There is a particular reproach I hear again and again, and it always arrives dressed as common sense. If you analysts live in a “probabilistic world,” the reader asks, then what is the point of your analysis at all? What can you actually predict? Or is probability just a convenient hiding place — a way of saying, after the fact, that the world is fifty-fifty, the forecast didn’t pan out, and nobody is to blame? It sounds devastating, and it is meant to. But underneath the sarcasm lies a misunderstanding so deep that untangling it tells us almost everything worth knowing about how to think about the future, the war, and our own helplessness before events.
So let me say plainly what kind of world I think we live in, because everything follows from that.
Two kinds of universe
There is a famous fantasy attributed to Laplace: give me the exact position and momentum of every particle in the universe, and I will tell you everything that has happened and everything that will ever happen, a thousand years out, to the second. In that universe the future is already written; it is simply not yet visible to us. Prediction there is nothing more than computation. If your forecast is wrong, you were either ignorant or careless, and there is no third option.
We do not live in that universe. We live in a probabilistic one, shaped at its core by free will, and the difference is not a technicality — it is the whole game. In a probabilistic world a serious forecast does not come as a single dot on a calendar. It comes with a range, a spread, a set of branches, each carrying its own likelihood. And here is the part people resist: living in such a world demands a certain preparation, and not only from the experts. It demands something from the audience too. If you expect a hundred-percent accurate forecast, the honest diagnosis is not that the forecaster failed you. It is that you are not yet ready to live in the world as it actually is.
Consider the weather. A forecast of rain is probabilistic; everyone understands this without complaint. Economic analysis is the same — it offers you odds, not certainties. Now imagine someone proposing that we line meteorologists against a wall and shoot them every time the rain came an hour late or a degree too cold. We would recognize this instantly for what it is: not a quest for accuracy but a Stalinist method, the demand for guaranteed results backed by the threat of punishment. And yet this is precisely the spirit in which people approach political and military forecasting. Wrong about the timing? You must have been lying, hiding, covering yourself with talk of probability. The executioner’s logic creeps in wearing the mask of rigor.
Why the reasoning matters more than the verdict
If the world were Laplacian, the only thing that would matter about a forecast is whether it came true. In a probabilistic world, something stranger and more important is the case: a wrong forecast built on sound reasoning can be worth more than a correct one that was pure luck.
This is not a paradox and it is not an excuse. Think about what a correct prediction actually proves. Sometimes it proves that the forecaster understood the underlying dynamics and read them well. But sometimes it proves nothing at all — the coin simply landed the way the forecaster called it, and a stopped clock is right twice a day. The verdict alone cannot tell you which kind of correctness you are looking at. Only the reasoning can. A forecast that turned out wrong, but rested on a clear, honest, well-built chain of argument, has taught you how the situation works, where its hinges are, what would have to change for the other branch to win. A lucky correct guess teaches you nothing and, worse, flatters you into trusting a method that has none.
This is why I keep insisting that the consumer of analysis has a job to do. If you want to sit back with your arms folded and say “take responsibility for your predictions” — well, yes, we do; our reputations rise and fall on being wrong, and that is fair, and it stings. But if that is all you want, you have reduced yourself to a spectator reading the final score. The serious reader follows the reasoning. They ask why this forecaster expects this outcome, what assumptions are load-bearing, what evidence would overturn them. In a more complex world, being a thoughtful consumer of information is itself an active task. You have to dig in. The forecast is the door; the reasoning is the house.
History has switches
The deepest reason the future cannot be computed is that history is not a single coin sliding forward along a fixed groove. It has switches — turning points where the track can divert, and where the thing that does the diverting is human will.
There is a grand and comforting tradition, running through Hegel into Marx, that says otherwise: that history obeys iron laws, that personalities are foam on a deep current, that whoever sat in this or that seat, the same outcome would have ground its way to the surface. Tolstoy gave this its most beautiful image — Napoleon as a child in a moving carriage, gripping the straps inside and convinced he is steering the horses. It is a wonderful picture. I think it is wrong. There are moments when a single person’s choices genuinely bend the line of events, when removing that person from the room changes what comes out of it. Whether a revolution seizes its hour, whether a faith crystallizes around one founder’s specific vision rather than dissolving back into the surrounding cults — these were not foregone. Personality plays a real role in history, sometimes a decisive one, and any model that scrubs it out in the name of inevitability has quietly thrown away the very switches that make the future open.
And if the future is open, then the reflexive pessimist is making exactly the same mistake as the rose-tinted optimist — just from the opposite chair. Both think they already know the ending. Look at how many developments were dismissed as naive right up until they happened. That Ukraine would receive F-16s was once treated as wishful fantasy; it occurred. That this war would grind on toward a fourth year, rivaling and perhaps exceeding the scale of far larger historical wars, would have struck most sober observers as implausible at the outset; here we are. I am not promising that every hoped-for thing will arrive on schedule. I am saying the planning horizon is genuinely short, that beyond a year or two the interference of factors grows so thick that responsible long-range forecasting becomes impossible — and that this same fog is the reason despair is unearned. The person who is certain things will get worse is forecasting from the same false confidence as the person who is certain they will get better. The honest position lives between them.
The observer is inside the experiment
There is one more objection I should answer head-on, because it is the sharpest. People accuse those of us who look for the optimistic branch — who insist the glass is half full — of practicing a kind of cheerleading rather than analysis. Combat optimism, they call it, as an insult.
I accept the description and reject the insult, and here is why. In physics there is a principle that the observer affects the experiment; you cannot always cleanly separate the measurement from the thing measured. We are in that position. We are not detached astronomers watching a distant galaxy through glass, unable to touch it. We are inside the situation we are describing. There is a well-known sociological version of this — the idea that if people define a situation as real, it becomes real in its consequences; that prophecies announced publicly reshape the behavior they predict. Declare loudly that a sound bank is failing, and depositors will run, and the bank will fail. The text influenced reality. The forecast helped cause its own outcome.
If that is true — and it is — then feeding people hopelessness is not neutral observation. It is a choice with consequences. There is even a grim little experiment from mid-century psychology, with rats placed in water, that showed something extrapolatable to us: the difference between giving up and holding on was the presence of hope, the learned sense that rescue was possible. Take that away and the creature drowns much sooner. We are not rats, but we are not made of stone either. This does not license lying, and it does not license the mindless optimism that paints the glass full when it is half empty. Between head-in-the-sand cheer and corrosive, fashionable despair there is a healthy middle, and that middle is where honest analysis under conditions of real stakes has to stand. To insist on seeing what is genuinely possible, including the favorable possibilities, is not a betrayal of objectivity. It is part of the job of an observer who knows he is also a participant.
What the analysis is actually for
So we return to the original question, but it should look different now. What is the point of all this analysis, if we cannot foretell the future and cannot personally bend it to our will? The answer is the same answer one gives to “why read books, why follow the news at all.” Should we bury our heads and wait passively for whatever happens to fall on us?
Understanding events does real work even when it changes nothing about the events themselves. It dissolves panic. It breaks the sense of isolation, the feeling of being alone and uncomprehending before forces too large to name. It lets people prepare — and people do shape their own destinies in the small, concrete decisions that are actually theirs to make, decisions that depend entirely on how clearly they see the situation they are in. For someone inside a dangerous regime, understanding may mean grasping that they must leave, or at the very least protect themselves and refuse complicity. For someone abroad, it may mean knowing where their voice, their vote, their help can actually land. And beneath all of that lies the most human thing of all: the need to talk, to be heard, to think out loud in company. Thought is not really born inside a single skull; it arises in dialogue. That is why these conversations matter — not because they let us predict the future like a clockmaker reading his gears, but because they help us live, decide, and endure inside a future that is genuinely, irreducibly unwritten.
Demand certainty and you will get either a liar or a coward, because no honest person can give it to you. Demand good reasoning instead, and you might get something far more valuable than a lucky guess: a clearer view of a world that was never going to hold still for the camera in the first place.