It's a fair line of argument. I think where one might draw the line will depend in part on how expansive one views the concept of "feints". But did you mean mechanically supported in the rules, or the fiction surrounding "feints"?
In the fiction, it's easy for me. I attend a fencing school taught by an older man who was taught by the Italians in New Jersey. One of his maxims is that, "a feint is anything that causes your opponent to react in a predictable manner." Now it is true that your feint doesn't force them to do that. They can not block your line of attack (in which case, you simply hit them) or do any number of crazy things. However the idea is, if they don't do one of two or three limited options, you've got them cold. (And if they've got any sense, they know you were leading them into an option, and they'll pick one of those, knowing you are expecting it, and then react to what you had planned. But that is getting afield.)
Despite all this, I have seen over and over fencers win touches by pushing their opponent to the back of the strip, suddenly feinting, and having the opponent retreat off the strip, and thus give up a touch. I even won a tied preliminary bout 5-4 once, doing that, against a guy that was better than me. And who cleaned my clock later that day. He felt pretty sheepish letting me get away with it once, and wasn't going to let up after that.
Combat is moving, and off a strip, fighting for your life instead of touches, all the experienced writers I have seen have agreed that this is magnified, not diminished compared to what is essentially highly-regulated sparring.
So for me, the only potential sticking point is the timing of the force. In a simulation with any fidelity, feints resulting in opponents moving--even when this was not necessarily their best course, will happen on an infrequent but recurring basis.