About Sun Roulette

A spin gives you a destination forecasting good weather right now. The wheel only includes places that scored “promising” or better on our weather rating today; nothing rainy, off-season, or too hot ever lands on the pointer. It is free, has no ads and no booking links, and needs no account.

How it picks

Open-Meteo gives us a 14-day forecast for every destination we track, refreshed once a day. A scorer rates each one on sun hours, temperature fit (peaks at 28°C, falls off a cliff above 35°C), rain, wind, and consistency. Shade mode turns that around and looks for daytime highs between 12 and 22°C, steady and dry, wherever you happen to live. The first five days carry most of the weight; days 11–14 are the model guessing and we mark them visually so you can tell.

The wheel pulls from the top 16 destinations sitting in “promising” or better. Within that pool, super-promising places spin a little heavier than promising ones, but anything in the top tier can come up. The label and the position always match — you can't land on a rainy week even if a destination is famous.

One limit is deliberate: the sun wheel only spins places within about eight hours' flight of northern Europe, because a spontaneous week away shouldn't start with two days of travel. Phuket, the Maldives and the Caribbean still have their own pages and live forecasts in the catalogue.

What this can and can't do

We can't promise sun. Nobody can. Weather two weeks out changes its mind regularly, and even a statistically reliable month throws the occasional wet week. What we can do is point you at the destinations the forecast prefers right now, with honest commentary on the months we'd actively avoid each place.

Treat this like a well-informed friend, not an oracle. Read the seasonal notes. Check the forecast again the day before you fly. Do your own digging on the things we can't see — festivals, renovations, the ferry that doesn't run on Tuesdays. We've tried to make finding the right place as simple as possible. You still get the final call.

Money

There isn't any. No ads, no booking links, no sponsored picks. If the site ever drew a crowd, affiliate links to hotels and flights would be the obvious way to cover its costs. That is a theoretical possibility, not a plan, and I doubt I'll spend time on it. Either way the ranking would stay about the weather.

Who built this

One person, A. N. Edvardsen, from Norway — where the sun problem is, fair to admit, slightly more pressing than for most. I was actually just looking for a small project to learn Claude Code with — I didn't want to point it at the real work before I'd felt out how it behaves. This was one of many half-baked ideas in my notebook (the kind that get hatched on a plane, or on a beach, or at a sidewalk café with sea view). I'd never taken it further because I was 100% sure there was no money in it. But I'm the type of person who has to experience every failure firsthand — reading about them just isn't the same.

I gave myself three days to play. While I was building the scorer, the destination data, the weather API, the parameters and the ranking — the actual useful bit — the “roulette” framing dropped into my head and added a couple more hours wrapping it in something you'd actually want to interact with.

Whether it goes anywhere, I have no idea. The fun part was making it. If it helped you find a week of sun, that's already a reasonable return.

If something is wrong, or if a recommendation feels off, send a note to hello@sunroulette.com. The site improves through corrections.

Last updated October 2026.