The mind sees what it choses to see.
If you plot cornell's rankings against feet of snowfall in Ithaca since 1990 I am sure there are people who will observe a correlation. However none of us will contest the claim that the correlation is mere coincidence.
Similarly while plotting change in cornell rankings to the change in index values(DJIA, NYSE etc) , You need to avoid data mining, articulate a clear hypothesis after plotting your graphs and subject your conclusions to hypothesis testing.
I applaud the effort but none of these conclusions are reliable in the slightest.