Method & sources
How our fishing calls work
Every “I’d go” on this site is computed from public marine data, then written up in plain English. This page shows exactly which feeds, how they are weighted, and where our judgement replaces the maths.
Where the data comes from
| What | Source | How often |
|---|---|---|
| Wind, temperature, cloud | Open-Meteo Forecast APIHourly forecast for the mark latitude and longitude, licensed CC BY 4.0. | Refetched per conditions cycle, cached a few minutes |
| Swell and wave height | Open-Meteo Marine APIHourly significant wave height for the same coordinates. | Refetched per conditions cycle, cached a few minutes |
| Tide times and heights | Marine Institute (ERDDAP) with Dublin Port timingIrish National Tide Gauge Network readings, with a modelled fallback curve when the gauge feed is unreachable. | Gauge readings every 5 minutes, cached 5 minutes |
| Conditions narrative | Anthropic (Claude)Writes the plain-English summary only. It never sets the verdict, the score, the lead mark or the window — those are computed before it is called. | Per conditions cycle; the page still works when it fails |
| Maps | Leaflet with OpenStreetMap tilesNo third-party mapping account, no tracking pixels on the map. | Static tiles |
The verdict card on the home page carries a “Conditions updated” stamp showing when that call was generated, in Irish time. If a feed fails we say the ranking is unavailable rather than showing you an older answer as if it were current.
How a mark is scored
Three factors, wind 40, tide 35, season 25 Each is scored 0 to 1 for your window, then combined. The result is shown as a number out of 100.
- 40%WindDirection against the mark facing
- 35%TideWhere the cycle sits in your window
- 25%SeasonWhether the target species is on
On this coast most marks face the same way, so scores cluster. When the top marks sit within a point or two of each other the number is not the deciding factor and we say so — the reason is what matters. Treat the score as a tiebreaker, not a verdict.
Wicklow and Dublin are ranked separately
The 20 Wicklow marks and the 17 Dublin Bay marks each carry their own decision, scored on their own model. We never rank a Dublin mark against a Wicklow one, because the two coasts do not share shelter, tide timing or access.
Wicklow marks are walked and verified in person. Dublin Bay marks are researched from documented sources and labelled as such, so you know which kind of confidence you are reading. See editorial standards for how corrections are handled.
Calculated versus curated
Computed from data
- The verdict tier (Excellent / Good / Mixed / Poor)
- Every mark score and its rank for the window
- The best window and the second window
- Tide state, phase and the high/low times inside your window
- Which species are in season this month
Written and reviewed by us
- Mark access, parking, terrain and hazard notes
- Bait, rig and method suggestions per mark
- Species identification, habits and local behaviour
- Safety warnings and who each mark suits
- Regulation values, checked against the Irish Statute Book and EUR-Lex
29 species profiles carry season, bait and method detail. Where a species has a legal bag limit, minimum size or closed season, that value is treated as regulated source data with a citation and a checked date — not as editorial copy.
What this cannot tell you
A model reads wind, tide and season. It does not know that the water coloured up overnight, that a netter worked the beach on the last ebb, or that the fish simply are not there. Forecasts also drift: a call made at breakfast for a midnight window is working from a forecast, not an observation.
Conditions on the shore change faster than any forecast. Check the sea yourself before you commit, and never fish a rock mark alone in swell.
Put it to work
See today's call
The plan names the mark, the window and what to bring.