>My hypothesis about this for a long time has been stimulants help everyone be more productive (with some tradeoffs)
Of course they do. They're stimulants, that's what they do. Some people just need them to be closer to normal, or whatever's considered normal in post-Industrial society. Modafinil promotes wakefulness in everyone, not just narcolpetics. Anxiolytics calm down everyone, not just the anxious, and psilocybin makes everyone feel euphoric, not just the depressed. It would be weird if stimulants only had an effect of ADHD patients.
> and ADHD is kind of a weakly differentiated diagnosis that could apply to most people.
I don't think we really understand it yet, but it's not something most people have. As the article mentions, people ADHD have a higher rate of transportation accidents, lower life expectancy, higher crime rates, higher addiction rates, etc. The differences show up in brain scans, performance tests, genetic biomarkers, heritability/twin studies, etc. Whether you think of it as a disability, or brain type, or whatever - ADHD is something real.
> Probably something like this was lost when people stopped smoking, obviously beneficial for health - but a huge amount of the public was taking stimulants regularly via nicotine until relatively recently.
Yes, and this is possibly why 35-55% of adults with ADHD smoke today, compared to 19% of the population. Studies have shown that nicotine is helpful for everyone but particularly helpful for those with ADHD. Nicotine-derived formulations are still being explored.