so, if we have an upper bound around 25% from the actual data, and i suggest the sick are being undersampled due to being in quarantine, how high can you guesstimate it is actually?

35%? 45%?

and, could we get herd immunity with numbers around or less than 40%?

if the reproductive number is low, perhaps.

but, that would contradict the basic deduction that it's actually higher than expected.

again: the key takeaway is that it is currently more widespread than previously modeled. more precise bounds are forthcoming.

my guess is that it is much higher in new york city, specifically.