Oh and now I read this. Cool. I guess I was somewhat close.Mr. Pizza, I am surprised at you:
The National Weather Service, meanwhile, has the same minimum standards for precipitation as AccuWeather, but displays the probability of rain at any given point in the forecast area, rather than at the specific point where the gauge is located, according to spokesman Christopher Vaccaro. (So if theres a 50% chance that it will rain somewhere in the area, and, if it does rain, 80% of the area is likely to be covered, the probability of precipitation is 40%.)
When the weather forecasters give it an 80% or 90% chance of rain, it rains 100% of those times? They're apparently afraid of commitment.
If they give it an 80% chance, then logically, on average, 1 out of 5 times that they predict an 80% chance of rain, it shouldn't rain.
Based on what I've been told by meteorologists at work, it's because weather forecasting isn't an exact science. There's too many variables at work that can affect things. So they could see a mass moving on the radar but the slightest shift of wind could push it north or south. So when they say 80%, they mean there's a 1 in 5 chance the storm system could change course.
Unless that is, you're saying that I'm an unlucky bastard, and EVERY time there's an 80% chance of rain, it rains where I'm at. And, somewhere, maybe 20 miles away, maybe 50 miles away, there's some really lucky bastard who, when they say "80% chance of rain" it doesn't always rain.
Yippee! All you said is that "if there's an 80% chance it'll rain, they mean there's a 20% chance it won't rain." I bow to your mathematical prowess. Okay, so let's say there's a 1 in 5 chance the storm system could change course - yet out of the last 500 times there's been a "1 in 5 chance", the storm hasn't changed. The person determining that 1 in 5 chance seems to be getting that determination wrong.
I understand this: "So if there’s a 50% chance that it will rain somewhere in the area, and, if it does rain, 80% of the area is likely to be covered, the probability of precipitation is 40%," and the rest of the point you thought you were making. Let's say that's the forecast 2000 days in a row. Then, 50% of the time, it'll rain "somewhere in the area." And, when it does rain, "80% of the area is likely to be covered." If you were somewhere in the area, that works out to... drum roll... 40% of the time that they say "there's a 40% chance of rain", it rains in any particular area.
Unless that is, you're saying that I'm an unlucky bastard, and EVERY time there's an 80% chance of rain, it rains where I'm at. And, somewhere, maybe 20 miles away, maybe 50 miles away, there's some really lucky bastard who, when they say "80% chance of rain" it doesn't always rain.
Could this be the gamblers fallacy in disguise masked behind the weather forecast?
Just because it's 90% going to rain and it doesn't is independent of the next forecast.
Would it be illegal to count clouds?Could this be the gamblers fallacy in disguise masked behind the weather forecast?
Just because it's 90% going to rain and it doesn't is independent of the next forecast.
It has been a long time since I took the course, "physics of the weather" but when I took that class, what we were taught was that 80% chance of rain meant that over some area of land, 80% of the land will receive rain.
I believe it makes much more sense that way and if you look at whether maps and storms, you can see how that makes sense.
EDIT: That doesn't mean "weathermen" use it correctly. I was a little off as it includes both. But you can see why it makes more sense to think of it in terms of area.
http://www.srh.noaa.gov/ffc/?n=pop
Let's put it another way - there was an 80% chance of rain. 100% of the area got rain today. The last time there was an 80% chance, 100% of the area got rain. The time before that, the time before that, ... 100% of the area got rain.
Yippee! All you said is that "if there's an 80% chance it'll rain, they mean there's a 20% chance it won't rain." I bow to your mathematical prowess. Okay, so let's say there's a 1 in 5 chance the storm system could change course - yet out of the last 500 times there's been a "1 in 5 chance", the storm hasn't changed. The person determining that 1 in 5 chance seems to be getting that determination wrong.
I understand this: "So if theres a 50% chance that it will rain somewhere in the area, and, if it does rain, 80% of the area is likely to be covered, the probability of precipitation is 40%," and the rest of the point you thought you were making. Let's say that's the forecast 2000 days in a row. Then, 50% of the time, it'll rain "somewhere in the area." And, when it does rain, "80% of the area is likely to be covered." If you were somewhere in the area, that works out to... drum roll... 40% of the time that they say "there's a 40% chance of rain", it rains in any particular area.
Unless that is, you're saying that I'm an unlucky bastard, and EVERY time there's an 80% chance of rain, it rains where I'm at. And, somewhere, maybe 20 miles away, maybe 50 miles away, there's some really lucky bastard who, when they say "80% chance of rain" it doesn't always rain.