The video above is nearly 5 years old, but it’s still the best general explainer of TV ratings that I’ve seen. Plus it has puppets! If you do nothing else, I recommend watching the video.
I’m going to address five specific questions below. The first three are the most frequently asked questions I’ve seen specific to sports TV ratings. The last two are more general.
Question: If I stream a sporting event online or through an App is it counted in the ratings?
No. Maybe someday, but not yet. The good news is that most of the TV networks release streaming data and many of them also release the streaming numbers in “minutes of viewing” which allows you to reverse engineer the numbers to apples-to-apples comparisons with the standard TV metric “average viewership.” At least for sports, streaming is still tiny compared to TV consumption.
Question: Are people in bars and hotels counted in the ratings?
No. None of that viewing shows up in the daily numbers currently posted. Nielsen is again working with some clients to test out-of-home measurement. The big issue, as always, is whether advertisers will be interested in paying extra for those viewers. Historically that answer has been somewhere between “no, of course not” and “we have already priced those viewers in to the price we’re paying you” (which, loosely translates to “no, of course not.”)
I’m hearing that Nielsen insists that at some point it will get those viewers into its daily numbers. In the meanwhile both Fox and ESPN are doing some custom reporting. If ESPN has released any info yet, I missed it, but Fox has trickled-out some numbers for FS1 and they are pretty staggering.
A Nielsen custom analysis reveals out-of-home viewing would have added a sturdy 67% to 18-34 viewership for last Friday’s USA-COL Copa match
— Michael Mulvihill (@mulvihill79) June 9, 2016
I hope to write about this topic more in the future. If you are interested in sports media and ratings research you should definitely follow Michael Mulvihill (@Mulvihill79 on Twitter) who heads up research for Fox Sports.
Question: Does DVR viewing count in the ratings?
Yes. If you’re in a Nielsen household, DVR viewing before 3AM ET is captured in the same day ratings that are commonly posted on the Internet (including the daily #s on this site.) The daily numbers are considered “Live+Same Day” or live viewing + same day DVR viewing. Nielsen also has standard reporting for 3 and 7 days of DVR viewing (Live+3 and Live+7). I don’t bother with those much because most of the DVR viewing that ever happens for a live sporting event is captured in the same day numbers. Those +3 and +7 day DVR reports are a much bigger deal for PR people who deal with ratings for scripted and unscripted entertainment programming.
The most important numbers aren’t commonly published: they are the C+3 or C3 ratings which are live viewing plus 3 days worth of DVR viewing of TV commercials. Historically the Live+SD program numbers are a reasonable proxy for those numbers, are published sooner and are more commonly available.
AdAge’s Anthony Crupi (@CrupiCrupiCrupi) is a good person to follow on Twitter for TV advertising in general and he sometimes does stories on the C3 numbers, too.
Question: Nielsen is BS! I mean paper diaries? C’mon, it’s 2016!
I’m not an apologist for Nielsen, but while it isn’t perfect, many of its foibles (like paper diaries) are overplayed. Paper diaries do not factor into any of the commonly reported national Nielsen numbers, including the +3 and +7 DVR numbers or any numbers you’ve ever seen posted on sites like this.
Even Nielsen’s customers, maybe especially its customers, take shots at Nielsen! But Nielsen gets blamed for things that aren’t necessarily Nielsen’s fault (like what advertisers do and don’t want to pay for.) But Nielsen is paid a lot of money so no need to feel sorry for it.
Nobody has come up with a system that’s better yet.
The Nielsen question has its own sub-frequently asked questions like:
1. why don’t they just do a census of everybody?! (too expensive, would take too long, not everybody wants to be measured and there’s reason to doubt whether it would change the numbers very much.)
2. why don’t they just use set-top box data!? (not everyone has a set-top box, not everyone wants to be measured and set-top box data doesn’t tell you who or how many in the household were watching, and Nielsen’s method generates numbers faster. That said, there are companies like Rentrak that sell set-top box data.
Question: I watched the puppet video but I’m still not quite sure what a rating or a share is?
Nielsen TV ratings are merely percentages of whatever is being measured. Sometimes it gets confusing because so many different things are measured.
if you see a 7.1 household rating that just means on average 7.1 percent of U.S. TV households were tuned in, so at least as of today that would mean 7.1 percent of 116.4 million U.S. homes with televisions or around 8.3 million households.
if you see a 7.1 “overnight rating” that’s sort of similar to the example above but instead of being the % of the 116.4 million TV homes in the U.S. it is the percentage of the TV homes in the largest 56 (out of 210) markets.
if you see a 3.2 adults 18-49 rating, that just means 3.2 percent of the 126.81 million adults who are 18-49 years old so a 3.2 adults 18-49 rating works out to around 4.1 million adults 18-49.
Ratings are a percentage of the total population being measured (be it households, adults 18-49, adults 18-34, kids 12-17 etc.) whether they are watching TV at the time or not. Share measures the percentage of people watching TV at the time. So if you see a household rating/share of say, 3.7/7 that just means 3.7% of total households were tuned in and 7% of the households watching TV at the time.*
*in the era of DVR viewing the “share” calculation is a little tricky, but the explanation above should work well enough for general purposes.
If you have questions not covered above leave a comment or ask me on Twitter via @SportsTVRatings.