Jake, that is a nice summary.

The only thing I would add is that the variance due to boat design is much smaller then the variance between sailing ability. See the F16 discussion.

Second point... the nature of the rules contribute to the lack of precision.
Consider... that for a 30 foot One design boat...and 10 boats in the race... a port tack boat coming into the lay line TIED for first with other perfectly sailed boats will loose three percent in time when the rules force them to bail out and fall into line in 10th place to round the mark. So... the variance will be 3% racing on time even in one design boats because of the rules.

Taking Times will not do better then 3% for 10 30 foot boats and so on.

The major advantage is transparency. Numbers don't lie... the formula is evaluated against the real world. considered by a committe for accuracy and fairness, and applied to all boat classes equally and it is published and sanctioned world wide.... you plug and chug to get your rating...

Everyone has an opinion on the accuracy of the table. If we want to average opinions... that is called PHRF and it can work... it is just not very transparent.

There is no argument that Sarah and JC turned in the rating on the Nacra 17 against joe noname who has the only A cat in the race.... So.. they have created a very hard rating to sail the N17to... The weekend sailor cant sail the boat to it's rating and compete against joe blow on the A cat...
The bar talk about the formula lasts 10 seconds.... the bar talk about Sarah and JC versus joe blow on the A cat lasts indefinitely and is meaningless.

Choose.