The article “Earthquakes, Patterns and Predictions” appeared in Wired Magazine. It points out that one attribute of being human is seeing patterns, even ones that don’t actually exist. This has something to do with the fact that it paid our ancestors to see patterns that even if some of them didn’t pan out.
You can use statistical analysis and scientific studies as ways of checking whether patterns are real or not, separating the wheat from the chafe. This analysis for earthquake shows that the distribution of earthquakes is random and not affected by such things as tides or weather.
This is an interesting question for the Cascadia fault, since there does appear to be clusters of earthquakes grouped together with shorter intervals between them and then longer periods between clusters. Real pattern, or just the luck of randomness? This is a topic of research for now, awaiting either more data or identification of the mechanism that would account for the clustering.
It would be nice to know if there is clustering. It would mean either that our expectations of when one the next one would occur either is more likely and less likely. If more likely, we may want to step up the pace of getting ready. If less likely we will have plenty of time. For now, we assume a random distribution and hope that the science is able to tell before the next Cascadia earthquake happens.
If it happens to occur on a hot still day on a full moon in the summer, well that’s just luck.