Hash, I can assure you that the Waddington effect is real.
In relation to maintenance the probability of stuffing something up when you reassemble and install it is not zero and the defect may not be discovered in initial testing. This effect gives rise to what is termed "infant mortality" - failure shorty after installation in components.
I independently proved the existence of this effect by studying and analysing failures of autopilot gyros in the Ansett aircraft fleet in the late 1970's. Almost all of them occurred shortly after overhaul and reinstalllation. They had very delicate bearings which were easily damaged. We increased their reliability by doubling their life limits.