I wonder why there isn't anything comparable for stamps?
You yourself are of course welcome to collect eBay
data and present a statistical analysis online. The sample number looks large, but Roman coins are a drop in the bucket against the overall US coin market. The author gives some off-the-wall stats for things that most collectors don't bother with. A breakdown of coins of Trajan by sales to current countries? Maybe that's the author's favorite era. Distribution of prices of coins by emperor is not a very useful or meaningful stat.
I would like to know how many are fakes.
For Roman coins, probably far less than you might expect. Anthony's is in the coin business, too, though.
There are lots and lots of corroded/badly oxidized coins around, fun to rummage through and try to identify, but worth little. There are fair number of some pretty nice quality ones that are only worth a few dollars. People get excited about Roman gold coins, but the percentage of gold in many of them is relatively low and their prices often just mostly reflect the amount of gold within. In this world, the problems are more about misidentification (easy), overgrading and overpricing (same old thing) and sales of stolen high-end coins.