Also, it feels like this statement:
Is relying on a heavy amount of narrative bias, since by your own admission there isn't data to quantify whatever you're trying to exhibit:
It's all relative to the resolution of how that data is collected.
Personality is consistent across long periods of time. There's always gonna be a high error rate, but my guess is it correlates to behavior in such a way it's impossible to remove it from an equation.
Personality profiling is really useful in figuring out how people will perform in modern work environments. Environments that are far far harder to define than skating between nets.
Again I'm unsure of what I'm asking because I'm unsure of what exists.
To ignore personality seems absolutely absurd. I'm convinced it must have been done, I'm not that clever and it isn't a trivial thing.
It's gonna heavily correlate to the stability of players, their ability to adapt etc
Power play killing, drawing penalties, cohesion with new team mates, success when down points, ability to travel, ability to improve and the streakyness are all things heavily rooted in personality.