Was I right about the math when the covid mRNA's were tested and released into the market?

I can’t find the old Q’s, but when Moderna and Pfizer first were released, they were said to b earound 94% effective at preventing disease. On several Q’s I questioned it, because I live where some of the testing was done, and I knew that we had very little covid where I live, so there was not much chance to catch covid during the testing time. I argued if there is not much chance to catch it, how can we really know how effective it is. Some of the math savvy jellies told me why the statistics would still be accurate.

I also argued that J&J showed lower efficacy, probably partly because it was tested in places with more disease going around, and in fact even scientists were saying it. So, it seemed to reinforce the idea that vaccines tested in low incidence areas would not be well tested for prevention.

Fast forward today when we know the mRNA vaccines did not prevent disease 94% of the time. It does look like they help prevent severe disease, I am not talking about that for this Q.

Why were the scientists wrong if not the low disease rate in the community? What would be the other reason for the mathematical estimate for efficacy to be so off?