It’s a bunch of nonsense really. They greatly simplify models, otherwise they’re too complicated to calculate. So they simplify the data points, assume point

It’s a bunch of nonsense really. They greatly simplify models, otherwise they’re too complicated to calculate. So they simplify the data points, assume point particles, assume no interactions due to electromagnetism, no tidal locks, and Newtonian gravity instead of relativity. And then it turns out galaxies sometimes rotate too quickly. Yeah, no shit. If your data is known to be wrong and your model uses the wrong theory of gravity and makes known false simplifications it would be quite strange if it somehow did predict without some discrepancies