When you say the Portra was "overexposed", on what basis are you making this judgement? Scans? My guess is that you're scanning and your inversion process is wrong, e.g. incorrect selection of white-point.
I ask because you can overexpose Portra by 3 stops and have it print nearly identically to a normal exposure. Being C41, it's extremely overexposure-tolerant and therefore I doubt that you could even tell that a properly-printed image was 1.5 stops overexposed by looking at it. You could probably tell with a side-by-side A/B comparison but the differences will be subtle, e.g. increased shadow detail and slightly reduced highlight contrast.
If you want to check your DSLR-preview/metering skills, expose some chromes. They too should come out right if you do it carefully.
A DSLR is a perfectly valid means of checking an exposure but you need to make sure that it is setup correctly, i.e. make sure the dynamic-range optimisations are all disabled. The ISO film speed standard is a standard (from ISO, who'd have guessed?), and the digital world follows it just as rigorously as film manufacturers.