The Emmys Drew 6.7 Million Viewers. Whether That Is Down 9, 10 or 12 Percent Depends on Who Is Dividing
The 78th Emmys drew 6.7 million viewers and the same number was reported as down 9 percent, down 10 percent and down 11.7 percent. All three are arithmetically correct, which is the problem.
Every outlet covering the ratings for the 78th Primetime Emmy Awards printed the same viewership figure: 6.7 million, across NBC and Peacock, for the ceremony Mariska Hargitay hosted on 14 September. Then they printed three different declines.
Variety said down nine percent. Deadline and the Los Angeles Times said down ten. TheWrap said down 11.7. Nobody is wrong, and only one of them showed the working.
The trick is in the denominator
Last year's ceremony has two viewership numbers, not one. The 77th Emmys initially came in at 7.42 million on Nielsen's fast nationals. When the final nationals landed, that was revised upward to 7.59 million — a four-year high, up ten percent on 2024.
Divide this year's preliminary 6.7 million by each of those and you get the whole spread:
- 6.7 ÷ 7.42 = −9.7%. Round it down and you have Variety's nine. Round it up and you have Deadline's and the LA Times's ten.
- 6.7 ÷ 7.59 = −11.7%. That is TheWrap's figure, and TheWrap is the one outlet that stated its base in the copy: down from "the 7.59 million viewers brought in by the 77th annual awards ceremony last year," while noting that last year's initial number was 7.42 million before adjustment.
Same numerator, two denominators, three headlines.
Which comparison is actually fair
The largest number is the least like-for-like. This year's 6.7 million is preliminary; it has not been through the revision that pushed last year's figure up by 170,000 viewers. Measuring a preliminary against a final overstates the fall, because the preliminary will also move.
The honest comparison right now is preliminary against preliminary — 6.7 against 7.42 — which is a drop of about 9.7 percent. Which is precisely why "nine percent" and "ten percent" are both defensible roundings of one sum, and why the two outlets that disagree with each other are in fact agreeing.
TheWrap's own copy says the story will be updated when final numbers land. It is the correct caveat and the largest decline, at once.
The record-low headline going round is four years old
A Fox News story headlined "Emmys reach record-low audience with 5.9 million people" keeps resurfacing next to this year's coverage. It was published on 14 September 2022, about the 74th Emmys, hosted by Kenan Thompson on NBC, which fell from 7.4 million the year before.
6.7 million is comfortably above 5.9 million. This was not a record low. It was the third-best Emmy audience since that 2022 floor.
The figures that hold up
- 6.7 million total viewers, NBC plus Peacock, preliminary Nielsen. Every outlet agrees.
- 7.42 million preliminary and 7.59 million final for the 77th Emmys in 2025, with a 1.28 rating among adults 18–49. That demo figure belongs to 2025 only; no 2026 demo number has been published.
- Up roughly ten percent on NBC's previous Emmys telecast, in 2022. The Emmys rotate between networks, so NBC's last outing is the like-for-like comparison nobody reached for.
- Peacock's simulstream posted the highest reach of any non-holiday NBC entertainment special on the platform, behind only the Thanksgiving parade.
- The ceremony moved to a Monday this year, which is not nothing when comparing to a Sunday telecast.
What is not known yet
- The final 2026 number does not exist. Expect the 6.7 million to be revised upward, which will quietly shrink all three percentages.
- The 2026 rating among adults 18–49 has not appeared in any reachable source.
- Deadline's and Variety's exact wording are behind paywalls; their headline figures are confirmed, their reasoning is not.
Three numbers, one ceremony, no dispute
There is a version of this story where outlets are sloppy and a version where they are contradicting each other. Neither is what happened. What happened is that Nielsen produces two numbers for every telecast, weeks apart, and a story written the morning after has to choose which one to compare against — and almost nobody says which they chose.
The decline is real either way. The Emmys drew fewer people than last year, in a year when the ceremony moved nights and the streaming half of the audience grew. It is a nine-ish percent story that got reported as three different stories, because a percentage without its base is not a fact. It is a ratio with a piece missing.