate advertisers with more ad space on other shows and platforms.

Amazon doesn’t release financial performance data related to TNF or its broader Amazon Video ventures (company officials also see TNF as a gateway to signing up Amazon Prime customers). But it’s hard to see how Amazon, which is paying $1.2 billion annually for TNF, emerged a financial winner from the deal this year. For context, consider that Fox paid $660 million annually last season for TNF and shed minimal tears about losing the rights package.

Patrick Crakes, a former Fox Sports executive who now provides media consulting services, told The Athletic that Amazon’s viewership numbers were “a pretty amazing achievement” for a streaming platform, but the tally represented “a pretty big decline” from linear broadcasts.

Across the pond, the results were decidedly more disastrous for London-based DAZN Group, the global streaming company focused on soccer and combat sports. 

Bloomberg reported Thursday that DAZN (pronounced “da-zone”) totaled $2.3 billion in losses in 2021 and has now burned through at least $6 billion, according to financial records reviewed by the outlet. The enormous losses stemmed from massive investments in sports streaming rights—DAZN exclusively broadcasts some soccer matches between teams in the top-flight Italian and Spanish leagues—and comparatively modest subscriber revenues.

The latest reports build on underwhelming returns to date for companies that have invested heavily in live sports broadcasts. 

Disney continues to rack up losses from its ESPN+ offering, which includes some exclusive combat sports, hockey, and soccer broadcasts, among other sports, according to annual financial reports. The company doesn’t break out operating losses tied to ESPN+, but available data suggests it represents a small fraction of Disney’s $4 billion in streaming losses in fiscal 2022.

Live sports also haven’t been enough to prop up teetering streamers Peacock and Paramount+, which broadcast English and Italian soccer games in the U.S., respectively. Both platforms are expected to post losses in the neighborhood of $2 billion this year, leading to speculation about industry consolidation. (In fairness, neither platform’s parent company reports sports-specific financial results, and the vast majority of losses are tied to other programming.)

Industry analysts still argue it’s too soon to deploy the mercy rule on live sports streaming—and they make a few fair points.

On a micro level, the sports streaming experience should improve in the coming years, as video quality gets better and telecasts mature. (A good first step for Amazon: breaking up the surprisingly dull TNF broadcast pairing of Al Michaels and Kirk Herbstreit.)

On a macro level, streaming should keep growing in popularity as younger viewers spurn traditional cable subscriptions. The growth of mobile sports gambling, particularly in the U.S., also bodes well for rights holders. 

But even if there’s a lot of game left to play, sports streamers are starting way behind the profitability sticks.

Want to send thoughts or suggestions to Data Sheet? Drop me a line here.

Jacob Carpenter

Editor’s note: This article has been updated to correct information about Disney’s operating losses related to ESPN+.

NEWSWORTHY

Look what they found. A bankruptcy attorney representing disgraced cryptocurrency exchange FTX said the company has recovered more than $5 billion in liquid assets, a significant increase from earlier accounts, CoinDesk reported Wednesday. The total likely isn’t enough to fully compensate FTX customers who used the exchange, and bankruptcy officials have not yet made plans for liquidating and distributing company assets. A bankruptcy judge set a mid-March deadline for FTX to complete the work needed to estimate an amount owed to debtors.

Screening out more vendors. Apple hopes to begin using homegrown custom displays on its devices starting in late 2024 as part of the company’s effort to bring more production in-house, Bloomberg reported Tuesday, citing sources familiar with the matter. Apple officials want to begin swapping out displays bought from outside vendors, including Samsung and LG, on many of its products in the coming years. Company leaders plan to begin the process with high-end Apple Watch devices and bring the iPhone into the fold later.

Getting bigger in Texas. Tesla has filed expansion plans for its Texas electric vehicle assembly plant, proposing additions that could total up to $770 million, CNBC reported Tuesday. The new facilities would bolster Tesla’s manufacturing and battery cell testing capabilities, among other aspects of operations. Tesla opened its Austin-area electric vehicle and battery factory in April 2022, aiming to produce its line of Model Y and Cybertruck automobiles.

Generating some cash? OpenAI publicly discussed this week the possibility of monetizing its ChatGPT chatbot, citing the need for revenue to ensure its long-term viability, TechCrunch reported Thursday. Through a link posted on the company’s official Discord server, OpenAI officials surveyed users about their willingness to pay for the generative A.I. chatbot and potential perks given to paying customers. OpenAI executives have spoken openly about the enormous computing and server costs associated with their generative A.I. products, some of which could be offset by venture capital and corporate investments such as the potential Microsoft deal.

FOOD FOR THOUGHT

Finally, a blue-chipper? Intel’s much-discussed revival effort received a kick-start this week—albeit one arriving almost two years behind schedule. A New York Times report Wednesday detailed the chipmaker’s stop-start effort to bring Sapphire Rapids, a long-awaited microprocessor, to market after five-plus years of development. Intel officials hoped to debut Sapphire Rapids in 2021, aiming to hit the market with a best-in-class new microprocessor for data centers and stunt the fast rise of rival Advanced Micro Devices. Yet repeated design flaws, a slow testing process, and manufacturing challenges kept pushing back its release.

From the article:

The bumpy development of Sapphire Rapids has implications for whether Intel can rebound to deliver future chips on time. That’s an issue that could affect scores of computer makers and cloud service providers, not to mention the millions of consumers who tap into online services likely to be powered by Intel technology.

“What we want is a stable cadence that is predictable,” said Kirk Skaugen, the executive vice president leading server sales at Lenovo, a Chinese company that is planning 25 new systems based on the new processor. “Sapphire Rapids is the start of a journey.”

IN CASE YOU MISSED IT

Inside the structure of OpenAI’s looming new investment from Microsoft and VCs, by Jessica Mathews and Jeremy Kahn

Sales of Trump NFTs down 99% since mid-December debut generated $4.5 million, by Marco Quiroz-Gutierrez

The A.I. lawyer’s courtroom debut is a stunt, but it raises some big questions, by Alexei Oreskovic

President Biden really wants to boost chip manufacturing and he needs Mexico’s help to do it, by Josh Wingrove, Akayla Gardner, and Bloomberg

Crypto just got its first insider trading sentence, as the 27-year-old brother of an ex-Coinbase worker gets 10 months in jail, by the Associated Press

Seattle public schools’ lawsuit against media giants like TikTok, Instagram and Facebook faces uncertain legal road, by Gene Johnson and the Associated Press

Ripple: ‘Congress must examine the SEC’s role in the recent crypto contagion’, by Stu Alderoty

BEFORE YOU GO

Bamboozled, I say! Roughly this time last year, I wrote about my excitement for the imminent arrival of higher-speed 5G wireless internet in the U.S. Twelve months later, I feel bamboozled, snookered, and generally double-crossed. The reason: I don’t see any real difference between the new and old 5G. It’s apparently a common frustration, according to the Wall Street Journal’s Joanna Stern, who authored a useful guide Wednesday on why the latest wireless internet upgrade isn’t delivering the game-changing speeds we were promised. As Stern noted, the biggest perks of the new 5G are limited to certain areas, including outdoor spaces close to cell towers and crowded areas with high wireless internet demand. Alas, we’ll have to wait even longer for a true wifi revolution.