@augieray You have no basis for asserting the distance is small... what are you basing that off of?
Depending ont he exact value of the variables linear growth can exceed exponential growth at any arbitrary point. It is just as possible exponential curve doesnt win out until the last 1 mm of distance.
The point is you can not say it overestimates or underestimates... All you can say is it incorrectly estimates., far away in overestimates the risk, but at some point arbitrarily close it will begin to underestimate the risk.