What does a 20 percent chance of rain really mean?

"There's a 20 percent chance of rain today."

I can't tell you how many times I've uttered these words throughout my career. I also couldn't tell you how many times I've been asked what a 20 percent chance of rain means. Does it mean 20 percent of the area will see rain? Does it mean it will rain 20 percent of the time? What does it really mean? Well, put on your math thinking cap for a minute because the correct answer is "none of the above."

A 20 percent chance of rain literally means that if weather conditions in your area were identical for 10 days in a row, statistically it would rain on 2 of those 10 days, or 20 percent of the time. In other words, rain today is very unlikely, but wouldn't be unheard of. If that answer is good enough for you, you can stop reading now, but if you're the analytical type like I am, read on for the math behind the percentages. However, I have to warn you--your head may hurt after reading the next 2 paragraphs. grin

To understand this, think about tossing a coin (which I often get accused of doing for my weather forecasts!). We all know there is a 50 percent chance the coin will land on heads and a 50 percent chance it will land on tails. If you toss the coin 100 times, the results will be split right down the middle, or at least very close. If I "forecast" a 50 percent chance the coin will land on heads and it lands on tails, does that mean my "forecast" was wrong? Was there really a 100 percent chance? Of course not. Taking it a step further, does it mean there is a zero percent chance the coin will land on heads next time? Of course not. The chance is still 50 percent because the coin still has 2 sides and must land on one of those sides.

Now consider what would happen if one side of the coin were heavier than the other? That changes everything.

Unfortunately, for me as a meteorologist, there are many, many factors that weight the proverbial weather coin in one direction or the other. Among the things that can affect how the "rain/no rain coin" lands are heat, humidity, winds, approaching cold front, low pressure systems, high pressure systems--the list could go on and on and on. These things all add up or cancel each other out to determine which side of the "coin" is heavier. For example, a cold front would tend to weight the coin toward rain, but if humidity is expected to be very low, the coin would lean back toward "no rain." As a meteorologist, I look at all of these dozens of factors and decide which side the weather "coin" is leaning toward and to what extent. If all factors completely cancel each other out, we have an unweighted coin--therefore an equal chance of rain vs. no rain. If all factors point toward rain, we have a 100 percent of rain. Hopefully, you get the point.

My job as a meteorologist boils down to analyzing this proverbial coin to as accurately as possible predict which side the weather "coin" is most likely to land on today. Like a real coin toss, it's a good science, but not an exact science.

Posted by on 05/12 at 07:53 AM

If you ever quit “weathering”..you’d make a decent teacher..

Again..I have learnt something that I had no idea what it meant..yet heard it thousands of times..

I’m starting to wonder just how many other hidden details there are in “weather” that are this interesting..

I’ll have to hand it to you..Melissa and Frank..your blogs are always packed with information that is educational..yet fun to read..

It’s amazing how many little facts you miss along the way in your schooling and then come here and pick up a better understanding of what is being said in a forecast..

Great job Chris!

Posted by  on  05/12  at  06:51 PM

Too bad Chris talks with rocks in his mouth, Frank is a boring statue and Melissa...well, is simply not good on TV....

Posted by  on  05/12  at  11:55 PM

I have always been told to explain it this way… A 20% chance of rain means that 20% of the area will see some type of rain.  So if you forecast 100% all locations will see rainfall.  And a 10% chance verifies if it rains in 10% of the area.  So if I forecast 30% and about a third of us see rain, then that’s a pretty good forecast!

Posted by  on  05/14  at  10:24 AM

Is Melissa a met?  Her explanation is completely wrong.  Any met who took statistics in meteorology could answer this question.  She attempts to take a complicated answer and make it simple, but it is far from accurate.  I suggest you brush up on exactly how the NWS uses and verifies percentages.  You’ll find there is no simply answer, but yours is wrong.

Posted by  on  05/18  at  07:46 PM

Hey Chip,

I am indeed a meteorologist.  I understand that a 20% POP that the weather service uses is different than the 20% that I use when forecasting for television.  NWS verification is based on a point forecast, while mine is for a region.  It just depends on methodology.  My forecast is based on the fraction of the coverage area that I expect to see rain, rather than the percentage of rain at one specific point in space.

Since all forecasts are subjective, I wouldn’t say that mine is incorrect or that a point forecast is the only correct verification method.

Posted by  on  05/18  at  08:38 PM

With the number of critics on the rise..

I’m glad I chose to become a beach bum instead of a meteorologist…

The benefits arent as good..but it is near impossible for someone to find fault in the way you lay around the beach..doing next to nothing..smile

Posted by  on  05/19  at  04:55 AM

Page 1 of 1 pages

Log In | Register as a new member