SEC Basketball Power Poll: Week 2

… again, a W/L record right now is just a snapshot in time. It is also a poor indicator, absent the context of opponent quality, of how a team should actually be rated. Keep in mind that UK has only one win that is in or near the top 25 at the time it happened, and that was against Tennessee. It is wise to avoid reading too much into that victory. My system uses a lot of objective information to derive the ranking, not just a W/L record or how a team “looks” to an observer.

I don’t really count the SEC W/L independently yet. I consider it a relatively tenuous measurement when it comes to actual team quality at this early stage. It becomes very important later in the game, though, when the games start to run out. I’ll begin factoring it in about halfway through the season. My polling is designed to respond to the question, “Who is the best team in the SEC,” not “Who is the best team in the SEC today.”

My system attempts to assign more of a time-independent value, more predictive than “as of today.” That’s why it changes relatively slowly. It is also worthy of note that it takes not only the SEC performance into consideration, but the entire body of work. This is essentially a subset (although slightly displaced in time) of the Blogpoll ballot.

With respect to the “beauty contests,” I refer to the AP and Coach’s poll. Most of them vote based purely on how they feel about a team, rather than using any objective standard. Hence the term, “beauty contest” — those polls are mostly who the press and coaches think are pretty right now. :-)

But those polls aren’t worthless. They provide a component of “heat” to the system I use, and when a team is getting love from the coach’s and AP poll, it shows up in my results much faster than it would in a purely stats-based system like Pomeroy, BBState or even a bastardized measure like the RPI. That’s the biggest reason why UT is ranked above UK right now.

A final note: The Tennessee victory, in isolation, means relatively little. It has to be supported by continuing good play, and since the Georgia game was outside the window of submission, what seems so obvious now was not so obvious on Thursday. What I mean by that is that a conservative system requires teams to “validate” big victories before jumping them up over other teams.

… when my poll ballot was submitted? I’m thinking not — at least, I’m hoping not.

… nor ignorant, and neither is my spreadsheet.

I’m thinking you are either uninformed or so unjustifiably certain of your position you can’t consider the possibility that you may be mistaken. In my opinion, you are manifestly wrong, the best team in the SEC right now is not “clearly” anybody. Many people think it is UK, but MSU is arguably playing just as well, and most people think Florida is playing better than either one. For evidence, see the AP and Coach’s poll. I am absolutely positive based on this discussion that you are no more qualified than those voters are to propose who is best by any measuring stick you care to apply.

Perhaps you should reconsider the use of such words as “clearly”, “ignorant”, “uninformed” and especially, “pathetic.” It also might help to read Pomeroy’s discussion of how his rankings work, something that you evidently have not done.

Based on the facts in evidence, it seems likely that the object of your derision is misplaced. Some introspection may suggest another target — namely your own failure to respect the opinions of others, especially when theirs are at least as defensible and arguably more so than your own.

You know the rest.

Pomeroy’s system does include MOV in his calculations. BBState does as well, I think, although their computations are more complex.