In a lottery draw the drum turns, the balls tumble, and one drops into the tray. Nobody can say in advance which. Yet inside the drum, every ball goes where a gear pushed it.
Alexey Semikhatov, a theoretical physicist at the Lebedev Physical Institute in Moscow and a Doctor of Physical and Mathematical Sciences, asks us to hold that picture for a moment. Quantum mechanics has no drum. Apparently there are no gears inside, and nobody can reach in to nudge the outcome.
That matters more than it sounds, because of how we use the word "chance". You run into a schoolmate on a tram, a complete coincidence. But look closer: you set off this way, they set off that way, they were held up by a phone call, and the two of you ended up in the same carriage. Every chance we know from daily life is a heap of causes too tangled to sort out. In the quantum world nobody has found a heap like that.
Here is what the quantum world does. Take a system with few moving parts, so you can prepare exactly the same situation again and again, exactly. Measure it. You get result one, then two, then one, then three, then one again. Same preparation, different answer, and no way to predict which comes next.
It isn't chaos, though. Not just anything turns up. The possible results are all written into the wave function, the mathematical object that carries the system's inner life, and the odds can be read straight off it. Run the experiment ten thousand times and you might find result one in 80 percent of runs, result two in 15, result three in 5. If the number for option five has shrunk to zero, you can measure forever and never see it. If it is tiny but not zero, it will turn up now and then.
A fair coin gives the nearest comparison, and it misleads in a useful way. Fifty-fifty does not mean heads, tails, heads, tails. Real tosses come in streaks, sometimes long ones. The odds describe a crowd of trials. A single toss, a single quantum measurement, cannot be predicted. Just can't.
Does the wave function secretly know what is coming? No. It is prepared the same way every time, and it still hands out different results.
Here Semikhatov makes the turn that surprises people. Between measurements, the wave function follows the Schrödinger equation, which is smooth, orderly and reversible. It runs backward as easily as forward, the way astronomers run the planets backward to find a solar eclipse in the first century. He calls this part of the theory "kind". Then you measure. The measurement is "evil": it breaks the equation, one possibility falls out, and there is no going back.
The fairy-tale hut on chicken legs is his picture of it. You ask the wave function for an answer, and it turns to you with its right side, then its left. You cannot influence which. The wave function grew out of causes, so causality is there, but the final turn of the hut has no mechanism.
Quantum mechanics has nothing like that. That is, how probability, how chance, works in the world with no mechanism behind it, and this thing is, you know, philosophically tricky.
В квантовой механике ничего такого нет. Каким образом вероятность в мире, случайность в мире действует без обеспечительного механизма, и эта вещь такая, знаете ли, философски непростая.
— Alexey Semikhatov, Gleb Solomin Podcast, 26:49
The theory itself works superbly. It gets the electron's magnetic moment right to twelve significant digits. It simply stays silent on the middle of the story: who, where, how. Physicists have a rule for that silence: shut up and calculate.
That randomness is solid enough to build on. A paper in Nature in May 2026 describes a device made of superconducting qubits that starts from weak, correlated randomness and produces virtually perfect random bits, with a Bell test certifying the result.
So the coincidence on the tram has a story behind it, and the click of a detector in a laboratory does not. The equation says what the click can be. What makes it this one and not that one, no one can yet say.
