A Computer Science Grad May Have Figured Out How To Predict The Hottest 100
If you don’t want to know the results of this year’s Hottest 100, close your browser now.
A computer science grad from the University of New South Wales has come up with a way to predict the winners of triple j’s annual music countdown, and he’s confident it’s legit.
100 Warm Tunas is the brainchild of Nick Whyte. Collecting data from Instagram posts tagged with the hashtags #hottest100, #triplejhottest100, #triplejhottest1002016 and #hottest1002016, the website has tallied thousands of votes and allowed Whyte to generate an accurate prediction of what the final count will look like.
“I did this sort of thing last year, and shared the results with five or six friends,” Whyte told Junkee. “Last year I got the top three, and if you don’t count the order of the top ten I got eight out of ten.”
Whyte says he was inspired by the Warmest 100, a short-lived website that accurately predicted 92 out of 100 songs in the 2013 count by cracking the algorithms that let users share their ballots on Facebook, Twitter and other social media channels. The website drew the ire of triple j management, who ended up changing their social sharing system to make the outcome harder to predict.
Last year we also saw the Tepid 100 use Instagram data to guess the end result – although the order was somewhat askew.
So who does Warm Tuna say is going to win? At the time of publication, with data still rolling in, Amy Shark’s ‘Adore’ currently sits at number one, followed by Flume’s ‘Never Be Like You’.
“As it currently stands I’m pretty sure that Flume’s going to top it,” says Whyte. “‘Adore’ being up at number one, I don’t know about that. You can see that ‘Adore’ has got 25 more votes than Flume, but that’s really nothing. 25 votes in a two per cent sample size, that can swing heaps.”
“I wouldn’t be placing any bets” he adds with a laugh.
Feature image via Flume/Facebook