Here's something I see a lot in submissions:
The protagonist is a member of, or working with, an organization that's either been around for a while or is run by people with a lot of relevant experience.
Protagonist is on a mission, or in the field, or sitting around the office, when Something Goes Wrong.
The thing that's gone wrong is the kind of thing that even a non-expert, given ten minutes of thinking about the ground situation, might expect to go wrong.
But nobody in the organization has ever planned or prepared for it.
And so the protagonist has to improvise and figure out what to do.
This setup would make for some pretty dramatic and compelling action—were it not for the fact that I'm usually thinking “Why didn't it occur to anyone in this organization that this particular thing, or something similar, might go wrong? They've been in this business for decades; why didn't anyone make allowances in the schedule for the possibility of this happening, or come up with a procedure to follow in case of this happening, or mention to the protagonist that they should be on the lookout for this happening?”
I know that in the real world unforeseen disasters happen regularly.
But in the industrialized world, a lot of disasters are avoided because people with relevant experience have thought about the possibility and have prepared for it.
And when seriously bad stuff does happen, like a bridge falling down or an oil rig exploding, it's usually despite a lot of experts having worked hard to prevent it happening; it's usually not because none of the experts gave any thought to the possibility that something simple and obvious could go wrong.
There are exceptions. (It didn't occur to anyone to check whether the units were metric or not?) But like so many things in real life, those kinds of things (imo) tend not to make good fiction, because they sound implausible.
This is, in a sense, a subspecies of the “idiot plot”; it's a plot that works only if a bunch of experts make a significant mistake about something basic. I just don't find that satisfying.
There are genre conventions that contradict me. For example, in real life, as I understand it, car gas tanks don't generally explode when cars run into each other; that's partly due to physics and chemistry, but partly because of design decisions: fuel tanks are carefully placed to avoid certain dangers. But in action movies, it's more exciting and satisfying to have cars exploding, so they do.
So I'm not saying you can't have fun explosions in your fiction, nor that you can't have characters make bad engineering or technical design decisions. (That too happens all the time in the real world.) I'm just saying that if your entire plot hinges on a bunch of smart people never having thought of something basic, it might be worth taking another look at the plausibility of your plot.
(Written in June of 2010; somehow didn't get around to publishing 'til now.)