I didn't imply that repeated doses are unpredictable.

Regular repeated dosages can be predictable as long as the patient remains stable with regards to infection/health and other drugs they ingest. Initial dosages, dosage increases and returning to a previously safe repeated-dose after cessation are not at all predictable.

This is why I mentioned titration in a previous comment: subjective response to drugs and subjective tolerance cannot be accurately predicted or measured.

This is also why drugs like Subutex are titrated when beginning suboxone treatment: a doctor cannot accurately measure tolerance to determine how much of the drug is needed to keep a person out of withdrawals.

> you absolutely can approximate how long tolerance lasts

Naloxone or naltrexone induced withdrawal rapidly diminishes tolerance faster than it would naturally.

This suggests that there's a function that can accurately output a safe and effective dose of opioids based on the variables of previous dosage and time since cessation. I'm arguing that there is a missing variable that's unable to be measured accurately: each person's subjective tolerance. That subjective tolerance depends on a multitude of factors that cannot be determined clinically.