The reason quitting is easier than moderating

Two people reduce their drinking. One stops entirely; the other allows two glasses at weekends. A year later the abstainer has held the rule and the moderator has drifted. The moderator’s target was easier at every individual moment and harder across all of them.

What everyone sees

Moderation appears obviously more achievable, since it demands less at each point. That intuition is correct about any single occasion and wrong about the aggregate, because it counts the difficulty of each instance without counting the number of instances at which a judgement is required.

What is actually happening

Ainslie described bright-line rules as a commitment device precisely because they eliminate case-by-case evaluation, which is where preference reversal occurs. Ainslie and Herrnstein’s work on hyperbolic discounting shows that the relative value of an immediate reward rises sharply as it comes near, so a rule evaluated in the moment is evaluated under the least favourable conditions. Baumeister and colleagues reported that repeated acts of self-regulation degrade subsequent performance, though the strength of that depletion account has been debated in later replication work. The structural point survives the debate: a limit requires a judgement per occasion, and each judgement can go the other way.

Why it stays hidden

The arithmetic hides because difficulty is imagined one instance at a time. Asked which is harder, a person pictures a single evening, and in a single evening the limit is plainly easier. They do not picture three hundred evenings, each carrying its own decision, and the compounding of small probabilities of failure is not something intuition renders.