The one grammar rule that really counts

I once read, long ago, that the one grammar rule that really counts for anyone writing (or translating into English) and wanting to convince, persuade or cajole (as opposed to just showing off one's possibly superior knowledge of grammar is this: Never use any expression that is likely to take the reader's attention off what you are trying to say.

In Give some ground to the pedants in the ‘I/me’ battle, subtitled You can argue with many grammar rules, but there are sound reasons for sticking to them, FT columnist Michael Skapinker comes to the same conclusion (my bold).

So Mr Wulf’s students could write “he is taller than me” or they could write “he is taller than I am”. Most people find either acceptable, while they regard “he is taller than I” as strange and stilted.
Having said that, I usually opt for the spelt-out “he is taller than I am” form. Why? Because I know I will not upset those who insist that “than” has to be followed by the nominative. They will concentrate on my argument rather than my grammar.
It is the same with infinitives. There is nothing wrong with splitting them. Those who insist it is bad are the same tiny group who are trying to squeeze English into Latin.
But unless not splitting the infinitive results in an awkward sentence, I leave it unsplit. So do writers on The Economist, whose style guide says that while the ban on split infinitives is pointless, “to see it broken is so annoying to so many people that you should observe it”.

then adds:

The problem is that most objectors do not complain. They simply assume you do not know the rules. ...

Most people ... are not grammar historians but people with their own linguistic shibboleths.

I would not give a job to anyone who confused “its” and “it’s”, even if Kamm points out that “it’s was a possessive until the 19th century”.
You never know when you are going to cross what we can call someone’s “stickling point”. (I had hoped to claim authorship of this term but Ian Mayes of The Guardian used it in 2000.)
You can argue with many supposed grammar rules, but it is best to know what they are and to break them only when you are sure it is not going to do you any harm.

On the basis of this reasoning I also avoid using 'whom' wherever possible, not because I don't know when and how to use it, but because there is always a serious risk that some of my readers will think more about my grammar than my (or my client's) message.

26 April: Further support for this line of thinking comes from no less an authority than Fowler’s Modern English Usage. The review, by Michael Quinion, of the fourth edition quotes author Jeremy Butterfield concerning the use of the word literally (my bold):

{Butterfield] ... concurs with the recent decision of the Oxford English Dictionary to recognise the figurative use of literally to mean “figuratively”, a sense that goes back at least to Dickens. But he cautions, after nearly two pages of discussion: “Knowing that your readers may have the screaming abdabs (dated British slang for ‘have a fit’) if they read literally prefacing a metaphor ... you might want to avoid using it altogether.”

So, when it doubt as to what your reader might think or how they might react, abstain.