So what's the complaint here? That GPS isn't always active and recording every single path you take (incl. when not exercising)? Or that steps are misaligned with what the phone reports? Because throwing in GPS won't just 'fix' the latter; the discrepancy is most likely getting counted elsewhere, not because the Watch is calculating less steps than the phone at the same time since that information is hashed together in Health Book (right?).
But, that's based on my experience where Health Book and Activity consistently report similar numbers, though.
It's important to remember that
arm swings aren't the only thing getting counted as you walk; the Watch and iPhone are sensitive enough to feel the impact of your foot on the ground and
that, among other things including arm swing, is a contributor to how stride is determined, too. However, while having an always active GPS would give you the best results, bringing down the interval is a good way to win on a lot of fronts with acceptable tradeoff. There is no such thing as right or wrong unless you don't understand the complexities of the system as a whole.
Then again, we don't have all of the facts that go into determining how they come up with numbers, let alone why they chose to generate those numbers. Example, for all we know, when the iPhone switches towers, it could trigger a GPS fetch and update with the distance and stride formulas on the Watch since it wouldn't have to explicitly activate the GPS chip (If you don't do this, do it Apple).
Of course, all I say is speculation. How would I know?
