Depends if you actually ment what your wrote. I think his point is that if you did, you would have chosen a different wording from a psychological point of view.

I'd argue that it worked perfectly, actually.

What were the poster's true motivations? Not what they stated, but what they truly meant.

What this kind of analysis needs is a blind test. Ask three groups to rate text independently in the same metrics - 1) General public. 2) Psychologists 3) Algorithm.

The results should be interesting. However, if the authors claim subliminal effects, then I do not how that could be tested.