[Caleb Sykes]

Originally Posted by Lil' Lavery

Even if we suppose that there is currently a correlation between OPR and team retention (which I have some dispute with, at least until more data is released), that doesn't mean that correlation will carry forwards if you take steps to increase OPR. That is to say, the concept of eliminating bag day to raise OPR of teams doesn't mean that fewer teams will fold since we have yet to establish a causal relationship between OPR and team attrition. If teams are folding from a variety of other stressors (under funding/under mentorship/no school support/burn out/etc), raising their OPR will not save those teams from folding.

To put it another way, basing actions purely on the correlation is treating a symptom, not the disease.

I completely agree. We can't know that the correlation between OPR and team retention will hold if/when we change fundamental aspects of FRC. I thought that I had made that clear.

Originally Posted by Caleb Sykes

We can disagree about how specific policies will impact OPR distributions, team retention rates, or the correlation between the two, but at the present time, OPR does indeed seem to be a reasonable proxy for a given FRC team's success.

I was responding to this statement:

Originally Posted by gblake

Scoring points in matches is simply not the correct proxy/metric to use to gauge the success of FRC teams.

I am fully 100% aware that correlation is not causation. However, gblake is going a step further by essentially denying that any correlation even exists in the present. This correlation could very well be diminished if we eliminated bag day, or it might increase. We don't know, what we do know though, is that OPR and team retention seem to be relatively strongly correlated in the present. If gblake wants to deny the usefulness of OPR as a metric for team success, that is fine, but it is fallacious to say that OPR is a poor metric for team success in the present just because it might not be correlated with team retention in the future.