the160,000 sex non specified
You can indeed produce range of numbers by using different denominators, trying to interpret the Census.
You can plausibly get rates between two and five times men in general - it's not really possible to get down to equal to men in general. Wings of Scotland has already gone through this quite comprehensively:
wingsoverscotland.com/the-rorschach-test/
The highest possible denominator would be to assume half the 262,000 are male, and actually trans. In which case it's 92/131,000 = 1 in 1423, which is still significantly higher than other men (1 in 2,212).
The actual calculation shows that it is about the SAME as MALE sexual offending,
No way.
You can only get to equal if you assume every one of the 262,000 including the "trans men" are male trans women, which is implausible.
But as I've said elsewhere, I'm convinced at least 50% and maybe up to 90% of the "no specific gender given" set of 118,000 are anti-gender responses from people who don't identify as trans. That's the only way you could possibly get such a big bar, and there's nowhere else they could be.
We need this clarified, but that means I would personally use (trans women + half of non binary + half of other + one quarter of the none) as my conservative denominator for male trans.
So 92/(48,000 + 15,000 + 9,000 + 30,000) = 92/102,000 = 1 in 1109 = twice other men.