The folks at the Hype Machine, the popular music tracking site, think that all of the Twitter music charts out there are “lame,” so they decided to make their own Twitter Music Chart. It encourages people to Tweet out links to their favorite songs on the Hype Machine, where you can listen to the full audio stream. They came up with a formula which gives people with more followers on Twitter more points for every song they Tweet. The songs with the most points, move up the chart.
It seems straight-forward enough, but it is way too easy for people with a large number of Twitter followers to game. I just RickRolled the chart by Tweeting a link to Rick Astley’s “Never Gonna Give You Up” from the TechCrunch Twitter account. The TechCrunch account has 916,735 followers, which gives each Hype Machine Tweet 2,997 points. A single Tweet was enough to put the song at the top of the chart, above Michael Jackson’s and Telepopmusik’s “Remember The Time” (1,972 points). Okay, so maybe it wasn’t fair to use the TechCrunch account, but what else am I gonna do on a Saturday morning plane ride back to New York (gotta love Virgin America’s WiFi in the sky).
Even before I Tweeted the link with the TechCrunch account, I Tweeted it first through my personal @erickschonfeld account, which only has 7,224 followers, and was able to get the song to debut on the list at No. 24. Every time I Tweet out a song link, it counts for 266 points, noyt enough to get a song to the top spot with one Tweet, but enough to move “Superteen” by The Care Bears On Fire from the No. 12 spot to the No. 5 spot.
The Hype Machine’s formula is flawed. No single person should be able to affect the rankings so easily. To be fair, it just launched, and as more people start voting, the system should self-correct. But the bigger problem with ranking songs based on someone’s popularity on Twitter is that just because someone has a lot of followers doesn’t man they have good taste in music (TechCrunch and myself excluded, of course). Ranking music based on roughly on how many Twitter followers someone has is just as lame as any of the other methods the Hype Machinists are trying to replace. (I like WeAreHunted). If there was a way to figure out who are the music experts or influencers on Twitter and give their Tweets more weight, that would create a more interesting list.
Otherwise, the Hype Machine’s chart is just going to keep on getting RickRolled.
so i guess they should factor in how many songs you’ve tweeted and add to the equation something that makes it so the # of followers coefficient isn’t linear (i.e. the more followers you have, the less each one is worth).
We Are Hunted has a much much cooler Twitter Music Chart.
WAH rocks!
Too easy to RickRoll?
Except for the little problem of trying to get 916,735 followers on your Twitter account.
I use the twitter music tracking site http://www.boil...gpage.com/music .. It does what it says, however, I haven’t compared this with RickRoll. But I believe BoilingPage-Music tracks all of twitter, not just pertaining to one music sharing site.
“Virgina America”? Freudian slip, Erick?
I’ll say! That’s what four nights in a row with less than 4 hours sleep will do to ya.
Give credit where credit is due. I believe Joe Laz pointed this out days ago in the comments of Hype Machine’s blog post:
http://blog.hyp...#comment-326526
Who reads comments?
Let’s make an agreement — I’ll give you some money, you tweet the songs I want tweeted. Sound fair?
Exactly, this is going to be overrun by Twitter spammers.
great song!
There’s typos everywhere Erick, get out your spell checker!
Also, it’s not rickrolling if they knowingly visit the video. Rickrolling is tricking someone into watching the video, not just mentioning the video somewhere. Learn your memes, Erick!
Nice “formula” that’s either the result of someone never attending a university and taking a single decent mathematics course with proofs or the marketing guy… I really really hope its the marketing guy.
so kids where does your 1/3…0.5…10…10 come from? if you marked D out of your arse you’d be correct!
We need to keep Rick Astley number one people
Better yet, we could let Rick rest in peace. Maybe then the RickRoll would become funny again.
…on second thought, it becoming funny again is as likely to happen as Michael Arrington professing his love for the former heads of Last.fm.
Using your twitter account to tweet out links to crappy songs so you can push them up to the top of the charts is one way to lose followers.
Great! Ashton Kutcher is now the arbiter of online music charts. I can’t wait until he games the charts with “Popo Zao” without a trace of irony…
Geez, Astley never had this many fans when his music was actually new.
Forget about the hype machine. Take a look at tweet-tunes.com – The Top 20 Realtime Twitter Music Charts!
hey erick, can you tweet a song for me?
And just because one has many twitter followers is no indication of musical taste.
or any kind of taste for that matter
Shouldn’t a song’s popularity on Twitter be linked to how many times it gets retweeted?
It would seem the only way to correct this is by weighting the algorithm based on twitter’s median user metrics, imho.
http://en.wikip...l_curve_grading
Speaking of which, anyone know where to find updated averages of twitter’s user-base?
I had a conversation with @haynes_dave of Soundcloud about how it would be easy to game this twitter chart. He pointed out something I hadn’t realised:
This is a chart of how many potential readers have seen about the track. So in this case, more people potentially will have read about the rick astley song.
For The Hype Machine, this makes sense. It’s about measuring ‘hype’, or how much promotion a track has got. So it’s easy to game, but the chart is a measurement, not a means to itself.
Get me? It is a MEASUREMENT, so it is still valid for what it is.
i like the recent conversations about charts and integrity. however we need to discern what these charts are for. Billboard for instance is based on units sold. The Hype Machine’s Twitter Music Chart seems to be based on views/readership. while We Are Hunted looks to be based on positive blogs/reviews out there. in effect, all my above mentioned charts are available to game and the debate to measure real popularity may be a cross-referenced search of them all.