How uncertainty can help you make more informed decisions

A couple of weeks ago, on one sunny afternoon, my family and I decided to go for a leisurely walk. As we strolled along, we couldn’t help but notice the clouds gathering ominously above. Undeterred we trekked on, and a cursory glance at the weather forecast on our phones, we were comforted by the fact that the Met Office had predicted only a 30% chance of rain. But alas, nature had other plans, and we soon found ourselves thoroughly drenched.

As we huddled cold under a tree trying to get a brief respite from the wind and the rain, I couldn’t help but ponder. how the Met Office quantifies and communicates their level of confidence with their weather forecasts. Even if it wasn’t something my children wanted to hear at that point in time.

The Met Office is the UK’s national meteorological service, and as it is in the UK, one of the most crucial aspects of their forecasts is the probability of rain. It’s not just about saying “it will rain” or “it won’t rain” (a binary outcome); they go a step further by quantifying their level of certainty. In our case, that 30% chance of rain was, in essence, a 30% chance of getting wet.

But here’s the catch: 30% may sound low, but it’s not zero. It’s like playing a game of chance, and the classic example of flipping a coin. At 50%, you’d expect the coin to land heads or tails roughly half the time. So, a 30% chance of rain is not as far from a random coin toss as we might hope.

Now, let’s talk about that 50% chance of rain. It’s the middle ground between rain and no rain. Wet or dry. But does it really provide us with much insight? Not really.

The Met Office’s meteorologists understand that weather can be unpredictable, and sometimes, they’re genuinely uncertain about the outcome.

They might as well flip a coin.

So, what can we learn from the Met Office’s approach to weather forecasts?

It’s all about transparency and communicating uncertainty.

They don’t just tell you what to expect; they tell you how confident they are in their predictions. In a world filled with uncertainties, this is a valuable lesson.

In our daily lives, we often make decisions based on incomplete information, just like choosing to take your family out for a walk with a 30% chance of rain. But I think there’s much to learn from the Met Office’s approach and as data professionals we should communicate the level of confidence and uncertainty we have in our outcomes. Nothing is perfect.

By acknowledging and quantifying uncertainty in our decision-making, it can empower us to make even more informed choices.

Perhaps few things are as uncertain as the British weather. But that’s up to you to analyse, decide and communicate. Maybe just take an umbrella in case.