50% Chance of Rain: The Most Useless Forecast

Yesterday afternoon as I was preparing to drive to a disc golf course for some Monday exercise, I was checking the weather every 5-10 minutes. I wanted to make sure that I wasn’t going to be caught in a thunderstorm after driving for 25 minutes.

However, the forecast kept reporting a 50% chance of rain. After seeing this a few times, I realized that this is the most useless forecast. If I’m checking the weather, I need the percentage to veer one way or the other–a coin flip doesn’t help me.

For example, if the forecast was for a 40% chance of rain, I could reasonably infer decent odds against rain. Or if the forecast was for a 60% chance of rain, it would appear that it’s probably going to rain.

But 50%? Right down the middle? I don’t know what to do with that.

Perhaps I’m just using the wrong weather app. I have several on my phone, but the standard iOS weather app has the cleanest interface. Which weather app do you recommend?

10 thoughts on “50% Chance of Rain: The Most Useless Forecast”

  1. Jamey,

    As it was described to me by a weather officer years ago in the Air Force, the percentage has to do with the amount of rain in a given region, not that God is rolling dice.
    A 50% chance of rain simply means that 50% of that area should expect rain.

    • The way the BBC does its forecasts, a forecast of 50% at 1600 (say) means that if you put a bucket in your garden at 1500, there’s a 50% chance that there will be water in it at 1600

      That’s a combination of “it’s definitely going to rain somewhere in the area, but where is uncertain” (100% chance of rain over 50% of the area) and “if it rains, it’s going to get everything wet, but it might not rain at all in the area” (50% chance of rain over 100% of the area).

      And it’s all statistical anyway, so if you hsve a 50% chance of rain, that means if you go out on ten days that had that forecast, you’d predict that you’d get rained on on five of those days. With a 60% forecast, you’d get rained on on a sixth day; with a 40% forecast, you’d still get rained on four out of ten days. And with only a ten day sample, you wouldn’t be surprised to be a day or two off the average anyway.

      The important question is not “is it more or less likely to rain today?” but “how often am I prepared to get rained on?” If you’re happy to get rained on once every seven times you go out, then you should go out whenever the chance of rain is under 15%; if you want to stay dry nine times out of ten, hold out for 10% or lower. And so on.

  2. Any percentage is useless unless the report is 0% or 100%. Whatever happenes there was always a chance that it would, or wouldn’t. To clarify that actually 50% of the district refered to will get wet is equally as useless. You are equally none the wiser.

    I used to belong to a games club where one member would always list his % chance of being present. So after 10 minutes I would say, ‘He’s not here lets start., But others would chime in ‘But he said there was a 73% chance he would he here.’ So we’d waste another 30 minutes waiting for him.

    So circumstances in life need binary options.

  3. 50% chance of rain seems quite a useful forecast if it is not a typical probability for your area. If you lived in a desert, knowing 50% chance of rain on a given day might significantly impact your actions for the day. Consider the more extreme example of 50% chance of a natural disaster destroying your house today–surely that cannot be a “most useless forecast” to most people…

    • Consider the more extreme example of 50% chance of a natural disaster destroying your house today–surely that cannot be a “most useless forecast” to most people…

      It is useless because given a choice no one is going to risk their life with only a 50% chance of survival. In those extremes even if there is forecast a 10% chance of the volcano going off, you sensibly assume it will be 100% fatal if it happens and get the hell out. The apportionment of a % is still useless.

      Mindful that we aren’t discussing life and death extremes though, most sensible people plan for a worst case scenario. In this case as this is whether or not you will get wet when you get out of the car no one is going to agonise unduly over whether an umbrella is a 25% good idea.


Leave a Reply

Discover more from jameystegmaier.com

Subscribe now to keep reading and get access to the full archive.

Continue reading