Page 1 of 1

The Base-Rate Problem in Assessment of Results

Posted: Sun Dec 03, 2017 11:19 am
by R. Eugene Laughlin
When a magician performs an operation for a specific outcome, then the target outcome occurs, it's natural (human nature) to assume that the magic performed caused the result. On the face of it and for most people, that's consistent with good reasoning:

1. do X for Y
2. get Y
3. X caused Y.

It also happens that from time to time the target outcome is not achieved, and the most common explanation for magicians is that they didn't do the right kind of magic, or there was something wrong with the way they did the magic. That too is consistent with natural reasoning:

1. do X for Y
2. don't get Y
3. there was something wrong with X.

The above is both common and reasonable. Less common in contemporary occult studies discussion, however, is an acknowledgement and consideration that everyone sets goals, and everyone achieves some of them. If that much is acknowledged, then the reasonable companion idea is that if magic operations represent something extra to what non-magicians do, the success rate among magicians should be higher than that of the Average Josephine, when the operations are done well. If we assume a good magical operation and that magicians are as likely as non-magicians to do non-magical goal-relevant things, this simple inequality should hold:

Set Goal + Regular Efforts + Magic > Set Goal + Regular Efforts

It usually goes without saying for magicians: magic should add something extra to push the success rate higher.
While the above represents a reasonable theory, there’s a problem with validating it: there’s no reasonable way to estimate the base-rate of success for any given goal among non-magicians. My experience in forums like this suggests that most people who practice magic never consider the base-rate of success in the general population when thinking about their own results, so I’m asking the readership to consider it now, and respond to the following statement:

Because of the Base-Rate Problem, making one-to-one, working-to-outcome assessments of one’s magical results is inherently fallacious.

What do you think?