An interesting proposition, however I would argue that it is actually even harder in the case of civil wars because you don't know what to compare to. Let's pre

An interesting proposition, however I would argue that it is actually even harder in the case of civil wars because you don't know what to compare to. Let's pretend that we compare the set of all worldlines where country A goes to war with itself to the set of all worldlines where it does not. The variance within each of those sets is going to be enormous with regard to the relative rates of progress. Another thing to consider is that in the case of a civil war we could simply treat the two parties as distinct states for the purpose of analysis. Yes, there will be some peculiarities such as supposedly being governed by the same set of laws, however if we take only the United States and France as examples, the factions that ultimately developed basically lived under two (or more) distinct sets of laws. I say this because in the US federalism was much weaker before 1865 and in France the class system divided the population and the difference in the treatment of different classes under the law has been argued to be one of the key factors contributing to 1789 (see Tocqueville).