Thing about everything being connected is “to what degree?” We are all related and us both being white you may conclude we are 4th or 5th cousins but actually we’re only 10th cousins at the closest point. However we’re also 12th cousins on 3 other points so we’re more similar than just 10th cousins. I think you have a problem with seeing more substantial or greater degree connections than are there. Sometimes that still works and sometimes it’s better to not let those vague connections alter the equation by overstating their importance. Balance is very important.

Reply to this note

Please Login to reply.

Discussion

To minimize the error is all we can do with the data we have. The PCA algorithm is an example of this, and ironically mentions Hooke's Law.

https://stats.stackexchange.com/questions/2691/making-sense-of-principal-component-analysis-eigenvectors-eigenvalues#140579