From its sensitivity to shocks in global supply and demand, to outside factors such as the weather, oil prices are notoriously difficult to predict.
After all, hardly anyone foresaw the current rout which has wiped around 70 per cent off the value of crude since the middle of 2014.
But companies around the world have been forced to make big bets on when oil prices will recover. Leading players in the US shale gas industry have set their sights on oil at $40 per barrel. Meanwhile, big oil majors such as Royal Dutch Shell and BP are hoping oil prices eventually stabilise at around $60.
Similarly, oil exporting economies such as Saudi Arabia and Russia are burning through large foreign reserves which they'd amassed when crude prices were high. If the black stuff doesn't recover soon, they'll have to make some tough decisions.
But future changes in oil prices could be forecast using previous statistics and computer algorithms, according to new research by academics at the Gulf University for Science and Technology and Plymouth University.
They found a gene expression programming (GEP) fed with several years of data “almost perfectly” predicted subsequent years’ oil prices. They said that it actually outperformed more commonly used statistical techniques.
"The price of oil affects people everywhere, whether they live in countries that are net importers or exporters of the commodity," Dr Ahmed El-Masry, associate professor of finance at Plymouth University, said.
"If policy makers and economists had a tool which could accurately predict future prices, it would enable them to plan for the future at the same time allowing consumers to have an idea of the rising or falling costs they might incur."