Calculating Risk of Nuclear War

This activity was motivated by an Oct 9, 2022 tweet by MIT physicist and machine learning researcher Max Tegmark. A screenshot of the tweet is given below

tweet screenshot

In this tweet and the subsequent thread it is claimed that the model implies that the risk of nuclear war is 1/6. I have no idea what the true risk is (but obviously have a preference), my interest here is to:

To start we look at the diagram of the full Markov model

diagram of Markov model.

Issues in published interpretation in tweet/thread

The author estimated the risk of nuclear war was 1/6 based on
$$ 30\% \times 80\% \times 70\% \approx 1/6 $$

Is this calculation appropriate? Under what conditions can one multiple probabilities of separate events to get the joint probability?

Implementing the Markov model