Will you need a coat? Should you bring an umbrella? Should you check if you’ve got some de-icer in the glove compartment? It’s January in Northern Ireland, so the answer should probably be ‘yes’ all round – regardless of what the weather forecast says.

Sometimes these forecasts will be wrong. A prediction of what will happen in the future can only ever be just that: a prediction. They always contain a degree of uncertainty. The point is that they can help us make relatively informed decisions in the face of otherwise total uncertainty. If you were a passenger on a plane, you would probably prefer to know that your aircraft’s assigned route took into account the expected weather conditions calculated by tried and tested models, rather than being told that the pilot just decided to wing it and see what it’s like along the way.

Prediction, despite its inherent uncertainty, is a way of trying to reasonably reduce uncertainty. As the UK prepares to leave the EU, the country is navigating an unfamiliar course. It makes sense to try and plan a route that avoids the nastiest pockets of turbulence. Some routes may be much bumpier than others.

That’s why the government tasked civil servants with calculating the potential effects of different types of Brexit on future economic growth. As reported by BuzzFeed, this leaked analysis suggests that the UK’s economic growth will be 2% lower over the next 15 years if it negotiated a Norway-style deal, 5% lower if the UK ended up with a comprehensive free trade agreement with the EU, and 8% lower if the country exited the EU with no deal.

These predictions have had some predictable reactions. They have been seized upon by those arguing for the softest possible Brexit and by those arguing for Brexit to be abandoned altogether. Nicola Sturgeon described the leaked analysis as a “watershed” moment.

Others have been outright dismissive. Iain Duncan Smith and other prominent Brexiteers pointed to the flawed economic analyses conducted by the Treasury and the IMF before the referendum. Philip Davies branded the latest official analysis as “dodgy figures” cooked up by “London-centric Remoaners.”

Both types of reaction are misguided for their certainty. The leaked projections are indeed economic predictions, not economic facts. But just as the Remain campaign did itself no favours by overegging the certitude of economic forecasts before the referendum (‘Project Fear’), so too might Brexiteers be foolish to bury their heads in the sand.

Some predictions can be better than others. Forecasting the weather and the economy are hampered by a common problem: they are both dynamic systems with many moving parts. Getting forecasts right depends on taking the correct variables into consideration, often through trial and error.

“Officials believe the methodology for the new assessment is better than that used for similar analyses before the referendum,” BuzzFeed reports. In other words, they examined the problematic assumptions of their previous models and have amended these analyses accordingly. Indeed, the leaked models make a number of generous assumptions: that the UK will strike trade deals with the US and non-EU blocs, a swathe of regulations will be cut, and a hard border will be avoided on the island of Ireland.

Indeed, responding to the leaked analysis, a government spokesperson acknowledged that it “contains a significant number of caveats and is hugely dependent on a wide range of assumptions.” This is precisely why the report, if and when it is complete, must be published in full. It might contribute little beyond confirming people’s existing biases, but it might not. That’s the inescapably uncertain nature of predictive analysis.

If that conclusion feels a bit too wishy-washy, here is the real punchline from the leaked forecasts: leaving the European Union is not simply about economics; it is very much about politics. Of that, we can be much more certain.

Also published on Medium.