My guess is that at higher RPMS the digital pulses from the RPM sensor are close enough together and the voltage is high enough that
they "blur" the pulses so they appear to be closer together than they really are resulting in the higher RPM reading. A resistor will
bring that voltage down a bit so the pulses are a bit cleaner. In essence the pulses are wider at higher RPMS and the width is
dependent on the voltage. It would be an interesting experiment to connect an oscilloscope to the signal from the sensor
and see what it looks like with and without the resistor.

V.