So the reflected power is down ~10dB on every band. Huh.

I don't know if we can yet say it's the op-amp, but it does sound plausible.

What I can't think of is a way for it to be damaged by the previous owner unless maybe he killed it when he tried to replace parts himself. (Did we ever get confirmation that was the explanation for that crappy soldering ...

Since you're getting different values of SWR that get worse with worse SWR but not on the right scale, does that seem to implicate the op-amp itself as the issue?

REV_PWR on the PA/Filter board schematics seems to be relevant. First page of that doc, bottom right corner.
Looks like there's an op-amp LM358 for FWD_PWR U1009B and one for REV_PWR U1009A (edit: two sides of a dual op-amp I guess) which are ultimately fed to the SDR board via the 2x5 alex header ...

Exactly how does the SWR get measured?

May I suggest that you use a dummy load for issues such as this in order to take more variables out of the 'equation'.

Tom
W4TMW

Please excuse my ignorance. How might a dummy load help him isolate the difference between three SWR readings, where one is presumed erroneously low?

OP: No idea but ...