Ordinary models assume that infection is like a steady flow of water that washes over everyone equally. In reality, the stream from a sick person fluctuates: right after infection it barely trickles, then turns into a raging downpour, and later subsides. The chance of catching the virus depends on the strength of this flow at the moment of contact.

The new model describes this unpredictability through the 'age of infection'—the number of days since infection. Scientists broke down the dynamics of contagiousness into time segments, and using Markov's ideas about random transitions, they created vivid forecasts. It turned out to be all about the spread of the virus and the dose received.

This explains why sometimes in a crowded bus no one gets infected, while a fleeting conversation leads to illness: it's not the meeting that matters, but the dose and timing.