In my more serious pursuit of Latin as opposed to my high school Latin days, I have learned that word order does matter, that it is not random. But would there have been a facet of every day Roman life that people would have spoken or written with sentences of free, random word order, perhaps if only for their own amusement?
Latin stands in stark contrast to German as that language is highly inflected and yet in any type of sentence(declarative, interrogative, etc.) word order follows rather strict word order principles with only a little variance allowed, which to me in my limited experience of studying languages appears odd because it seems to negate the advantage of inflected endings.
Some interesting articles:
https://www.jstor.org/stable/3288352?seq=1#metadata_info_tab_contents
https://www.jstor.org/stable/3287962?seq=1#metadata_info_tab_contents
But surely there is something more recent, especially with all the work being done in the latest fad of discourse analysis?
Here’s a very recent take on word order.
https://magisterp.com/2019/07/08/how-weve-been-wrong-about-latin-word-order/
I think this makes it more challenging 
Jim
Ah, no. This is based on an article written in 1918, one of the articles, I believe, that I cited above.
Here are some fun facts from this 1918 article I accessed on JSTOR looking at SOV frequency of esse, main clauses, and subordinate clauses in Caesar’s De Bello Gallico I & II, and Cicero’s De Senectute
I was hoping to find material written more recently, particularly from a discourse analysis perspective.
I don’t know how to Latin, but I do know how to Google.