Hacker News Reader

Isn't that the very definition of something being predictable?

That there is a pattern which predicts a certain outcome?

For example a certain simplified weather pattern (e.g. "if it rains for three consecutive days 9/10 times it rains a fourth day") makes it possible to predict a certain weather ("it has rained the last three days so there will be rain tomorrow as well")

Obviously that is not a prediction with absolute certainty but nonetheless a prediction.

It's a semantic argument over the meaning of "pattern". Your example is mathematical, the meaning in the linked article is more along the lines of "statistical model".

I mean, to provide a counterexample: we have about as good a model for beta decay of H3 as we are ever going to get[1], but we still can't tell you when your geiger counter is going to click.

[1] Pause for some physiscist to jump in and tell me this is totally not true and there remain deep theoretical holes here.

Unless you have some unique definition of what a pattern is: yes, it does.

David Hume's Problem of Induction strikes again! But that's not how science works.