I retract the “ALL” and “impossible” part of my statement, assuming that the Babylonians astronomers did indeed believe the Earth was flat. The Saros Cycle was based on timing. I was correct that their accuracy was not perfect and they couldn’t predict the path.
I believe that Babylonians COULD have tracked eclipses using the Saros cycle ... IF they had more data, readings/observations from different places, since their predictions of eclipses were for those that they could see.
In any case, the bottom line is that if you have enough data and do enough math, you can calculate any kind of recurring, periodic movements, even if from different coordinate systems.
You had Tycho Brahe with his geocentric-based Tychonic circles, and most planetarium software is actually geocentric.
In the end it doesn't matter whether the sun and the moon are the same size (roughly, as many FE conceive of them) or whether the sun is four times as large and four times farther away (such as when Greek astronomers, who also predicted eclipses, consider it to be a million miles away) or 400 times as large and 400 times farther away, as those amount to the same thing.
If you have enough data points, e.g. readings / observations, and do enough math, you can predict anything that has a regular geometric / periodic / recurring movement.
There are some anomalies that are not fully explained with regard to modern science's claims about how eclipses happen, so their model is far from perfect either.