Statistics are predictive for individuals. If they weren't then, insurance companies wouldn't use them and they wouldn't make large profits off of people by using them.
Judging correlates to the behavioral trait Conscientiousness, and I thought this interesting:
The big five and organizational virtue
"Genetic influence was estimated as 41 percent for neuroticism, 53 percent for extroversion, 61 percent for openness, 41 percent for agreeableness, and 44 percent for conscientiousness."
So, you can even predict the probability of someone being born with a behavioral traits.
What it originally was created for is still important, but its developed way beyond that. Just because people have given bad descriptions of it, doesn't say anything about it as a predictive behavioral trait.J/P was created to keep track of whether a percieving process or a judging process was being extraverted. Then some people theorized about what preffering extraverted judging vs. extraverted percieving would look like. Most of the ideas drawn from that were way too general, or completely untrue.
I'd think that the test results would be more reliable than experts most of the time(why does it matter at all what experts guess about?). The theories are only useful to the extent that they correlate with what is measureable. And, in this case, that means behavioral traits.(by the way, the way to figure this out is NOT to use MBTI test results as evidence. Look at type guesses made by experts)
To tell you the truth, I don't even know what we're debating. I have no fixed opinions about the matter and that is what I'm trying to discuss. We don't yet know many things about typology. Its a young science and hasn't proven its validity in many areas besides J/P. Its pretty much been developing and changing ever since Jung came up with it, and personally I see that as a good thing. We don't as yet know everything about J/P, but why is that a problem? The MBTI Step II is still new. It will be years before we have enough data and research to draw any useful conclusions about how valid is the J/P dichotomy. However, we have enough evidence to show its valid in a general sense even if it needs to be clarified.
It doesn't matter what your opinion is or what my opinion is... or even what Nardi's opinion is. By the way, there really wasn't much that Nardi said that disagreed with anything I've said so far. I guess I disagree with his dismissal of statistics, but he really didn't give much of an argument for his opinion and so I didn't fully understand what he was getting at.