Not sure I understand what you're saying.

* THC threshold for a first-time user to register as "completely stoned": $foo

* THC level of someone who smoked the maximum amount they could 30 days ago, and hasn't smoked or eaten an edible since: $bar

Are you saying that $bar >= $foo?

The degree to which any given dose of THC affects an individual varies so wildly as to be completely useless as a baseline. To guarantee that 100% of debilitated users are detected, the threshold would have to be so low as to also detect a heavy user who smoked a joint two days ago.

Your argument is built on the supposition that there's a hard lower boundary on debilitating dosage. There is no such magic number, or it is nearly indistinguishable from zero.

It's sort of shocking. Compared to alcohol, where someone might plausibly get as drunk off of 1 drink as someone else might from 5, with THC the plausible ratio is more like 1:50 between naive and heavy users (IME).

That's the way it was in my state until it was overturned on appeal, and they only did that due to an accident of the way weed was legalized.

That’s what Germany does (roughly)

>"Phosphatidylethanol (PEth) is a direct marker, formed only in the presence of ethanol, and can detect heavy or binge drinking for up to 4 weeks after consumption."

We should simultaneously use this marker to prove drunk driving instead of the clearly outdated direct measurements.

Totally agree.