Yeah but I have yet to read any research that says those two lead to scoring more points, or that the risks of an interception/fumble (not unlikely outcomes of rushed delivery) are worth the negative impact of a sack.
Sabermetrics works because it is built on 30 years of research that went into understanding what increases the likelihood of a Run. It seems to me that these football wannabe stat nerds skipped the underlying, fundamental work that is required and simply started assigning random values to Visible outcomes.
I think my favorite exchange ever with someone at PFF was his insistence that QB's had absolutely, literally, without a doubt ZERO impact on yards after the catch. That one was a real winner for me, and showed just how far these lazy clowns have to go.