When in Rome, do as the Romans do...murder someone.
If you're reading an ancient mystery at the moment, then odds are it's set in Rome or Roman Britain.
Steven Saylor and John Roberts both write in exactly the same years of Late Republican Rome. Not only that, they use the same historical characters and sometimes even the same events. Both have a novel centred around the Catiline Conspiracy. This has its cool aspects. You can read side by side Saylor's Catilina's Riddle, and Robert's SPQR 2: The Catiline Conspiracy, and get totally different viewpoints from two great writers of exactly the same historical event.
I happen to know (because John told me...how's that for a name drop?) that Saylor and Roberts have a gentleman's agreement not to read each other's books. I can't imagine anyone confusing the voice of Saylor for Roberts, or vice versa, but I suppose they're being understandably cautious.
Caroline Lawrence, Lindsey Davis and Simon Scarrow all write in the period of Vespasian and Titus. In fact Vespasian and Titus appear in all three series.
This means five of the biggest names in historical mysteries are focussed on two precise periods, separated by less than 120 years, whose total duration is less than 40 years, even though there are 3,000 years of ancient history to choose from. This is remarkable.
It's probably no accident that these two periods bracket the the end of the Roman Republic, to the end of the first Roman dynasty, the Julio-Claudian line founded by Julius Caesar and Augustus. In between, David Wishart has his Marcus Corvinus working crime in the reign of Tiberius, Robert Harris has a thriller in Pompeii, and PC Doherty has Roman mysteries and also wrote Domina, the best novel of Agrippina, the mother of Nero, which you're ever likely to read. There are lots of other mysteries set in the time of the Julio-Claudians, but I'd go mad if I tried to list them all, and I don't know them as well as these authors.
If your ancient mystery isn't set in Rome, then chances are very good it's in Roman Britain. Kelli Stanley, Ruth Downie, Rosemary Rowe and Jane Finnis for starters. Roman Britain is huge for mysterious deaths. (And I can only assume the surviving characters all settled in what later became Midsomer County, where their descendants carry on their homicidal habits to this day).
I think it'd be fair to say that Rome + Roman Britain covers 80% or more of all ancient mysteries, and I'm being deliberately conservative because I don't have exact numbers. I suspect the true percentage is 95%+.
If you're not in Rome or Roman Britain, then you might be in Ancient Egypt with PC Doherty, who has a successful Egyptian mystery series starring Amerokte, the Chief Judge of the Temple of Ma'at, or Anton Gill's Huy the Scribe.
After Egypt comes Ancient Greece with Margaret Doody's Stephanos at work during the rise of Macedon, and PC Doherty (again...the man's amazing) with his mysteries set in the time of Alexander.
It's astonishing how concentrated the historical mysteries are. Why these particular periods? The same question applies to later times: Mediaeval mysteries occur mostly in the abbeys and palaces of Britain. Victorian mysteries are mostly in London.
Where are the mysteries of, for example, Carthage? Or Phoenicea? Or India? Robert Hans Van Gulik wrote an excellent series of mysteries set in mediaeval China, starring Judge Dee. Technically mediaeval because the stories are 7th century, but they actually have an ancient feel to them because we tend to associate all things Mediaeval with the Church and Europe, which is a cultural bias. Yet his fine series seems to be sady under-appreciated. I think it's because English readers in the western world have trouble seeing what Mediaeval China means to them.
I've pondered long and hard about this. I think authors are writing, and people are reading, the periods which people can easily see had a major influence on their modern lives. This explains why Roman Britain is huge but not, for example, Roman Hungary. Many historical readers live in the UK, or are descended from there. When they read a mystery of Roman Britain, they read about their ancestors. The doings of the Roman Empire affects us to this day. When we read of Rome, even if we're not Italian, we can say to themselves, "Yes, the past was like this, and that's why my modern world is as it is." Every popular period for historical mysteries has a foundation point for some major aspect of modern western life.
If I have this right then the Golden Age of Athens should be screaming out for a mystery series. It's when western civilisation was founded, after all. I will be testing this theory in October when The Pericles Commission goes on sale. I'm confident I have it right.
Conclusion: You can't go wrong if you set your ancient mystery in Rome or Roman Britain, but for all our sakes, I beg you consider the other 3,000 years of ancient history and a whole globe of cultures. Go out of your way to show the reader why they should care not just about the story and the characters, but also the time they live in.