To a large extent, the way statistics are expressed depends on the original data. For nearly 200 years in UK, we have measured rainfall at 09UTC each day, probably originally because that was a convenient time to do so throughout the year. This means that standard references, such as the Monthly Weather Report and British Rainfall, can give you the number of days (09h-09h) on which measurable amounts of rain fell in particular places during a particular month or year. If you have a rain-gauge near you and this has not already been done, it should not be hard to do this for yourself. For most parts of the UK you may expect to find big differences from month to month, and between the same month in nearby years. To get anything like a stable statistic, you'll probably need to do the job for about 30 years of data.
A small number of stations are equipped with continuously recording rain-gauges, hyetographs. It might be worth asking the Met Office if they have any hourly returns of rainfall for stations near you, but don't be surprised if there is not one very close.
For some years, I used to walk regularly from home to the station and from the destination station on to work each morning and then reverse the journey each evening; say four journeys each day; each of ten minutes. This was in south-east England. Only rarely did I get wet. Having lived for some years in Somerset, I can say that it rains more frequently (and more heavily) there! But, even there, it will only be for a relatively small fraction of the whole year. I suspect the same is true for most parts of the UK, even the north-west (though the lakes are there for a reason).
So, mostly, if you think of overall statistics, the probability of getting wet when you take your dog for a walk is small. But the weather when you are about to go out is the best guide. The weather now and the weather in a few minutes time, or even in an hour or two on most days, are highly correlated.
This approach can lead to some interesting questions. Your statistics can tell you, say, on how many days in the year it rains. Suppose it is 122; say one day in three. One day, it has not rained for five days. Is the probability that it will rain the next day less or more than the usual 33%? What if it does not rain for 10, or even 20, days? In summer in the middle of continental Europe we might think the next day is more likely than usual to be dry. But in UK?
We now have rainfall radar coverage of the UK. We see it on TV. In theory, it should be possible to build up statistics digitally, for every pixel (and so for more or less anywhere) of the strength of weather-radar echo and so the implied rainfall for every minute of every day. But in practice, to get useful figures, you'll need to amass the data for something like 30 years. They may eventually tell you that, on average, you'd better walk your dog on the east side of the hill rather than the west, but the best guide will still be: "Is it raining now– or does it look like rain?"