The article mentions some of the shortfalls of the scoring system used, but I’m going to repeat an important one in the comments since not everyone reads the wh

The article mentions some of the shortfalls of the scoring system used, but I’m going to repeat an important one in the comments since not everyone reads the whole article. The scoring system used makes it look like alcohol is about as harmful as crack cocaine or heroin. That’s because they don’t normalize every harm per capita (some are, some aren't). What the scores show is that the total harm of nearly every adult in the UK drinking alcohol is about the same as the much smaller groups of people using crack or heroin. If an individual was choosing a drug to use, alcohol is by far less harmful per user. I’m not totally disagreeing with the article; it makes good points but it’s a little unfair to compare legal drugs vs illegal drugs on the basis of their total harm. If you want a fair comparison you should come up with a way to compare the harm that crack or heroin would do if they were legalized- I realize this is hard to do because you have to estimate how many more people would use heroin and also estimate whether the supply chain would become safer. My opinion is that an analysis assuming legality would show that crack and heroin would still be pretty harmful but that many recreational drugs like ecstasy and marijuana would be less harmful than alcohol if legalized. Quick edit: the comparison that I’m talking about is what causal inference in statistics tries to do, but this particular example is an unsolved problem because of issues with transportability and generalizability that arise because we don’t have any data where hard drugs are legal in a population that is comparable to the UK.