Maybe the confusion is yours? You think the category is "drug" but it's really more like "taboo drug."

> I also find it odd that we treat substances that the data clearly indicates are less harmful than alcohol as though they were worse. We have alcohol staring us in the face as a counterexample to the claim that such laws are necessary. I think that avoidance of this observation can largely explain the apparent widespread unwillingness to refer to alcohol as a drug.

I think you missed a pretty key point: "shared cultural knowledge about how to manage the substance, including rituals for use (this is the big one)." In the West, that exists for alcohol, but not really for anything else. People know how it works and what it does, can recognize its use, have practices for its safe use that work for (most) people (e.g. drink in certain social settings), and are at least somewhat familiar with usage failure modes. A "less harmful" thing that you don't know how to use safely can be more harmful than a "more harmful" thing you know how to use safely. None of this is "data driven," nor should it be.