I've just been posed an interesting question. The only straight and flat real estate available local to the race course for laying out a calibration course contains a speed bump. Since riding the cal course would produce a longer distance than was measured using the steel tape, is this acceptable? My initial view is that it's probably OK. What does anyone else think?
Speed Hump on Calibration Course
It's probably no worse than a slight swerve. And the error is in the right direction. I suspect that it would be difficult to detect any difference between the course with a hump and one without.
Usually speed humps tend to taper to road level near the road edge - I'd say lay out the course as close to the edge as practical. That said, Pete's comment is probably on the mark.