When critiquing or beta-reading, my “first paragraph test” runs at about a 80% to 95% positivity rate. It goes like this: 80-95% of the time, the start of the story I’m reading would be better if the first paragraph were deleted entirely and the story started with the second.
For published works? It clocks in at about a 50/50 clip. I’ve found this is also is true when editing my own works…especially short stories that I think I’ve ‘finished’.
I haven’t had the chance to fully think through why this is—but, once I noticed it, there was no denying it.