If I could just null and get 3 places beyond the decimal.
The problem I see with the short is the center pin rotates. Say you bought a new SMA adapter and start twisting that center pin around in your new part. Worse, you didn't inspect the short to make sure it had no burrs. Now you plug your new load into a connector you just damaged. The damage spreads like covid and soon you are left with a mess.
You really want to know how well the standards behave at the frequencies you plan to use them. You won't get that answer with even an 8 place DMM.
Even if the parts from my V2+/4 were 51 ohms DC, they would still far out perform that standard I show, except at very low frequencies. The meter is still on, so for fun here is some more DC resistances without nulling the meter:
Custom PCB standard short: 0.015 This is the short shown in the previous picture.
NanoVNA V2+ standard: 51.021
Mini-circuits ANNE: 49.822 Used to replace the original part and shown in that graph I linked.
Cheap unknown BNC Ethernet terminator: 49.921
I would guess that BNC terminator is the worse of the four for RF work. I wouldn't be surprised to find an axial part inside if I cut it open. But again, if your goal is only to make measurements at very low frequencies, sure. Are you trying to measure SWR at 4MHz or 4GHz? You can get away with a lot below 100MHz. I made this RF circuit, which I am very proud of, to demonstrate this point:
https://www.eevblog.com/forum/projects/20db-rf-attenuator-seeking-feedback-to-improve/msg2924286/#msg2924286Skip down to my next post and you can see it running at 300MHz with one small change..
*****
That picture I showed of the two gentlemen teaching a class on how to use the nano makes sense if the standards they show are for the HF bands. Still, very poor construction like the attenuator I show in that link. Not something I would show other than to make a point.
Hope this all helps.