Data Analyst Spoils the World's Biggest Song Vote

littlekorea writes "A data analyst has successfully predicted the top ten songs of the world's biggest song contest — the Triple J Hottest 100 — by extrapolating voting intentions fans had posted on Twitter and Facebook. Nick Drewe's Warmest 100 list closely mimicked the Hottest 100 results, predicting the top three in correct order and predicting 92 of the most popular 100 songs."

Comments

How is this spoiling?

[TWX]

That's like saying that Nate Silver or Andrew Tanenbaum spoiled the presidential election... Just because they looked at the signs and figured out how others were going to vote doesn't mean that they spoiled anything.

Let's talk about this again if statisticians end up actually influencing the vote through their data analysis. Otherwise, let it go.

Re:How is this spoiling?

[Anonymous Coward]

That's like saying that Nate Silver or Andrew Tanenbaum spoiled the presidential election... Just because they looked at the signs and figured out how others were going to vote doesn't mean that they spoiled anything.

Let's talk about this again if statisticians end up actually influencing the vote through their data analysis. Otherwise, let it go.

Have you considered that maybe Nate Silver made his predictions and then God rearranged the Universe to make it so? Thus, Nate Silver spoiled the election (or is God).

Re:How is this spoiling?

[TWX]

Have you considered that maybe Nate Silver made his predictions and then God rearranged the Universe to make it so? Thus, Nate Silver spoiled the election (or is God).

If his prediction was important enough to warrant rearranging the universe, then perhaps the Republican fielded a worse candidate than we could have possibly feared...

Re:How is this spoiling?

[Anonymous Coward]

How do you know He didn't? Hitler never won a majority vote until he outlawed the other parties. It was alliances with the other parties that allowed him to become Chancellor. The election prior to passing the Enabling Act his party only won 44% of the vote. The Enabling Act required a 2/3rds vote (it passed with 83%). The cowardice of the other parties to stand up to Hitler is a major reason he was able to get so much absolute power so quickly.

Re:How is this spoiling?

[nabsltd]

Have you considered that maybe Nate Silver made his predictions and then God rearranged the Universe to make it so? Thus, Nate Silver spoiled the election (or is God).

When you do things right, people won't be sure you've done anything at all.

Re:How is this spoiling?

[mrsquid0]

Perhaps things would have been much worse without Hitler. Stephen Fry wrote an amusing book based on this idea.

Re:How is this spoiling?

[dgatwood]

Yeah, sure. Next thing you're gonna tell me is that Darth Vader is Luke Skywalker's father.

Re:How is this spoiling?

[Anonymous Coward]

Nietzsche didn't influence national socialism. His works were very selectively edited by his sister who was a Nazi to try to give a philosophical background to and synch with Naziism. Nietzsche was very critical of antisemitism and nationalism in his writings. If Nietzsche would have been alive when the Nazis came to power, he would have been one of the first ones targeted.

Re:How is this spoiling?

[Opportunist]

Actually, statistical predictions CAN influence voting results considerably. Independent of the system used, for different reasons.

Especially first-past-the-post systems are prone to influence from predictions. If the statistics tell that candidate A is leading with 90%, a lot of people might consider not going to the polls, considering their vote is moot either way, whether they like him or not. Not too long ago, election predictions foretold a landslide victory for a certain party over here which nearly turned into a disaster for them. Nice weather convinced quite a few of their supporters that their vote isn't needed and that heading for the beach instead of the voting booth is a good idea, while seemingly that "shocking" news made people go to the polls that wanted to "limit the damage".

Predictions have an especially big influence on the success of smaller parties in systems where a certain percentage of voters is required to get a seat in the parliament. A prediction that tells a party will make it into the parliament can easily convince people who actually want this group to have a voice to vote for them who would have voted for their "second best" choice instead that makes it for sure (and hence makes their vote count), while a prediction that they will probably not make it can prove disastrous, as has been shown lately with the German FDP that was actually eliminated from a few state parliaments in recent elections after predictions that they might not make it.

And I don't even want to talk about those strange individuals who want to feel better by picking the "winning" party instead of voting for someone that actually represents their goals.

Re:How is this spoiling?

[AK Marc]

I've not read that book, but I had postulated the same thing many times. But Hitler is a cog, not a driving force. The hardship caused by WWI caused WWII. I'd be interested in seeing what would have happened if WWI never happened. That was the senseless one. WWII was caused by WWI. Much like a bit of the trouble in the middle east was caused by the ending of WWII. So what would have happened if WWI didn't happen, or the ending was different? It would be interesting to see, though I'd never write a book about the alternative, as that seems plenty ambitious/presumptuous.

Re:How is this spoiling?

[iserlohn]

WWI happened mainly because of colonialism an it doesn't stop there; we can recurse into this until the neolithic.

Re:How is this spoiling?

[AK Marc]

I thought it was old treaties and perhaps byproducts of colonialism, but it wasn't caused by colonialism as a primary cause. The colonial world war was Vietnam. US vs USSR in SE Asia, based on French colonialism.

But the only way colonialism was a cause of WWI was treaties and alliances made when splitting up the world that didn't work for more domestic conflicts. The way you can be sure that there wasn't some underlying greater cause is that the US took longer to enter (like WWII) and the side wasn't even certain at first (unlike WWII). There was a large push to enter the war on the side of the Germans. Especially in PA and other loci of German immigration.

Re:How is this spoiling?

[Smauler]

Hate to break it to you, but I don't even know the last time any party in the UK won 44% of the vote. This is how most of the world works - only the US has quite such an exclusive 2 party system.

Just found some stats here. Last time a party got more than 44% was 1970, then only just.

Note that this page also shows how flawed first past the post is. In the '83 election, the percentages of the top 3 party's vote were 42, 27, 25. The respective percentage shares of seats in the house of commons were 61, 32, 3. This can also be seen to a lesser extent in recent elections.

k...

[Anonymous Coward]

did you even look at the page?

He goes very much out of his way not to spoil it, but giving you half a dozen very large warnings.

Re:This just in

[Anonymous Coward]

Sorry Alphatel, but we've never heard of you either. However we do deeply care about your opinion regarding our radio station. Please tell us more.

Regards,

Australia.

Re:subject

[jonwil]

I am an Aussie and love good Aussie music and I haven't heard of any of the songs in the top 20.

In saying that, I should point out to the foreigners here that Triple J isn't exactly mainstream (Triple J started out as a station to play all the artists that commercial radio was ignoring and still does so today in some cases)

Social media

[jamesh]

Whilst listening to the countdown yesterday I was pondering the idea that the warmest 100 is flawed because it only takes into account the votes from people who think the rest of the world cares about their every thought. Turns out I was wrong... I guess I'm a bit older than the Triple J target demographic these days - I wasn't 20 years ago though dammit!

Next year i'm posting my votes on facebook, just to throw it out a little bit :)

The votes are in!

[PPH]

And the winner is ....

Re:subject

[Trepidity]

I've never even heard of any of the top 25, and none of the top 66 are in my Google Music playlists.

From this comment, can we infer that #67, a track entitled "On Top" by musical artist "Flume" (featuring guest artist "T-Shirt") is on your playlist?

Re:Two things.

[deek]

1. Your cynicism is getting the better of you. This is a valid news for nerds story, about how data analysis successfully predicted the majority of a list meant to be kept secret. The story would have been interesting, had it come from any part of the world.

2. The Triple J station is known for playing many alternative and unknown artists. It's a government funded station, but editorially independent from government by law. Your list of known songs/singers/artists are likely influenced by your regional tastes, and also commercial interests of the stations you listen to. These would not match up with an alternative music station.

2a. Have a listen to Thrift Shop, the song that made #1. I'm not generally one for hip-hop, but I liked it. Great lyrics, has some good things to say about fashion labels and stylistic expectations. Interestingly enough, the song is independent of commercial labels, and has had great success in US/Canada/Australia/New Zealand.

for those who didn't RTFA or listen to Triple J

[ferret4]

It's a song contest where the public (anyone in the world) can vote for any song (released anywhere in the world) so long as that song was first released in the 12 months prior to start of voting. The station does not nominate songs or pick winners, they simply count votes. The metric is that the Triple J Hottest 100 competition consistently receives the largest number of people voting for a competition of this kind anywhere in the world.

Re:Two things.

[caffeine_high]

I've been listening to triplej for about 20 years and as usual there are plenty of songs I do not recognize in the list. That is part of the deal with triplej, no comfort zone of classics and far more new music than you can follow.

There will be a few teeny pop songs in the list but if you listen to a few, you'll see that the range is far more diverse and none of the usual pop artists.

Over the past few years triplej has been doing a lot of work to foster local music, their site http://www.triplejunearthed.com/ is great and you can legally download plenty of good music.

I guess one day my music tastes will crystallize and I'll join the host of others who think that music was best in year X (usually when they were a teenager). I'll fight that as long as I can.

Comparison chart with commentary

[diodegod]

For a slightly more detailed comparison of Warmest 100 vs. Hottest 100, here's a chart a mate did which includes some commentary:

http://public.tableausoftware.com/views/WarmesttoHottest1002012/Dashboard

World's biggest song contest ?

[Peachy]

Surely can't beat the Eurovision in terms of biggest song contest ?

I am the author of the ITnews article.

[justin.warren]

So yeah. Hi.

I wrote some more detailed commentary on my blog, if you want more details than would fit in a short news article.

http://www.eigenmagic.com/2013/01/24/will-the-warmest-100-prove-a-scorcher/ and http://www.eigenmagic.com/2013/01/26/warmest-100-updates/

It's nearly midnight here, but if I get a chance in the morning, I'll post up some more info including the code I used to replicate the Warmest 100's methods, and some scatterplots showing how the accuracy of the predictions got better the closer things got to number one, similar to the tableau link from diodegod.

Statistics is the new black. :)