I agree too, (have been in the computer business for about 40 years).
Final testing before shipment from the manufacturer gets most of the infant mortality end of the "bath tub" curve, and those shipped are very, very stable.
To have, as we are told, six similar "crashes" is statistically not possible.
Oh, first computer - IBM 650 (bi-quinary) with mercury-delay lines (think big shift register) and drum storage, circa late 1950s.