Mike Fahle Wrote
OCRA used to do a lot of racing using Portsmouth for any boat that showed up (we encouraged that); popular class boats, some "one-offs", any modification that a racer wanted to try, and many "dead" class or rarely raced class boats. Over the years we found it to be very even and predictive - the sailors seemed to finish pretty much in the same positions regardless of the boats they raced; the classes seemed to finish in an expected pattern to each other, etc.
It worked so well that way that one time when we went racing in CRAM we saw results that were posted that we knew could not be correct because that predictability was wrong - one whole class beat all of the next class that started, for example, and (very long story, short) it turned out that there had been a five minute scoring error in favor of the unusually well performing class.
I always thought that a measurement class could be even more accurate since I have always also raced big boats that use measurement handicap systems (both keel and multihulls) and that is the way the moneyed racers have gone (good old PHRF is still alive and well here in MI and OH). I have no experience with that on the beach cat level and wonder if they use race result feedback to check their accuracy as Portsmouth did.