Well, the guy in that link sure seems to think so, and is convinced that that's the problem. So now I'm stuck with the question of who to trust. You or him. And if you don't know, then I would have to believe him over you (no offense). Also, I'm still left wondering if "Hardware NAT" is the same as "Hardware Flow Offloading". Does anybody know?
I'm familiar with that. However, you said that video showed full gbit speeds, but, as I posted earlier, that isn't what I observed in the video. I saw him getting ~300Mbps on 5g wifi (at around 18:40 into the video). That's not full gbit is it? Am I misunderstanding something here? This is important as it helps to set expectations for me.
That is something I am not sure how to do. What is qca firmware?
Sure, it could be possible to get 270Mbps on an Archer C7 with OpenWrt, but, yes, I'm not sure what that combination is, either. 
But at this point it seems to be OpenWrt's lack of implementation of Hardware NAT support for the Archer C7, based on the info I've gleaned so far. But it does appear to exist for mt7621 devices.
Perhaps a better question, and one which could help illustrate my conclusion, is:
"Why is there no implementation of Hardware NAT support in OpenWrt for the Archer C7, but there is for mt7621 devices?"
Fair question? Because it sounds to me like this is the culprit.