> The ISS's changes like that though because it's constantly doing burns to keep it "afloat" an abandoned satellite in LEO is going to eventually just burn up in the atmosphere and have a much more consistent orbit.
This is not accurate. Even if you ignore the big vertical jumps at each burn, the downward trend is very obviously irregular. Satellites in LEO experience a slight drag force from the Earth's outer atmosphere, and the magnitude of this force can vary unpredictably by orders of magnitude depending on space weather conditions.
Satellites in orbit are moving very quickly, so a slight change in altitude (i.e. orbital period) results in a very large change in position at any given future time. More importantly, the inaccuracies are compounded with each subsequent orbit. As a rough estimate, an uncertainty of 1m in the height of a satellite in LEO translates to a positional error of more than 100m/day. It is not possible to accurately predict the position of a satellite without ongoing observations.