So why are alcohol and tobacco legal while marijuana is illegal?

In a sense, this has virtually happened in California. IIRC, the misdemeanor presumptive bail schedule specifies a bail of $75 for possession of less than an ounce of pot. That actually is almost within the range for parking tickets. However, there’s the added consideration of court costs, and possible fines. I don’t know if people actually get arrested and booked for misdemeanor possession, or if they only get a “ticket”.

Generally, just a 'ticket".

Stoned is stoned. You are not going to convince me someone stoned on pot is not dangerous and it’s just nonsense to think driving improves with dosage. Maybe lower levels of pot are “better” than lower levels of alcohol when driving. Couldn’t say. I can say that large levels of pot are just as bad as large levels of alcohol.

OK. I have tracked down the June, 1980 article in Car and Driver (not Road and Track), “Puff, the Dangerous Driver.” All I can say is, in the words of Inigo Montoya, “I do not think it means what you think it means.”

It was not a scientific test.
They did not match drinking vs toking.
The tokers did not get better the more they toked up.
Now, they did indicate that there appeared to be no appreciable loss of perception or control as they got high. There did not appear to be a serious impairment to their reflexes as there would have been with alcohol.

On the the hand, they became less and less concerned with the outcome of the test. When they concentrated on the specific maneuvers needed to complete the course, they were able to carry them out. However, they paid little attention to anything outside those maneuvers and displayed a reckless disregard for the situation while they were driving. On the last pass, one driver charged through at 50 mph (when they had been keeping to around 35 mph for most of the test) knocking over a few cones on one pass and making it through without a knockdown on the next. However, he was successful only in carryingout the specific task and the authors noted that he would have been oblivious to other traffic in a situation outside the test environment. In other words, he would have been quite capable of keeping his car in the center of the lane while he was driving, but it appeared that he would have maintained that “good” driving even if something unexpected had happened to which he should have reacted.

That’s the key weakness in such tests, as I see it. If one drives a test course several times (or does any task, really), performance will naturally improve as familiarity does. A fairer test would be to alter the course after each round of smoking/drinking, so the subjects can’t go by memory.

There was a controlled study published last year

Psychoactive substance use and the risk of motor vehicle accidents (PDF)

The odd ratios for various drugs

Amphetamines → 2.10
Benzodiazepines → 5.05
Cannabis → 1.22
Cocaine → 2.04
Opiates → 2.35

Alcohol
BAC < 0.5 → 1.00 (base reference)
BAC 0.5-0.79 → 5.46
BAC 0.8 >= → 15.5

Multiple drugs vs. no drug → 6.05
Drug-alcohol combo vs. no drug → 112.22

It should be noted that my specific point was merely to confirm (or not) the claim that perfomance improved under the influence of pot.

I am willing to believe that pot has a less deleterious effect than alcohol (with all the usual caveats that we cannot reliably measure how much THC (or supplemental agents) are found in a single hit or joint so that we have nothing similar to a BAC to compare).

I was just curious regarding the “improvement” claim.

Good to know.