Super Bowl Viewership Statistics

I gave the students in my media literacy class two different stories to read about the 2024 Super Bowl viewership statistics. One was from the Financial Times (FT), the second - from the NFL website. This was an exercise in reading the stats critically.

I have assigned these stories because statistics need to be always presented in a proper context and because I've seen too many media stories comparing Super Bowl LVIII viewership to the 1969 Apollo moon landing.

The FT story started out with,

A record 123mn US households watched Sunday’s Super Bowl game, making it one of the biggest broadcasts in the country’s history and underscoring the enduring pull of live sports even as the television landscape craters beneath it.

This reporting of viewership is obviously too high, as there's only about 131Mn households in the US today, and it is hard to believe that 94% of all American households watched the game. The mistake has probably crept in from the confusion about how Nielsen, which is still the dominant force in audience research, compiles its viewership data - it is a composite of both household estimates and individual watching of the game on all platforms, which the NFL website rightfully pointed out.

Here's the first paragraph of the NFL story:

CBS Sports' coverage of Super Bowl LVIII, which featured the Kansas City Chiefs' 25-22 overtime victory over the San Francisco 49ers, delivered the most-watched telecast in history with a 123.4 million average viewers across all platforms, including the CBS Television Network, Paramount+, Nickelodeon, Univision, and CBS Sports and NFL digital properties, including NFL+.

This is a more accurate reporting of "average viewers" (i.e. individuals, people), and as there's about 338 Mn people living in the US now, it gives us about 36% of the population. The same story gives us an estimate of about 202 Mn people who watched the game either entirely or in part (eg, people like me, watching the half-time show and checking in occasionally for the scores), and that gives us about 65% of the population. These are much more realistic stats.

I would still not call the game "the most-watched telecast in history" for several reasons:

1) Whose history? The US? or the world? (Olympic Games and World Cup finals since 2008 command billions of viewers).

2) "Telecast"? That's an olden term that even broadcasters are avoiding to use as most technologies are now using digital technologies and deliver their content on various platforms, not just TV.

3) this wording suggests inclusion of all events, not just sports, and this is where a lot of media are using comparisons with the moon landing.

Here's the quote from the FT story again, describing the game as:

the second most-watched US television broadcast in history, behind only the Apollo moon landing in 1969, which is estimated to have drawn between 125mn and 150mn US viewers when the country’s population was just over 200mn.

Let's look at this closer, shall we? Even if we take the middle point of the 125-150Mn estimate (in 1969, the estimates of viewership were not as exact as Nielsen claims today), let's say 137Mn, the population of the US at the time was 202Mn. Therefore, the moon landing viewership was about 68%. So, the FT description is more accurate than the NFL website, which must have taken the lower end of the bracket - 125Mn - to make the Super Bowl the most watched event (a much better word than telecast or broadcast) in the US history.

In general, comparing the moon landing - the event of national and global importance in the Cold War context - covered by TV, the only truly dominant mass medium of the time that people were almost literally glued to, with the beloved sports game in the modern context of multiple digital media platforms and over-saturated media environment is unfair. A media literate person should question and doubt these types of comparisons.

This coverage is also a good illustration of how it is important to frame the absolute numbers in the proper context of totality. 123Mn is indeed a large real, or absolute, number. However, it "shrinks" considerably if you look at it as a proportion of the total 338M population (36%) compared to 125Mn of the total 202 Mn (62%) in 1969. It's almost two times lower!

The original stories are here: