Just curious, I found this old video I made of the vacuum reading of my old '76 350 (RIP) and I was wondering, was this a good vacuum reading? How does one measure it properly? I had it hooked to manifold vacuum and had the engine idling in Park... is that the correct procedure? There is no engine to test anymore, but I will use the knowledge in the future.
Needle drifting usually from mixture being off and some from the idle surging (rpm). Not enough initial timing or low compression (depending on the big picture) causes lower readings like that. Some of the trick to the gauge is to try to throttle the engine in a way that holds at a number between 0 and the 21 seen, such as 10 or so, duplicating part throttle cruise under some load. There and the transitionary throttle changes can show up erratic movements like valve sealing, ignition issues, weak springs and other mechanical problems (seen just AS you crack the throttle open, before the gauge jumps up to the full amount, or what it is doing before the revs catch up.)
Temporarily advancing the timing to see what the gauge does near idle will answer some questions. Help tune the idle and drop in gear (with a friend) while you are there. Now you can see first hand that compression on the low side requires more lead. You can't tune the carb until the timing is close, it will vaporize better, draw more intensely, etc.