Coffee is very subjective, you can measure it‘s compounds but it‘s not very helpful. In that sense it‘s even more subjective than Hifi, but you can still rate it.
In order to rate products that are meant for subjective experience you need a group of people that are calibrated to each other.
So like with coffee you would have to agree on parameters that are important like timbre variation, dynamics etc…
Then you let the group listen to different gear and everybody rates each variable, after which they compare their findings. If your experience, of let‘s say brightness, is way of compared to the others you would adapt a bit (calibrate to each other, so you speak the same „language“). When all are calibrated you pick the gear you wanted to test and rate it again on the same parameters as before. Now everybody would be very close to the other in their rating as they are speaking the same language and the goals (what good sound would sound like) are clearly defined.
Now, if you would have a calibrated group of people comparing every other DAC to a Dave it would be the winner in terms of timbre variation (the gap between the brightest/hardest sound and the warmest/softest sound is the biggest). There are a lot more variables in which Dave would be better but this is one of the most obvious imo.
When Rob said some people are just not good at listening tests it would mean that a person is not able to calibrate to the group that are performing the test together.
Such a person would always be way of to the others even after lot‘s of comparisons. This is not a problem of course, you can still enjoy your Hifi. Just don‘t work as a tester