Surfers say the app promised chest-high and delivered ankle-slappers, again

PACIFIC BEACH, San Diego – More than sixty surfers converged on a stretch of sand Tuesday morning after a forecasting app predicted chest-to-shoulder-high waves with a “good” rating, only to find what several described, independently and with striking consistency, as “basically a lake with opinions.”

The forecast, generated by Surfline’s predictive model, had shown a green rating and a two-foot-plus swell arriving overnight. By 6 a.m., the ocean had produced what local surfer Tomas Reyes measured, using what he called “the eyeball method, which has never let me down,” at roughly eight inches.

A Pattern, Not an Incident

This is not, surfers say, an isolated case. A group calling itself the San Diego Forecast Accountability Collective, formed initially as a joke group chat and now boasting 340 members, has begun keeping an informal log comparing app predictions to observed conditions, and claims a discrepancy rate they describe as “statistically insulting.”

“I’m not a data scientist,” said collective founder Priya Nandakumar, who is, in fact, a data scientist. “But if you told me a model was wrong this consistently in any other context, I’d assume it wasn’t measuring what it claimed to measure. In this context, I just assume it’s messing with us specifically.”

Surfline has not responded to the collective’s spreadsheet directly, though the company’s public buoy and forecasting data remains among the most widely used in the sport, and its models account for factors, including swell period, wind direction, and tide, that are notoriously difficult to translate into a single color-coded rating.

The Psychology of the Green Rating

Local sports psychologist and occasional surfer Dr. Melinda Auer, who was not involved in the collective’s research but was asked to comment after overhearing the argument at a coffee shop, said the anger makes sense even if the app is technically doing its job. “A forecast is a probability, not a promise,” Auer said. “But nobody drives forty minutes and wakes up at five a.m. to be told a probability. They want a promise. When the promise breaks, that reads as betrayal, even if, statistically, it’s just Tuesday.”

Reyes, for his part, said he understands the science but remains unmoved. “I don’t need a lecture on swell period,” he said. “I need the app to stop showing me a green dot when the ocean looks like a Jacuzzi that’s been unplugged.”

An App Divided

Not everyone in the water Tuesday shared the frustration. Beginner surfer Casey Ohm, on only her third session, said the smaller conditions were “honestly perfect” for her skill level and expressed mild confusion at the volume of complaints around her. “Everyone kept saying it was flat,” Ohm said. “I caught four waves. I thought that was the whole point.”

The disconnect, according to The Inertia’s past coverage of forecasting culture, is a recurring feature of surf communities broadly: what reads as disastrously small to an experienced local can be an ideal session for someone newer to the sport, meaning any single rating is, by design, going to disappoint roughly half the people checking it on a given morning.

What the Model Actually Does

Forecast meteorologists who work with swell models generally caution against reading too much into a single color rating, noting that the underlying data typically includes a range, not a guarantee. A model showing a two-foot swell at four-second intervals with an unfavorable wind direction can look identical on a simplified chart to a much cleaner two-foot swell at twelve-second intervals, even though the two produce dramatically different waves at the sand.

Nandakumar acknowledged this nuance exists but said it undercuts her larger complaint rather than resolving it. “If the underlying data is that inconsistent, then maybe the app shouldn’t be handing out a single confident rating at all,” she said. “Show me the range. Show me the uncertainty. Don’t just give me a green dot and let me find out the hard way that green apparently means ‘maybe.’”

Reyes agreed, adding that he has started manually cross-referencing buoy data himself before paddling out, a habit he described as “exhausting” but necessary. “I didn’t get into surfing to become a part-time meteorologist,” he said. “But here we are.”

What Happens Next

The collective says it plans to present its findings, still unofficial and gathered largely over group chat, at an informal “forecast summit” next month, essentially a beach cleanup with a PowerPoint. Nandakumar said the goal isn’t really to change the algorithm. “We just want it on the record,” she said. “We want someone, somewhere, to acknowledge that the green dot lied to us. That’s it. That’s the whole ask.” She added that the group has already designed a T-shirt reading “TRUST THE BUOY, NOT THE DOT,” which she says has generated more interest within the group chat than the underlying spreadsheet ever did, a fact she finds simultaneously funny and slightly discouraging about the state of collective activism in 2026. Reyes said he plans to keep checking the app anyway, mostly out of habit, though he has started treating the color rating as “a suggestion, not a summons,” a mental adjustment he says has measurably improved his mood on flat mornings.

Full coverage of the collective’s ongoing spreadsheet, and San Diego’s broader relationship with forecasting technology, continues at bohiney.com.

Further reading from the humor desk: The Onion.

SOURCE: https://bohiney.com