To clarify I have independently calculated the error as the standard error on a set of Monte Carlo orbital elements but the result I get is around 10 times smaller than the error returned directly by find_orb and I was trying to understand why this might be.
Cheers,
Chris