oik01 FWIW, DC Rainmaker (www.dcrainmaker.com) has done a good bit of work WRT various power meter accuracy levels. It will take a bit of digging within his really extensive body of work to find it, however.

From what little I know about this stuff, I would not view the power #'s from a trainer as the 'power gold standard'. I would also observe that the power measured at the pedal would be expected to be a bit higher than the power measured further down the drive train (losses in the system - probably a few percentage points). Of course you are seeing the opposite in this case.

dave