As I said above, if the instrument assumed the target was on the ground, then the range estimate would've placed it on the ground:
A better explanation is that the instrument tracked the object as if it was a fixed location high above the sea level, which would make it appear to go slow because the starting assumption is that it's going nowhere. But this would make no sense in air-to-air mode since aircraft usually don't hover in place.
This idea has recently been proposed by somebody on Twitter, who has supposedly worked for 40 years in the aerospace/defence industry. I cannot verify if his credentials are legit, but he sounds like he knows that stuff.
Source: https://twitter.com/RayMoor18837824/status/1579196505927270401?s=20&t=U5B3LX-x7YExhgHGLtwFqg
Now it's interesting because setting up a ~100-Knots headwind in my 3D reconstruction, I retrieve a geostationary object at ~13000ft (speed=0) and range/closing velocities that are very close to the video at my two points of reference (0'11 and 0'31). In other words it really looks like the algorithm is giving the range to the intersect of the lines of sight. So either there is a geostationary object there (stationary relative to ground, i.e. moving against the local wind). Or range is wrong and the object is somewhere else along the lines of sight.
Having a 100-knot headwind gives a better match for the speed of the background in Sitrec. Without headwind the background moves too slow, a problem that was also noticed by
@Edward Current in his GoFast Blender model. I think it suggests the F-18 was more or less facing the wind that was reported that day (120 Knots to the West, assuming GoFast happened just before Gimbal).
This all will need confirmation, but the idea that the range is inaccurate because it is looking for a geostationary object is not absurd given the situation. Now it does not mean that the object is way far from the given range, but if the Navy has additional data that showed that the object was fast (?), this may explain the discrepancy with the displayed RNG. Lehto and Ryan Graves, among others, also believe the range is inaccurate because it is mostly used for air-to-ground, so this is not completely out of the blue.
EDIT: note this is a 3D model I made, it looks 2D because I'm showing a view from the top.