The only practical way to get a meaningful top speed (other than a completely windless day) is to do two runs, in opposite directions and take the average of the two speeds. I once recorded 84 mph in one direction and just 42 in the opposite direction. An average speed of 63mph.
However, even this result is misleading as it is the product of the peak speed in both directions and will therefore add any wind speed variation during each run to the resulting average speed. Thus, if the wind speed had changed by just 5mph during each run, then the average mph could be up to 10mph lower than the average of the peak speed measured in each direction.
I guess you could do two timed runs in opposite directions over a measured distance but this would be quite difficult to do with any accuracy.