What am I looking at?

Before the Olympiad kicked off, I ran 5,000 simulations of the tournament to come up with a prediction for the probabilities that each country wins (or gets silver/bronze). Then, after each round, I run new simulations based on the current results. My simulations also take into account the official FIDE pairing logic.

In case you are wondering, yes, it was a nightmare to program. And its not quite perfect. For example, Im tracking how many whites each country has played. So when two teams face off, whoever has had white on board one more times will get the black pieces on board one. But when they have played white the same amount of times, then the white pieces should go to whoever had white less recently. Right now, I just randomize who gets white - its a small detail, not materially impacting these predictions. But its a good example of the complexities of the Olympiad pairing algorithm.

That was a little bit of an introduction to the simulations for this particular event, and you can read more technical details here about the underlying model, method, and simulations. You can also find the code on GitHub.

Note: If a country is on the list above, it means they at least won Bronze one time in one simulation, even if it says 0.0. Countries that never win a medal in current simulations will not be in the table above.