If you are in the United States, nothing beats getting the forecast straight from the horses mouth (NWS that is). Some companies are notorious for producing 30+ day forecasts, which can't have any meaningful levels of skill.
NOAA/NWS should just create their own mobile application. I use Wx[1], which parses NWS data directly and can be found on f-droid[2] and g---le play.
Granted, Wx doesn't follow Material Design for Android in the slightest, but I like it that way because it's very information-dense, snappy, and light (unlike r--ct "native" and other JS toolkits out there).
[0]: https://news.ycombinator.com/item?id=19775291
I agree that NWS [text] forecasts are usually the best, though it can be time consuming to digest them. Nate Silver's book [1] makes the interesting point that commercial forecasts almost always reduce the quality of the input data they're given from NOAA, but that part of this boils down to incentives: Nobody complains if you forecast a small chance of rain and it turns out to be sunny. The problem spot is in ruining someone's picnic. Hence, forecasts tend to bias heavily towards rain.
[0]: https://www.bloomberg.com/news/features/2018-06-14/trump-s-p...
Your blanket statement doesn't always hold. This is more accurate.
Have there any been any comparisons of the commercial services to NWS? That would be an interesting casual study.
Edit: There are some additional interesting comments downthread: https://news.ycombinator.com/item?id=22740480
I am. Spend some time in the desert southwest, particularly on indian reservations. You'll find that NWS temperatures are regularly off by 10-20°. But that's still better than Weather Underground, which can be off by as much as 50°.
One thing I will give the NWS credit for is the wind forecasts. Those things are spot-on at least 90% of the time.
I've lived in 15 cities in a dozen states, and what I've noticed is that if you're in a large city, or east of a large city, the forecasts are great.
But if you're west of a large city, or in a smaller city, it's hit-or-miss. This makes sense, as the better forecasters tend to end up in the larger markets.
Even smaller cities like Minneapolis, Buffalo, Sioux Falls, Albany, and hundreds more are located where they are because waterfalls and rapids made them the farthest extent of water travel before rapids or waterfalls or other hazards made the route too difficult.
After finding no substitutes, I started making my own weather app (for my own private use) this week to duplicate that functionality. But of course I was using the Dark Sky API which will now get shut down. I just can't win this one!
https://play.google.com/store/apps/details?id=com.samruston....
Sadly at some point they weren't able to pay for the dark sky API anymore and since then the quality of forecasts is a bit less convincing. I'm still rather shocked by the low reviews though, for me it works well and I love that you can add a dozen cities and see the forecast.
Supposedly they're still fixing bugs from the new weather provider before it's made public, but it hasn't been updated in ten months. I doubt it will be improved anymore, and I don't really use it much since the new provider is mediocre at best.
It's beautifully designed, has an excellent selection of widgets, and is open source.
You can even set your lock screen to be a live wallpaper of the current weather.
Now it's probably the worst app on my phone and I should just delete it. It takes forever to start, more than half the time it loads no data, and even if it is loaded the UI redesign has made the whole thing slow and useless.
Would love suggestions for another app on Android.