(I haven't linked to the actual blog because I don't want to seem like I'm criticising them personally.)
You might think that people who teach language know linguistics. Most people seem to think that linguistics *is* teaching languages. In fact, lots of linguistics graduates go on to do language teaching, but they are far outnumbered by language teachers who don't have a linguistics background. TESOL MA programmes frequently have a core linguistics component, but people who've chosen to do TESOL but not linguistics are probably not that interested in linguistics anyway.
The reason I mention this is that the question above is taught on any first-year linguistics course as the absolute go-to example of phonologically-conditioned allomorphy in English. Typically, we point out that native English speakers do this subconsciously and consistently, and it's an example of how there are rules and we can discover them by careful investigation, as well as to teach the theoretical concept.
Of course, speakers of other languages with different rules will not do this subconsciously, because the rules of their language might not automatically force the same result as English. They'll need the rule taught to them. The fact that someone who (presumably) is teaching English doesn't know how to explain this simple fact is quite shocking to me, but I doubt the asker is alone. A few basic linguistics classes would make life as an English teacher so much easier, I can't even imagine why it wouldn't be standard.