I remember hearing an author tell a story about, when she was a child, writing long, long stories with dozens of characters who started on a grand adventure.
After a few twists and turns, the author couldn’t think of anything else to do with them, so they died in a fiery bus crash.
The author said she did this again and again: develop a few too many characters, start them someplace, and then watch the whole thing fizzle out.
Eventually she learned to hone down the characters, the plot, and the focus, and she became the kind of author who told stories in front of crowds.
Writing science can be like this too. Too molecular (or too expansive) and the only way to wrap up is an atomic explosion. Not molecular (or galactic) enough, and a reader will never understand perspective or purpose.
I’ve had my share of drafts that end in fiery bus crashes. I doubt I’ve had my last.