This isn't fully true. For clinical reference to known "good" number ranges, then yes, you want to make the comparison as accurate as possible.
Butm there's a lot of potential unlocked with more data.
For example, we could discover that daily BP averages are more predictive than isolated doctor's office measurements. Imagine two people with equal BP measures at the doc's, but one has a more stressful job with higher average BP. We might reasonably expect them to have worse cardiac outcomes. Or consider the person who's anxious about docs and consistently shows higher numbers in the office than outside; we'd expect the reverse!
Another factor might be examining the variance. This already has precedents, since we know there are things that heart rate variability (HRV) correlate with more strongly than heart rate itself.
This is a great point. Blood pressure readings are only useful in the context of stationary measurements under known conditions. Blood pressure is inherently volatile and depends on activity and recent movements, or to extreme events like cold exposure.
There are even finger-cuff blood pressure monitors designed for continuous, constant blood pressure measurement in the context of a lab or hospital setting. Their measurements would be useless in the context of someone moving around all day, but it can be useful in the context of someone under anesthesia or undergoing a controlled lab test.