CMJ Scandal!

I was reading the Wikipedia entry for CMJ and I have to share. Toward the end of CMJ's existence there was a scandal. Well not much of a scandal but that's what someone on Wikipedia called it.  [SOURCE] So let's quote that passage, as it appears in Wikipedia:

Minor scandals followed. CMJ was accused of manipulating their charts in order to push their own compilation into the Top 200; however, CMJ claimed it was an accident and the compilation was only used as a placeholder. This resulted in CMJ changing the name of their New Music Report compilation from Certain Damage to On Air. In addition, the magazine was criticized at the time by many in the independent music community for focusing too much on major label acts, which resulted in Beggars Group pulling ads from the publication.

So this is partly accurate. It cites as a source, earshot-online [LINK] written by one Michael Barclay. Barclay is not a bad egg. He's a legit music journalist in Canada. (He has a substack now) Canadian college stations did report to the CMJ chart so Canada did have a horse in that race. I should disclose that Earshot also published it's own charts which is a conflict of interest, as far as criticizing other music charts but I can confirm what they reported is materially accurate.

First lets discuss how charts work. Hundreds of college radio music directors submitted top 30 reports to CMJ every week. They were mostly albums, though some singles were sprinkled in. This was by phone and fax back in the 80s, and gradually email submission became the dominant method by the mid-1990s. The problem is that Johnny can't count to 30. A surprising number of kids would submit charts of 29 records or fewer. They simply miscounted, or accidentally included duplicates. In the plain text environment of free Hotmail or Yahoo accounts it was easy to miss. Spell check and grammar check existed but the context-sensitive spell checkers that could detect a number sequence came later.

I personally knew the fellows who edited that chart. I won't name names, but in a well-intended effort to include more stations reports, i.e. net greater accuracy, they would insert the most recent Certain Damage compilation as a place-holder for the missing album. This was such a small number of corrections that they did not expect it to affect the top 200 report at all. The average week incorporated more than 1,500 charts per week, so it was a drop in the bucket. But then it did show up. Oops. 

The reason some indie labels called foul, was that record labels had to pay to be included on the Certain Damage compilation. From the outside it looked kind of like payola. It wasn't, but some indie purists got indignant. I'm not picking on Beggars Banquet records. The source of the claim is un-cited, and if it happened I was unaware at the time. The Editor-in-Chief of CMJ, Kevin Boyce, wrote a nice apology letter with a full mea culpa, which was published in the March 10th 2003 issue of CMJ. [SOURCE] 

“The decision to replace unverified albums with Certain Damage was a foolish one... of course, in retrospect, we should've picked a more benign placeholder. At the time it was our genuine belief that this was a temporary fix in advance of curing a larger internal technical issue.”

In the before time, when there were fewer chart reporters CMJ staff would sometimes send the top 30 back to you if it contained errors. I know this because it happened to me when I was the college kid who couldn't count to 30. But when there are 1,500 charts to collate —that is not a scalable process. The later online submission form put an end to the problem. No one, no matter how hungover, could submit partial reports anymore and no chart with unverifiable records would be accepted anymore.

This new verification process led to a second lesser scandal where real but obscure albums caused Top 30 charts to be rejected. That became the subject of the March 24th 2003 issue. [SOURCE] It was a classic over-correction error.  (Earshot and CMJ  describe a letter from CEO Bobby Haber addressing the issue, but the URL was not archived [LINK] and it appears to be lost)

But you may be wondering... how did a few dozen manual corrections skew the chart in the first place?  Oh that's a thing even fewer people know. and since CMJ is no more, I think it is safe to share the thing that almost nobody knows... It starts with station weights. 

Not all airplay is created equally. It is somewhat intuitive that airplay on a 10 watt station will generally be heard by fewer people than a 50,000 watt station. CMJ used station weights such that "big" stations charts were weighted more than "little" ones. It wasn't super scientific, but the basis was logical. Bigger markets, taller towers, more watts all added up to bigger station weights. NACC actually used a near identical set of weights. [SOURCE]  A caveat before the next paragraph... this is from memory, decades later. It might not be perfect.

Anyway, every record promoter and record label had a list and the biggest weight was a 5 and the smallest was a 1. Not a huge scale but it all sounds pretty sensible so far right? That's the basis of the old CORE chart, just the airplay from the big stations: 4s and 5s. WRAS is the first 5 that comes to mind. I think KEXP, KCRW and WFMU were as well. (I wish I'd kept some of those reports.) All carrier current stations were 1s. Most stations didn't know about the weights at all, but the industry folks did, it was on all the tracking reports because that's critically important to know if you are trying to get a record to chart.

Here are 5 things you probably disn't know about the CMJ charts: 

1. Station weights were not static. Industry folks could lobby for revisions to station weights. For example: I personally got CMJ to downgrade WYBC in New Haven, CT down from a 4 to a 1. They had been miss-assigned a weight based on the 3,000 watt 94.3 FM stick. But the actual reporting station was 1340 WYBC-AM with only 1,000 watts on a tower which was actually in West Haven. (Today they don't even have the AM stick, they're WYBCx. [LINK]  Sorry/Not sorry.)
 
2. The station weights were not the actual station weights. The weights the promoters and label folks saw were actually mapped to another set of numbers. This was top secret. This was known inside CMJ to some, but considered the secret sauce of the chart so to speak. Most people did not know even inside the industry. What I remember is that a 1 was a 0.3, a 2 was a 1, a 3 was a 3 and a 4 was a 5. I think a 5 was actually a 7 or 8. My memory is a bit hazy; its been a few decades. Anyway with that kind of multiplier, you can see how even the #30 spot on a few "big" station top 30 charts might accidentally push a Certain Damage compilation onto the top 200 chart. Before you cry foul, station weighting and even more complex "momentum" formulas are standard practice in commercial charts and even in retail music sales numbers. Its always a secret formula. Don't hate the player, hate the data model.

3. There were fake records! As a prank, some promoters would ask stations to chart records which did not actually exist. Other promoters indignantly would report these ad hoc to the beleaguered editor of CMJ and they would be manually removed. This was most prevalent for the final chat of each year in December but it could happen any week of the year. 

4. There were lots of fake Top 30 lists. Some promoters were extra dirty and reported fake top 30 charts impersonating real stations. This was surprisingly easy to do. There was no validation process with emailed charts. Every station with a subscription could report. So all you had to do was impersonate a college station who wasn't reporting... and there were many to choose from. Some had seasonal gaps, others just never reported for whatever reason.

5. Charting was mostly fake anyway. Most Top 30 reports in reality were popularity polls driven by free T-shirts, tickets and tchotchkes. It was generally not reflective of actual airplay. This was before monitoring services like RAM and Mediaguide offered data-driven alternatives and the industry largely embraced that when it came. Later, self-reported charts like Spinitron pushed college radio charts into unexpectedly egalitarian territory. I don't even know what they do now for charts. Do kids listen to the radio anymore? 

Both Billboard and The Awl declared CMJ dead in 2016. It wasn't the first or the last time. [SOURCE] Vice Magazine declared CMJ dead in 2017. [SOURCE] The most recent reboot was an unexpected revival of the CMJ marathon in 2021. Their social media stopped updating after the event and the CMJ website is remains frozen in time. Ultimately the mergers, acquisitions and lawsuits probably did much more damage to the brand than any of the chart problems. More here, here and here.