The Met Office and the BBC are keen to introduce probabilistic weather forecasts, to stop us grumbling about them getting the weather wrong.
Instead of giving one maximum temperature for the day, a range of possible temperatures could be displayed on a chart with a percentage probability attached to them. And "scattered showers" could become a 30% likelihood of rain.
But there are worries that people might not understand them.
Head of Ensemble Forecasting Research at the Met Office, Ken Mylne says the complexity of weather systems mean "there will always be uncertainty in forecasting, and that uncertainty should be communicated."
He says telling people about the probability of weather events would increase their confidence in weather forecasts, and help them make better decisions.
"If there's a 10% chance of a storm, and you decide not to mention it, there are a lot of people out there who could lose a lot. For example, a builder who has scaffolding up, or a farmer with his stock outside," he adds.
But although the BBC and the Met Office have been talking about doing probabilistic forecasts for a long time, BBC Weather Centre manager, Andrew Lane says there is a big stumbling block: "The capacity to confuse is immense.
Would weather probabilities really tell the whole picture?
"The concept of 90% chance of rain, say, is not too complex but when the probability approaches 50% people interpret it as guesswork, and complain you're supposed to know what the weather is."
And, he points out, even if the chance of rain is 90% it still may not be very clear what that means.
Does it mean 90% of the area will get rain or that 90% of the time it will be raining over the whole area?
But Mr Mylne says the Met Office's research into people's understanding of probability forecasts has been encouraging.
Researchers split 140 undergraduate students at Exeter University into two groups and asked them questions about how likely they thought various temperature scenarios were.
The first group were given a temperature graph with no information about uncertainty. The second were given a temperature graph which also expressed a measure of uncertainty.
Mr Mylne says, "We got a very clear signal that the second group made better decisions."
The Met Office already include percentage probabilities in its severe weather warnings. But, at the moment, weather presenters mostly try to express uncertainty in the language they use.
For example, they might say "patchy light rain is possible" or that "there are likely to be showers for many places".
Mr Mylne says it was not always like that: "When I first trained as a meteorologist, you were told to make a best estimate of a story and then stick to it."
He says the big storm of 1987 helped change that approach, as forecasters began to accept their own fallibility, and they have been developing uncertainty forecasts ever since.
So why is weather forecasting so uncertain a science?
It comes down to chaos theory, which says that a small, unpredictable event in the atmosphere can have a significant impact on the weather.
The effects of industry on local weather can be hard to predict
For example, on a still day big plumes of hot air from power stations can form clouds in the air, which affect the amount of radiation reaching the surface of the earth, which in turn affects the temperature.
This can create extra moisture in the air which can form local showers, which in turn can then create larger clouds.
So how do forecasters arrive at a percentage probability?
Meteorologists gather data from land, air and sea and feed it into a super computer which uses mathematical equations to create a forecast.
They then make small changes to the original data and feed up to 50 different versions into the computer. They then have a set of scenarios of what the weather might be, what is called an "ensemble" forecast.
If 10 of the 50 forecasts predict a big storm, there is a 20% chance of that storm happening.
A number of companies already use probabilistic forecasts. For example, an energy company might want more information about the likelihood of a one degree difference in temperature, because that can have a significant impact on demand.
Andrew Lane of the BBC Weather Centre says: "It's still fashionable to disregard forecasts - to say they're never right."
He thinks daily probabilistic forecasts could help counter that: "It's about sharing science with the public.
"Lifting the bonnet to show how things work."
But he also thinks there would need to be a big educational campaign if the BBC switched to probabilistic forecasts across the board.
Any initial trials would only take place online, where it is felt people have more time and inclination to take in more information.
And Mr Lane believes more work still needs to be done to find a clear way of communicating probability, without confusing or overloading people with information: "We appear very deterministic and often the forecasts are not, or can't be, and we fail the audience by not communicating that.
"But if probability is not communicated in a clear way, we could make things worse."