The engineering approach is that if 30mm/s is working best on your equipment (at least on that day), you should use it; at least until it doesn't work so well for you. You might also try more different speeds (closer to 30 than to 10 or the assumed 60) in case one works even better than 30mm/s - though you might also want to question the assumed 60 (or test a set 60) since assumptions can bite you.
There's not going to be a fixed speed that's always best independent of the printer - everything is dependent on the printer, and its environment - if the temperature of the room varies a lot, it could change the "best setting" on days of different room temperatures. Perhaps the "10mm/s" that you saw working well was using a more effective cooling fan while bridging, or had other differences from your particular unit - or had different settings (other than speed - nozzle temperature, for instance) which impacted the result.
(note that I have limited tolerance for videos, and have not actually looked at yours, so if you're going to come back with "but I don't have a cooling fan", please post a picture of the printer or results that does not involve needing to sit through a video. And perhaps add a cooling fan...)
As for "understanding how to best tune your slicer settings" - far more testing, and keep track of the results. So, do tests (without altering other parameters) at 20, 25, 35, 40, 45, 50, 55 mm/s.
Depending on the results of those tests, say you find the best results at 35-40mm/s, perhaps you try 47 and compare that. Perhaps there's no real difference between the 3, so it's in that range, but not picky.
Then alter one other thing (nozzle temperature comes to mind first, but perhaps it's "fan speed for bridging" or some other, single, item) and do more testing - likely the best speed will change, so you'll need to re-run a range of speeds at the new setting. And keep track of which results (test prints) go with which settings, so you can go back to the ones that work best. Then alter that parameter more, or alter another parameter.
For the most part, there comes a point where it's good enough, or at least where you're sick of testing for a while and results are good enough until you want to test / tune some more.
If you change many things at once, it's hard to isolate cause and effect. If you test many settings but lose track of which settings were used for what result, it can be hard to progess towards better results consistently. When you find a range that seems to make no difference (and it's as good as it gets), be happy that the parameter is not too sensitive, and tend towards the middle of it.