While this could be good, on the face of things, I have had many discussions with people on the face of scoring/ rating items lately. Oddly with consumable items things can be very personal, as there is no real way to quantify a right vs a wrong like there is in certain school subjects. In regards to what you said about for some people 3 being decent to good, and others it being not that great, comes from the fact that some people feel that when they have a rating system you should be able to use the entire rating system. Basically what is the use of scoring out of 100 if you are only going to use the numbers 70-100?
From the standpoint of a teacher, I feel I have a bit more of a unique view on scoring systems. In theory a tea or basically any other consumable is a sort of finished product. Like a project or other in theory polished homework assignment that a student turns in, it shouldn't be something someone wrote up in streaming consciousness in a few minutes trying to just turn something in ( doesn't stop it from happening though). So in my opinion the grading scale for these finished products should naturally be tilted towards the higher score end, if you take the view point you are grading a tea on its merits.
A though F could help fix that, but at the same time the same people that want to set 0 to be the worst item they have tried in a similar vein, would probably also set F to be the worst item they have tried in a similar vein, even if it is perfectly drinkable, or just contains flavors of which they themselves are not fond.
Not to mention I have been around higher education long enough to hear of some weird grading styles. While I tend to prefer the grade on merit approach, where points are given for good things, and only if something is not quite right do they loose points, I have heard of some scoring systems where every mistake results in the loss of points, the more points for how sever the teacher considered a mistake. In the later grading style, its not unusual for students with very strict teachers to get back a test or a homework assignment with a negative score on it, as in if out of 100 points they lost more than 100 points.
So I guess this is a very long post to say that every rating system is nearly pointless. The only way they can ever make some sense is if you understand a persons ratings well enough, and how they relate to your own personal ratings, to be able to naturally translate them to your own rating.