01One score, four words
Every spot gets a Glyph Score from 0 to 100 for each hour, plus a word: EPIC (80 and up), GO (60–79), FUN (40–59) or FLAT (under 40). The same score shows on the spot page, in lists, on the map and in alerts, so they always agree.
The score rates the surf itself: how much swell reaches the beach, how clean it is and whether the tide helps. It is the same for everyone. If you sign in and set your level, you also get a personal verdict on top (see below).
02What goes in
Only measured or modelled ocean and weather data from public sources, credited on every forecast page:
- Swell: wave height, period and direction, plus water temperature, from the Open-Meteo Marine API. The height is the model's significant wave height out at sea (Hs), not measured at the beach: shoaling, refraction and a wide shelf change it on the way in, and we don't adjust it per spot yet. That's why we show it as a range, from the typical wave to the bigger sets.
- Wind: speed, gusts and direction, plus air temperature, rain, cloud and UV, from the Open-Meteo weather models.
- Tide: NOAA station predictions where a station covers the spot, otherwise Open-Meteo's modelled sea level.
- Observed wind: where an airport weather station (METAR) is close by, its real measured wind is preferred over the model.
03How it's put together
Waves first. Size and period make the base, adjusted for the tide and for how sure the forecast is. Both adjustments grow with size and period: short-period chop doesn't get better on a rising tide. Then the wind: when we know which way the beach faces, wind is judged against it, from offshore (cleans the waves up) through cross-shore to onshore (messes them up). Where we don't know the beach's direction yet, only wind speed counts.
Then the swell direction: a swell has to be able to reach the break. Each spot has a swell window (its own from local knowledge, otherwise 70° either side of the way it faces), and swell from outside that window counts for less and less, fading out over 45°.
Last, the surface. Each hour is read as glassy, light texture, choppy or blown out, and that caps the score: a choppy hour can't be better than FUN (59), a blown-out one stays FLAT (39), however big the swell. A sea with a mean period under 6 seconds is wind chop, so it also stays at FUN at most, whatever the tide or wind.
04How sure it is, and how fresh
Each reading carries a confidence from four checks: how complete the data is (35%), how fresh it is (25%), how reliable the sources are (25%) and how much they agree (15%).
Every score shows its age. The forecast is refreshed when the weather models publish a new run (about twice a day), more often for the spots people are watching. Freshness is judged against that cycle: the latest run counts as fresh however many hours old it is, and the confidence only drops when a run is missed.
05The reason line and the best window
Next to the score is one line saying why, like "clean lines, light offshore wind, tide on the push". It is written by rules from the same numbers that made the score. No AI decides the score or writes the reason, so the same conditions always give the same answer.
The best window is the stretch of at least two hours of daylight with the highest score in the next day and a half, with its reasons.
06Your own verdict
Signed in with your level set, you also see whether the day suits you. One level above you caps your verdict at FUN; two or more above says SKIP. The public score never changes for this.
07What it can't tell you
A forecast is a model of the ocean, not the ocean. Sandbanks move, local winds do their own thing and crowds aren't in the data. That's why surfers' own reports and live cams, where a spot has them, sit alongside the forecast.
Use the score to decide whether to go and check. Always look at the water yourself and surf within your limits.