As the person who did these measurements I feel like I need to chime in to correct this. I did this in response to some comments on a podcast where they felt the remakes had a longer flipper delay than the originals. When I initially did the measurements there was a grounding issue on the original which made it look like it had no delay at all but this wasn't the case. This was corrected in the first page of that thread. Here is what was found:

On original WPC95 games like AFM, MM, MB there is a variable 2-5 ms delay between when the switch on the flipper is closed and the flipper is energized.

On remakes there is the same amount of variation as the originals with an additional delay of about 1.5 ms. This additional delay is likely not perceptible but this its unknown as I don't know if anyone has ever looked at this in a rigorous way.

The engineers at CGC said on future versions of the remakes they would introduce code that allowed the delay to be reduced or eliminated. Perhaps even eliminating the variation that was present in the originals so people could decide if this made a difference or not. As far as I know such changes have not yet been introduced.

I personally preferred the play of the originals to the remakes but skill can't put my finger on why. The games are different at many levels and the differences between them can been seen from across the room so impressions of the games may be shaped by many factors. There are people who cannot detect any difference in gameplay and some people on both sides are very passionate about their opinion on this. In short preference for an original vs remake is an individual choice that you should make for yourself.