RIP 100 Warm Tunas: Hottest 100 Prediction Site Shuts Down

It’s the end of an era for triple j tragics – cult-favourite Hottest 100 prediction site 100 Warm Tunas has officially called last drinks.

The site’s creator, Sydney software engineer Nick Whyte, has announced he won’t be running the project this year, wrapping up a side-hustle that’s been quietly ruining group chats and office sweeps for almost ten years. If you’ve ever smugly told your mates “nah trust me, Warm Tunas has it at #1”, this one hurts.

“Thanks to everyone who submitted votes, shared the site, and checked the predictions each year”

View this post on Instagram

Launched back in 2016 as a passion project, 100 Warm Tunas used a home-brewed algorithm to trawl social media for screenshots of people’s Hottest 100 votes – mainly via the hashtag #hottest100 – then crunched the numbers into a live leaderboard.

At its peak, the fishy little website was freakishly accurate, predicting around eight of the Top 10 songs each year and nailing the #1 track roughly 67% of the time.

Speaking to the ABC back in 2022, Whyte said he’d knocked back offers to sell the project, refused advertising, and even turned down a sponsorship deal from a gambling company. The whole thing ran at a loss – but he kept it alive because it was a creative outlet.

“It’s not about predicting number one for me,” he said at the time. “It’s about seeing how far I can take the problem-solving process.”

So why kill it now?

In a brutally honest farewell post on the site, Whyte explained that the project has become almost impossible to run thanks to the slow death of public social media data:

Instagram now hides chronological hashtag searches
Story hashtags are gone
Bot detection nukes scraping tools
Twitter/X has paywalled its API
People share less publicly and more via private stories and locked accounts

For the past few years, Warm Tunas was surviving mostly on people manually submitting their votes – which skewed the data towards hardcore triple j die-hards and broke the accuracy.

Despite that, Whyte used the project to level up hard: migrating the site to Next.js, building a new FastAPI backend, training an ML model to correct for bias, and fighting off spammers trying to tank the results. Not bad for a “side project”.

In his goodbye, he wrote: “It’ll be a shame to wrap it up… but I think the project’s prime time is now over. I’m excited to see what new projects others spin up in this space.”

And in a shorter social post, he kept it simple, writing: “After nearly a decade of predicting the Hottest 100, we’re winding things down. Thanks to everyone who submitted votes, shared the site, and checked the predictions each year; you made this ride possible.”

So this year, for the first time in a long time, we’re flying blind. No Warm Tunas safety net. No smug spreadsheets.

Pour one out for the fish that knew too much