No, I dont think there's much difference between a 65 tester and a 70, but if there wasn't a difference, why is the 70 the minimum? The Olympic committee doesn't take 5 seconds off the Jamaican bobsled team's time. Is there a difference between a team that comes in 32nd vs the team that comes in 31st? Neither is going to medal. Now if taking 5 seconds off the time means that they'll get on the podium, then that's a problem. In this case, we're letting guys stand on the podium that wouldn't necessarily be there.
Can we stick with just one analogy?!?
There's not just one "Olympic committee" deciding how to treat all the professionals in the world. There are 53 "Olympic Committees" in the U.S. alone that exercise their authority over their jurisdiction (the 50 states, DC, PR, and Guam... if I haven't missed one or more) - and they're free to choose how they "qualify their athletes". NCEES is *not* the Olympic committee.
NCEES provides a service that just about everyone follows unchanged. Georgia is an exception. The cut score of 70 is deemed the border of "minimally competent" by NCEES. There has to be *some* number but there's also got be be a fairly large 95% confidence interval!
Bottom line (for me): If you accept that someone who takes the exam five times before passing with a 70 is "competent enough" to be a registered engineer, then you should accept that someone who fails with a 65% is probably "good enough". And remember, no state looks at just the exam on its own - they all use a combination of education, experience, and exam.