We’ve all been there: You’ve pounded a couple of high-ABV beers at happy hour when The Strokes start blaring from the jukebox. It’s the band’s old stuff—think Room on Fire-era Strokes—and the opinion forms in your head that, Man, Jules and the boys really peaked in 2003. At that moment you realize you have to share this thought on Twitter because the world deserves the truth.

You inevitably hang your head in shame looking at the tweet the next day. But it doesn’t have to be that way. A team of researchers from the University of Rochester has developed a computer algorithm to pinpoint exactly when you’re sharing an inebriated, often-incoherent thought on the social-media site. Turns out drunk tweeting is pretty predictable.

Gizmodo reports that, in 2014, the team analyzed over 11,000 geotagged tweets in New York City and Monroe County, NY. From there, the researchers filtered boozed-up tweets from sober ones using the following strategy:

  • Searching for terms like “beer,” “party,” and “drunk”
  • Using Amazon’s Mechanical Turk crowd-sourcing service to determine if the Tweet was sent at the same time the Tweeter was drinking
  • Developing an algorithm to verify that the tweet was sent while drunk

Gizmodo went on to qualify that the study had an altruistic motive in cracking the code on Twitter users’ drunken habits.

Mapping the boozy tweets can show where people are drinking—or, more specifically, drinking too much—and could help cities point out important trends when it comes to alcohol policy and public health. As MIT Technology Review points out, this kind of research could prevent some of the 75,000 alcohol-related deaths in the US every year.

At this point, it doesn’t appear that the algorithm is yet equipped to determine a drunk tweet without certain buzzwords, so your Strokes #HotTake might go undetected. However, if the University of Rochester team needs to expand its sample size in future studies, it can always look towards the Twitter feeds of Adele and Sam Smith.

You can read the full breakdown of the study over at M.I.T. Technology Review.

[via Gizmodo, M.I.T. Technology Review]