A star rating alone isn't enough because it tells you how good the surf is supposed to be without telling you why, or for whom, so you have no way to check it against your own level, your beach or the tide. Most surf forecast star ratings are a black box: a number goes in one end, a few stars come out the other, and the reasoning stays behind the curtain. Ours is printed in public. Every one of the 115+ spots on Peel Watch gets a 0–5 score per hour, per skill level, eight days out, and the score is multiplicative: 10 × swell quality × wind factor × tide factor × skill fit × daylight, halved and printed to one decimal. Because the terms multiply, one dealbreaker sinks the hour. A howling onshore wind, a dead-wrong tide or a swell too big for your level drags everything toward zero, however good the rest looks.
What do surf forecast star ratings actually hide?
Three stars on a Tuesday. Three stars for whom?
A star rating compresses swell size, period, direction, wind, tide and daylight into one symbol. That compression is the whole point, and also the whole problem. Once it's done, you can't get the parts back.
Here is what usually goes missing:
- Who the rating is for. The same morning can score a 4 for an intermediate and a 0 for a beginner. One star count can't hold both.
- Why it's low. Small swell and a blown-out sea look identical as two stars. They are very different mornings.
- What would change it. If the rating is low only because of the tide, waiting two hours fixes it. A black box won't tell you that.
- How wrong it might be. A star has no error bar.
The practical cost is familiar to anyone on this coast: a forty-minute drive to a beach the nortada (the northerly wind that builds along the west coast on many summer afternoons) has already turned to mush.
How are surf ratings calculated on our dashboard?
We'd rather you could check our arithmetic than take our word for it. So here it is.
The formula
score = 10 × swellQuality × windFactor × tideFactor × skillFit × daylight
The result is halved and printed to one decimal, from 0.0 to 5.0. Each factor sits between 0 and 1. A perfect factor leaves the score alone; a bad one cuts it.
The bands
- 0–1: Not so great
- 1–2: Marginal
- 2–3: Decent
- 3–4: Great
- 4–5: Fantastic
The skill levels
Skill fit is where a forecast stops treating a beginner and an expert as the same person. Each level has an ideal breaking-height band and a hard cap, in meters, as the site prints them. Above the cap, skill fit collapses and so does the score.
| Level | Ideal breaking height | Hard cap |
|---|---|---|
| L1 Novice | 0.3–0.9 m | 1.3 m |
| L2 | 0.5–1.2 m | 1.8 m |
| L3 | 0.8–1.8 m | 2.6 m |
| L4 | 1–2.6 m | 3.8 m |
| L5 Expert | 1.3 m+ | No cap |
Why multiply the factors instead of adding them?
Because the ocean doesn't average. A clean, well-sized swell does not compensate for a gale blowing straight onshore. You can't surf the average.
A hypothetical hour, purely to show the arithmetic:
- Swell quality 0.9, wind 0.9, tide 0.9, skill fit 1, daylight 1. That's 10 × 0.729 = 7.29, halved to 3.6. Great.
- Same hour, but the wind swings onshore and its factor drops to 0.1. Now it's 10 × 0.081 = 0.81, halved to 0.4. Not so great.
If we added the factors, that second hour would still look respectable, because four out of five things are fine. Multiplying makes the one bad thing visible. It's a harsher model, and that's deliberate. We'd rather under-promise a session than send you to sit in the parking lot with a flask watching whitecaps. Though, to be fair, we've done that voluntarily in February.
What goes into each spot's profile?
The formula is only as good as what feeds each factor. For every spot we store:
- Coastline facing, measured from OpenStreetMap rather than estimated.
- A hand-authored swell window with a shadow factor, for headlands and islands that block certain directions.
- A size factor for how the spot handles bigger or smaller swell.
- Wind tolerance and sheltered wind sectors. This is how the nortada is modeled: a north wind can be offshore at one beach and cross-shore at the next.
- Tide preference and tolerance, because some banks only work around one stage of the maré (tide).
- Per-level skill gates.
Neighboring spots share one offshore forecast. What separates them is orientation, seabed and tide, which is exactly what the profile captures.
When a spot is new, it carries a "provisional" badge on the dashboard, with safety caps on size and skill fit, and it's excluded from alerts until a person has checked the profile. That's the line between what the model computed and what someone has looked at with their own eyes.
Where does the raw ocean data come from?
The ingredients of a surf forecast in Portugal are public and granular. IPMA, the national weather and sea institute, publishes hour-by-hour marine data for individual beaches, including total sea height, swell height, wave period, peak period and wind in knots and Beaufort, on pages such as Informação para a praia: Costa da Caparica.
Underneath that sits a measured network. The Instituto Hidrográfico's multiparameter buoys record sea state, wind, water temperature and currents, and feed MONIZEE, Portugal's real-time monitoring and forecasting system (Jornal do Algarve). Coastal buoys sit in water under 100 m deep; oceanic buoys in zones over 1,000 m (Rede de boias ondógrafo dataset record).
Those buoys also record what storms actually do:
| Buoy | Maximum wave height | Event | Source |
|---|---|---|---|
| Nazaré | 17 m (about 56 ft) | Storm Hermínia, January 28, 2025 | Região Sul, Jan 2025 |
| Leixões | 15.8 m (about 52 ft) | October 17–23, 2023 | Marinha Portuguesa, Oct 2023 |
| Sines | 11.5 m (about 38 ft) | October 17–23, 2023 | Marinha Portuguesa, Oct 2023 |
| Faro | 9.9 m (about 32 ft) | October 17–23, 2023 | Marinha Portuguesa, Oct 2023 |
These are storm maximums, not surfable conditions for anyone on our L1–L4 scale. The point is that the measurements exist and are specific. Our own forecast runs on Open-Meteo model data, updated four times a day.
How accurate is a surf report, really?
Surf report accuracy is the question a star rating can't answer about itself. Ours tries to.
Roughly every 72 hours, we lock in a forecast and compare it with the day-of call. The results go on the accuracy panel on our features page:
- Average swell-height error: about ±0.1 m, growing to ±0.2 m at five days out.
- Average wind error: ±1–3 kn, depending on lead time.
What that means in practice: treat a 0.2 difference in the 0–5 score between two spots as noise. If one beach shows 3.1 and the next shows 2.9, the forecast is not telling you to pick the first. It's telling you they're about the same, and your own knowledge of the parking lot, the crowd and the walk down should decide.
The further out you look, the wider the margin. An 8-day hourly forecast is useful for planning the week, less so for choosing the hour on day seven.
A surf score is not a safety rating
This matters more than anything else in the post. A high score means the waves suit your level. It says nothing about rip currents, flags or lifeguards.
That's partly why we score swimming separately. The swim board ranks every beach 0–10 for calmness, a different scale from the 0–5 surf score, because a swimmer wants roughly the opposite of a surfer: less swell, less period, less wind. The site states plainly that rips, flags and lifeguards are not in that data.
The context is sobering. Of 121 drowning deaths in Portugal in 2024, 41.3% happened at sea and 97.5% occurred in places without lifeguard supervision (FEPONS report via Diário de Notícias). FEPONS noted only three of the 121 deaths were in supervised zones during the bathing season (Renascença, July 2025).
No forecast, ours included, can tell you a beach is safe. We describe hazards on each spot page. The all-clear is never ours to give.
Reading the working on a live spot page
Mainland Portugal has about 1,089 km (roughly 677 miles) of coastline by one GIS measure (Riscos.pt, coastal territory chapter). That's a lot of water to summarize in stars.
So here's our one suggestion. Next time you check a forecast, open a spot page on the dashboard instead of the headline number. Set your skill level. Look at the spot's best swell, best wind and best tide, then scroll the hours and see which factor is dragging the score down. If it's the tide, wait. If it's the wind, try a beach with a different facing. If it's skill fit, that's the forecast being honest with you, and it's worth listening to.