The mirage of rap's popularity: How streaming stats mislead vs. sales and airplay

Looking over the top 100 songs for 2020, I was struck yet again by how much rap there is. I've noticed this for the past several years, and figured I was just out of touch with that area of pop music, since I don't listen to rap radio stations, don't know anyone who's really into rap, and don't go to clubs that play rap. Still, it just seemed like a growing presence that was not my cup of tea.

But having studied the seismic shift that streaming has had on the music culture, I can finally make some sense of it. Rap is in fact a niche genre, with a few crossover hits to mainstream audiences. However, the media elites who construct the chart formulas have put their thumb on the scale in order to over-emphasize rap's popularity and influence, at the expense of truly popular genres like pop, country, and dance.

The formal trick they use is giving substantial weight to streaming stats, which do not distinguish between breadth vs. depth of exposure. And popularity is about breadth of exposure across the entire music-listening population, not the depth of devotion among fans who expose themselves repeatedly to the song, as opposed to casual fans who expose themselves to it far fewer times.

So if you too were wondering why there's so much emphasis on rap in the media, you weren't crazy. It's another example of woke representation practices, meant to buy off the "talented tenth" of African-Americans, who get employment in the entertainment sector, and placate the bottom 90% of them with cultural cred -- rather than the entertainment and media elites using their high status to lobby the Democrats into providing desperately needed public goods and services.

* * *

First, a clarification -- the Billboard charts are a product of the media sector, not entertainment. The record labels produce the music, the artists perform it, and various other channels distribute it to the audience (radio stations, clubs, streaming platforms, etc.). Billboard is more of a trade publication than a cultural commentary one, but it's still part of the media sector, albeit the entertainment-focused media.

So, this is not a case of the production side making a ton of rap songs that nobody wants to listen to. Nor is it the distributors taking a niche genre and foisting it on users of their platforms like YouTube, radio stations, and so on. Those two sides are too driven by the cold, hard laws of supply & demand to attempt to deliver a bunch of stuff that is largely unwanted.

However, the media who describe and comment on entertainment are under no such constraints. Or rather, their supply & demand laws are different because their audience is not listeners of music per se, but readers of music-themed discourse. Their audience wants to read takes, and spit out takes of their own, which is orthogonal to what types of music they enjoy listening to.

It's perfectly possible, then, for Billboard to mischaracterize the state of supply & demand in the music industry, if doing so will satisfy the cravings of discourse junkies. It would be wrong to call this "foisting" their narrative onto the music-listening public, because the typical music fan probably never looks at the Billboard charts. Those who do consume media commentary on music, though, evidently eat up the narrative about how influential rap is, so the media outlets are not foisting the misleading description onto their audience either -- they're just supplying the demand for a certain narrative.

But if you do want an overall accurate picture of what the zeitgeist is like, out of curiosity, just bear in mind that the media chart creators can and do rig the outcome in order to please their target audience above all else. It won't be totally outta whack, since such a picture would not even be plausible, and the audience wants the illusion of reality as well as the ideologically soothing distortions. Still, something to take into account.

* * *

Now, the basic problem. For the 2020 Hot 100 chart, I count 30-some rap songs. Not a majority, but still a sizable share. And to anyone who's been in touch with music this year, way too many. What gives?

Well, the Hot 100 chart is actually made up of several component charts. The three with heaviest weighting are sales of singles (i.e. digital downloads), radio airplay, and streaming plays. As of 2013, these carry weightings of 35-45%, 30-40%, and 20-30%, respectively, and the weightings change week to week.

The shifting weekly weightings is the first obvious sign that the chart creators are using these stats to rig the desired outcome, otherwise the weightings would stay the same in order to judge all songs by the same set of standards. In one week, one weighting will accomplish the goal of maximizing rap at the expense of pop, dance, and country, while a different weighting will be needed for another week, since each week's batch of songs perform somewhat differently relative to one another.

For example, if the sales stats favor pop over rap by a huge amount in week 1, and to a lesser degree in week 2, then the chart-riggers will have to give a lower weight to sales in week 1 than in week 2.

Let's turn to Billboard's Digital Songs sales chart, of which their website lists only the top 75. For the 2020 chart, I count about 10 of the rap songs that are also on the Hot 100 chart, and several others that are not. Scaling that to a list of 100 by sales, that would be about 15-20 rap songs -- in other words, only half as many as actually appear on the Hot 100 chart.

Not only is rap less pervasive on the sales chart, but big hit songs you've been hearing all year do in fact show up, while they're mysteriously missing from the Hot 100 chart. "Kings and Queens" by Ava Max, "Midnight Sky" by Miley Cyrus, and "Stupid Love" by Lady Gaga in the dance genre. "Lover" and "Cardigan" by Taylor Swift in the adult contempo genre. And a slew of country songs I don't recognize because I'm not in the target demographic. Again, that's only for the 75 songs listed on their website; if they had a full list of 100, there would be more staples of the zeitgeist in those genres that were kept out of the Hot 100.

When I looked over the sales chart, it felt 10 times more familiar than what's listed on the Hot 100 chart. And that's not just from what I seek out deliberately -- it's from what I hear in any public place that plays music, what's on while changing radio stations, what's popular on Tik Tok trends, what a popular streamer like Pokimane plays in the background while chatting, what anyone is talking about, what online memes refer to... literally every source of pop culture other than the music media itself.

There is a heavy overlap between the sales and Hot 100 charts, since the big-picture narrative from the Hot 100 chart cannot be totally divorced from reality. Still, it's striking how disoriented a normal music listener -- who doesn't care about woke ideology being reflected in the list -- would get from there being a sizable minority of fake rap songs shoehorned in, and the same number of actually popular dance / pop / country songs erased from the record.

* * *

As for radio airplay, Billboard doesn't list their top 100 songs online, but by genre it must reflect the distribution of radio station listenership size by genre (called a "format"). See Nielsen's overview of the top radio formats at the end of 2019.

There is no dedicated rap or hip-hop format, but they are included under the urban contemporary and (to a lesser degree) the urban adult contemporary formats. So however well those formats do, rap must do worse, since urban contempo (and especially urban AC) also includes a bunch of R&B songs.

Among music stations playing current music, the formats with the largest listenership size are adult contempo, country, and pop, followed by hot AC and urban AC, with urban contempo lagging down with Mexican regional. Notice that this is the opposite ranking of which genres are artificially boosted on the Hot 100 chart.

Even more telling is the fact that Mexican regional does not get over-represented on the Hot 100, despite the potential wokeness cred that the chart-riggers could enjoy from doing so, and despite it being as popular or more so on the radio than rap. However, the cultural commentators and their audiences don't find Mexican culture interesting, other than the food, and they don't feel as strong of a need to "heal historical traumas" or "correct the historical record" by amplifying recognition of Mexican culture today, as compared to African-American culture.

The same goes for the Spanish contempo format, which is about as popular as rap (if not urban contempo as a whole, due to R&B's popularity), and is as popular as alternative. But listening to reggaeton enough to have an opinion on it one way or another, let alone actually dancing to it in a club with thicc-booty Cuban girls, would absolutely mortify the dorky white liberal males who control the music-themed media sector, and make up most of the audience as well.

"Spanish music" is still too dance-oriented and corporeal for it to appeal to the cerebral types who are take junkies and media consumers. There's no pretension in the lyrics about "telling a larger story," "raising awareness," etc., and there's no raw angsty attitude like in modern rap, which in many ways has become the black version of punk -- non-musical angst, verbal focus, and a basic beat without melodic instruments.

When rap is (slightly) more melodic, dance-oriented, and aimed at dudes and dudettes grinding on each other in a club, rather than individuals stewing in angst alone in their room, music media people lose all interest. Suddenly it's just a pretext for animalistic booty-shaking. Like all good cerebrals, they only condone sex-having and lust in music if it's centered around boobs (elevating) rather than butts (sinful).

* * *

That leaves only streaming as the component of the Hot 100 that must be artificially boosting the popularity of rap, and diminishing that of pop, dance, and country. This component reflects all the big platforms -- YouTube, Spotify, et al.

The first problem with streaming is the youth bias: 55% of Spotify users are aged 18-34. And in Nielsen's radio listenership article, urban contempo skyrockets to 4th place in that age group, while Mexican regional and Spanish contempo lag far behind. Giving greater weight to streaming means giving rap more representation compared to pop, country and Spanish-language music.

But there's a far greater distortion that comes from using streaming stats. Billboard uses the total number of plays across streaming media (in the US), which could be a lot of people listening a few times each (a broadly popular fad song), or a few diehards listening to it over and over again (a niche song that will last forever in the fanbase).

That makes streaming unlike sales of songs, where one sale could be listened to one time or one million times -- it's still measuring the song's exposure to just one individual. Diehard fans do not skew the overall results by listening to their purchases many times more than casual fans who bought the same song. Total sales of a song = total individuals exposed directly.

And it's also unlike radio airplay, which takes audience size into account. They use the share of radio listeners tuned into a certain station for at least 5 minutes during a 15-minute interval. If a song is played in that interval, that audience size is about how many people were exposed to the song.

Even if someone tunes into the station and hears that song every day, it doesn't skew the overall results because they aren't adding to the audience size by tuning in every day -- it's roughly the same size from one week to the next. Diehard fans may be tuning in every day, while casual fans only listen some days of the week -- but the total size stays about even because some casual fans who are absent on one day are made up for by other casual fans who are tuning in that day. (And when those casuals tune out later in the week, the casuals who were tuned out earlier in the week show up to replace them.)

So if anything, radio airplay is skewed more by casual listeners than by diehard fans, since the total number of casual listeners is hardly all present on any given day. Thus, the station's listenership is larger than it would appear from a snapshot in time. And assuming some song is a regular in the station's playlist throughout the week, it's reaching a broader audience than it would appear from a snapshot. Like sales, radio airplay measures breadth rather than depth.

The streaming stats could be made accurate by measuring the total number of unique individuals who played the song during a given interval, regardless of how many times they played it. Spotify could do that, since their users have to have downloaded the app and be signed in. But YouTube does not require you to even have an account, let alone be signed in, to search for and play videos. And big hits get in the 100s of millions of views on YouTube (albeit globally), so that is no small problem with aggregating streaming stats across all platforms.

I guess YouTube could try tracking how many unique IPs within the US played a video in a given interval, regardless of number of times played. But they're not doing that.

And the larger point is they don't want to -- it would ruin their goal of over-representing rap in the Hot 100 chart, to construct the narrative of how influential a certain part of African-American culture is in the broader society. Using youth-biased streaming stats, with a substantial and shifting weighting, and measuring total plays rather than audience size, are just the technical means toward the end of narrative construction.

If you think they aren't aware of these problems, you think they're stupid, and these people do not have low IQs. Even if you didn't think of the breadth vs. depth issue beforehand, your BS detector would be screaming when you checked the results of your algorithm and saw 30-some rap songs on the top 100 for the year. "I knew there'd be some, but not this many -- something's wrong with our formula."

Any naive techie geek who pointed these issues out would be gently ignored by his managers, and fired if he pressed on it. If these problems were solved, the results would only have 15 rap songs out of 100, and 20 more songs from pop, dance, and country -- ummm, lame mayo music much? Gotta get more hyped-up rap songs in there somehow, or else our descriptive narrative won't fly with the target audience of woke take junkies.