What a 40% chance of rain actually means

Ask ten people what a 40% chance of rain means and you will get at least three answers. It rains 40% of the day. It rains on 40% of the area. There is a 40% chance of getting wet at some point. Only the last one is close, and it is still not quite it.

Here is the actual definition. In a US National Weather Service forecast, a 40% chance of rain means that at any given point in the forecast area, there is a 40% probability that at least 0.01 inches of precipitation falls during the forecast period. That threshold is deliberately tiny. It is the smallest amount a standard gauge can reliably measure, so the number is answering a much smaller question than most people assume: will anything measurable land here at all.

Two numbers hiding inside one

Forecasters build that percentage from two separate judgements, and then multiply them.

The first is confidence. How sure is the forecaster that rain will develop somewhere in the area? The second is coverage. If it does develop, how much of the area does it cover?

An 80% confidence that showers will develop, covering about half the region, gives you 40%. So does near certainty about a shower covering only 40% of the region. Those two afternoons look nothing alike. One is a sky that might stay dry all day, the other is a sky that is definitely going to rain on somebody. The forecast reports them with the same number.

This is why the percentage feels unreliable when it is not. A 40% day that stays dry where you are standing is not a failed forecast. It is the 60% resolving, exactly as advertised.

What the number cannot tell you

The percentage is missing all three of the things a decision actually needs.

It does not tell you when. A 40% chance spread across twelve hours and a 40% chance concentrated in a twenty minute window at half past four are the same number and completely different days.

It does not tell you how long. Duration is not an input to the calculation.

It does not tell you how much. Measurable rain starts at a hundredth of an inch. A day that delivers exactly that and a day that delivers an inch both count as rain.

So the honest reading of "40% chance of rain" is something like: rain is plausible today, nobody can promise you either way, and this number is not going to help you decide whether to cancel anything.

Read the timing instead

If the question is whether to move a plan, the percentage is the wrong field to look at. The useful ones are the hourly view and the rain rate. When does the precipitation start, when does it stop, and how heavy is it while it lasts.

That is a genuinely different question, and it is answerable with much more precision over short horizons. Radar based nowcasting can put a start time on rain in the next hour or so with real confidence. Past a couple of hours it degrades, and by tomorrow you are back to probability, which is the correct tool for planning but a poor tool for deciding.

This is the gap Ora is built around. Rather than leading with a daily percentage, it shows when rain starts and when it clears, because that is the shape of the actual decision. Jacket or not. Leave now or in twenty minutes. Move the thing outdoors or do not.

The one time the percentage is the right number

None of this makes probability useless. It is the right number when you are choosing between days rather than reacting to one. A 20% Saturday and a 70% Sunday is a real signal, and it is exactly the sort of comparison probability is designed for.

The mistake is only ever using it for a decision it was never built to answer. At nine in the morning, "40% today" is fine. At three in the afternoon, with the sky going dark and a plan to keep or break, you want a time, not a percentage.