Oh, they’re using a 15-year old database, back when VASI installations were still marginally predominant in the US. Aside from the specific issue this thread addresses (longitudinal placement and the effect on landing aimpoint), the type and side of runway, and whether they even exist are as often as not completely off. If you extrapolate the error rate in my random sampling, it will affect thousands of installations (along with approach lights).
It’s simply not effective or reasonable to expect scenery designers to go in and fix each one. They basically need to get the new data (which are publicly published every 28 days in the US) and do a full, system-wide re-run.
I have a separate wishlist addressing the overall wrong/missing light types. But it only has 13 votes which tells me either people don’t care, or more likely, they don’t know what they don’t know and why they should care, for the sake of fidelity.