I'm not sure if I remember correctly but I think I've read ketamine makes neurons groe new dendrites in petri dish. I have no idea how this effect could happen in the brain without making a patient trip.
I interpret you as saying that unless there is a "trip" there is only a placebo effect left on the table so to speak. I think this is somewhat refuted by the fact that the standard ketamine infusion protocol (that does not normaly induce "trips") has an effect size that dwarfs the effect size of placebo (say 0.2 - 0.4). Some studies have shown ketamine approaching ECT-effect sizes. This makes me sound like a ketamine nut, which I am not. I am a bit of an ECT nut though. I have just not excluded the fact that the phenomenology of certain treatments actually mediate the treatment effects in a non-placebo way.
At least it is what authors of this study consider.
Also consider the difference between high and low doses of alcohol: just because having 1 beer won't make you blind drunk doesn't mean that the effect is a placebo.