Niels Bohr was supposed to say:
"Predicting is hard, especially when it comes to the future"
You may be surprised how perfectly the completely contradictory forecasts and theories can be justified.
However, many people still ask us: when will energy prices be the lowest, what will gas prices look like in the coming
months or even years, etc.?
Managing the risk of purchasing energy is too unattractive for many people. They are more likely to accept price forecasts. Based on forecasts, is it possible to effectively manage and manage your energy costs, i.e. make effective purchasing decisions?
We believe that the word "we forecast" should be replaced with the sentence: we analyze our purchasing portfolio (based on our one strategy), indicators, events, trends and prices on commodity exchanges in order to know if we are to make a purchasing decision and how much to buy today to avoid the risk of future price movements.
The key question here is the following rhetorical question:
Is it possible to predict (i.e. forecast) natural disasters, political decisions, or the next harsh or mild winter?
Of course, there will always be those who succeed and grow in the eyes of the opinions and CEOs of companies. This is how Paweł Motyl put it in the book Schrödinger's World that I read recently:
"The whole problem with the handful of infallible is that they are more or less as fallible as everyone else, but with one spectacular hit they start to be treated like an oracle."
And further:
"With a gigantic number of forecasts, the probability that someone will hit is close to 1, therefore, from the point of view of predicting the future, we should be interested not in a single success, but in the repeatability of accurate predictions."
Filip Tetlock, a Canadian psychologist and political scientist, conducted a very interesting experience. He asked himself: How accurate are the forecasts of the most eminent authorities? In Paweł Motyl's book we read "It took him 20 years to find the answer. In the first step, Tetlock selected 284 experts from various areas, people at the intellectual forefront of their specializations. In the following years, he collected 82,361 concrete, measurable forecasts from them along with the probability estimated by the authors. their implementation, which guaranteed to sow good luck or bad luck in forecasting. (...)
Results?
(...) Experts showed tragic predictions no matter what field they dealt with, how many years of experience they accumulated, and even no matter if they had access to inside information in preparing their projections. They were wrong in both short-term and long-term forecasts. They couldn't even figure out probabilities. From among the events deemed to be completely impossible, as many as 15% happened, and more than a quarter of the forecasts deemed absolutely certain turned out to be incorrect. Worse still, the Canadian observed another surprising correlation: the belief that his own prediction was correct was inversely correlated with its accuracy, that is, the more an expert was willing to give his head for his words, the more likely he was to lose it. "
Of course, it is human to be wrong. Mathematical models are also wrong. So is it worth betting on the competitiveness of our business, or on a set budget with forecasts that 'maybe' will come true, or put portfolio management in the hands of advisers, experts or consultants who 'know' how the markets will behave in the future?
In my opinion, the author very aptly concludes:
"Let's also allow a scenario in which instead of trying to predict the future at all costs, we learn to live with the surprises that will surely come."
The conclusion is obvious for me. If the future is unpredictable and energy markets are no exception to this rule, then there must be another model that supports making difficult and effective decisions. A leadership model where action scenarios are created much earlier and ready for any market behavior. A model that relates directly to the company we manage and for which we are responsible. There are no general best theories and patterns. For our companies, we have to develop them ourselves.
Our decisions are like with Schrödinger's cat - good and bad at the same time. The question is what will you see when you open the box?
Author: Bartosz Palusiński
Date: 7th February 2021
