Most of the story is usually that you are expecting them to encounter ideas that they don't know how to handle, ideas that are new and scary and difficult. They might be great writers with things that they are comfortable with, but once you challenge them? The whole thing becomes a hot mess. This doesn't mean that they are bad writers--it means that they are out of their intellectual depth. If you teach them the ideas, then the writing can catch up. But the writing has to catch up to their thinking--the writing isn't a stand-alone thing.
Sure, there are certainly students who lack the technical skills with writing to formulate grammatically correct sentences and coherent paragraphs, but the root cause of most bad writing is either 1) people don't know what to say, which is where FLG guesses most poor student writing falls, or 2) people are trying to obscure their true meaning, as Orwell explained so eloquently.
It took FLG a while, but he has an almost full-proof strategy for dealing with not knowing what the heck to say. He might have mentioned it before.
Plan A, draw a distinction / compare and contrast. Given that FLG benefited from a Jesuit education, no distinction is too small or obscure.
There are lots of options. Some of FLG's favorites include: long-term versus short-term, ancient versus modern, rational versus empirical, utilitarian versus deontological, sacred versus profane, and pirate versus privateer.
Usually, step one suffices to give FLG something to center his argument around. If he is still at a loss, then he proceeds to Plan B.
Plan B, comment on what somebody else said on the topic. FLG's first GoTo on this is Aristotle. Who is going to be against referencing Aristotle? And he talked about so many things -- tragedy, politics, economics, ethics, etc. Tons of material. In b-school, however, his frequent GoTo, for obvious reasons, was Adam Smith.
Plan C, combine Plans A & C. Find two thinkers. Elucidate and expand upon a distinction between them. Plato versus Aristotle happens to be FLG's favorite.
Plan A works about 95% of the time. Plan B or Plan C work the other 5%.