How we score
MonthPick scores are estimates from modelled climate data: averages for 2004–2023, not forecasts. They tell you what the weather is usually like in a month, not what it will be on your dates.
What the score out of 100 means
For each area, month and holiday type we compare the usual weather with what suits that kind of holiday: daytime high, rainy days, humidity (dew point), cloud and wind. Each factor scores from 0 (bad) to 1 (ideal), and they are combined with weights that suit the holiday. One really bad factor pulls the whole score down, so a hot month with a lot of rain cannot score highly. The result is shown out of 100, on the same scale for every area: we never stretch one area's scores to fill the range.
- Beach: ideal daytime highs of 26°C–35°C, 3 or fewer rainy days, low humidity, little cloud, light wind.
- City break: ideal daytime highs of 18°C–26°C and 5 or fewer rainy days.
- Hiking: ideal daytime highs of 12°C–22°C and 6 or fewer rainy days.
Within each ideal range there is a sweet spot (for a beach, a daytime high of about 30°C and no rain). A factor scores full marks only at the sweet spot and a little less (0.9) at the edge of its ideal range, so a perfect month is rare and good months can be told apart from great ones. Beach scores are only given to areas with beaches, and hiking scores only to areas with good walking nearby.
- Excellent: 90 or more
- Good: 75–89
- Mixed: 50–74
- Poor: under 50
When two areas have the same score in a list, the one we have more confidence in comes first (high, then medium, then low), then the one with more sunshine, then (for areas that show rainy days) the one with fewer rainy days, then alphabetical order.
Best months on an area's page are every month that scores within 5 points of that area's best score for the holiday type, up to 5. If more qualify, we show the 5 highest (earlier in the year first when scores are equal) and say how many more are close behind. Areas with low confidence do not get best months.
A rainy day is a day with at least 1 mm of rain. Sunshine is the usual hours of direct sun a day. Sea is the usual sea-surface temperature just offshore.
How we count rainy days
For most areas we estimate the number of rainy days from ERA5-Land, a climate model, counting the days when the model gives at least 2 mm of rain. We use this as an estimate of gauge-style rainy days (days with 1 mm or more on a rain gauge). We chose 2 mm because, compared with rain gauges at 16 areas, that count matched gauge-style rainy days better than the satellite count or the model at other thresholds. It is an estimate, not a gauge count. Dubai, Hanoi, Marrakech and Sal still count rainy days from CHIRPS satellite rainfall (days with 1 mm or more).
We could check rainy days against rain gauge records at 16 of our 42 areas. For the other 26 areas we use the same method but have not been able to check it yet, so treat their rainy days as rougher.
Tropical areas show the month's total rain in millimetres instead of rainy days, because that is what their score uses. It comes from CHIRPS satellite rainfall (from ERA5-Land where CHIRPS has no record).
For 16 tropical areas the city break score is the same number as the beach score in every month, so we show it once, under beach, and leave those areas out of the city break lists.
One point per area
Each area is scored from one data point, not from a survey of the whole area. Temperature, rainy days, humidity and wind come from the ERA5-Land grid cell nearest that point; those cells are 0.1° wide, about 11 km from north to south and a little narrower from east to west away from the equator. Cloud and sunshine come from a coarser ERA5 cell (0.25°, about 28 km), sea temperature from the nearest open-sea ERA5 cell, and the monthly rain of tropical areas from CHIRPS satellite rainfall (0.05°, about 5 km). Conditions elsewhere in an area, for example the other side of an island, a hillside or another stretch of coast, can differ from the figures for that point.
Tropical areas
In tropical areas temperatures barely change through the year, so we score a month mainly on how much of the day stays dry. Hours of daytime rain count for half the score, the month's total rain for a fifth, how sticky the heat feels for a fifth, and sunshine for a tenth. A short afternoon shower costs a little; a monsoon month costs a lot.
We calibrated this tropical model on 16 tropical areas and tested it on 8 others, against published travel guides and a climate site. That shows how well it agrees with those sources, not measured visitor experience.
Tropical scores do not include sea conditions or cyclone risk.
What confidence means
- ● High: after our corrections, daytime highs agree closely with nearby weather-station records.
- ◐ Medium: either there is no nearby station record to check against, or the nearest one does not agree closely with our corrected daytime highs. Either way the figures rely on a general correction for that kind of area (coastal city, inland city, small island or tropical). Tropical areas are never rated above medium.
- ○ Low: the climate model's grid square is mostly sea, so the figures are rougher. We show these scores as an estimate. A tropical area is also marked low when its yearly rainfall differs by more than half between our two rain sources.
Where the figures come from
- Temperature, humidity and wind: ERA5-Land climate reanalysis (Copernicus Climate Change Service), adjusted for the area's altitude and with a general correction for the kind of area, because the model runs cool by day on islands and in the tropics.
- Rainy days: ERA5-Land (Copernicus Climate Change Service), as an estimate of gauge-style rainy days. Tropical areas show monthly rain from CHIRPS satellite rainfall (Climate Hazards Center, UC Santa Barbara). Dubai, Hanoi, Marrakech and Sal still count rainy days from CHIRPS.
- Cloud, sunshine and sea temperature: ERA5 (Copernicus Climate Change Service).
We modified this data: we averaged it by month, corrected it as described above and turned it into scores. The ERA5 and ERA5-Land data are licensed under CC BY 4.0. Neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains, and neither endorses MonthPick. CHIRPS data are in the public domain.
Generated using Copernicus Climate Change Service information; CHIRPS rainfall (modified).
Commercial links never affect scores. Some pages may carry affiliate links (marked “Affiliate link”); they never change a score, a ranking or which areas we list. About affiliate links.
What the scores do not cover
Monthly averages do not capture storm or hurricane risk, and a city's figures are a proxy for the hills nearby when it comes to hiking. Always check the forecast and local advice before you travel.