Eric defines LL narrowly as “covid resulted from gain-of-function research in WIV”. His final total Bayes factor is 0.001255 (log-odds -7.19), giving P(LL|O)=0.07529%, i.e. ~1-in-1300. He explicitly does not endorse the precise figure but maintains the qualitative conclusion (strong zoonotic) is robust.