File under "Other things that happened last week."
Topics I have blogged about over the last 9 days:
- Tom Izzo and the Cleveland Cavaliers (10 posts)
- The World Cup (1 post)
If Tom Izzo can focus on other things while this process continues to (not) play out, so can I.
So let's talk about that other minor development in the world of college athletics: There's a new kid on the block. Welcome to the Nebraska Cornhuskers. Jim Delany reeled in one of the three schools with a big enough football brand to make one-and-done expansion worthwhile. And that looks like it may be the final product here. With Texas staying put in the Big 12(Ten), the Big Ten would really have to go all out to break up the Big East and force Notre Dame's hand to create an economically-productive 14- or 16-team scenario.
An even dozen appears to be the state of the near-future world. So, on to the implementation stuff. What about divisions?
Here are the criteria Jim Delaney has laid out for constructing divisions:
1. Competitive fairness 2. Rivalries 3. Geography.
Doc Saturday interprets thusly:
1. Splitting up Ohio State, Penn State and Michigan, the three programs responsible for eight straight conference championships/automatic BCS berths and four of seven at-large BCS bids since 2002; and
2. Preserving the prominence of the Ohio State-Michigan game in the regular-season finale.
Mathematically, those things add up to "Move Penn State, the easternmost school in the conference, by a wide margin, to the western division."
But is that really the most efficient arrangement? Let's step back and build the divisions using Delany's three criteria, but applied in reverse order. We'll then see how the resulting divisional alignment stacks up using Delany's order (I'll confess to having already played this out in my head; I think it's gonna work).
So, East/West divisions based on simple geography:
Adding a team on the western front, rather than the eastern front, makes the geographic split very simple: The Illinois-Indiana border is your (vertical) Mason-Dixon line.
How about maintaining natural rivalries? Well the current protected rivalries all occur within three subsets of the existing 11 Big Ten teams:
- Iowa/Minnesota/Wisconsin (with Nebraska becoming a natural fourth in this block)
- Michigan/Michigan State/Ohio State/Penn State
Any divisional alignment is going to break up one of those blocks of four teams into two pairs. It might as well be the block of four Illinois/Indiana teams. The Illinois-Indiana rivalry is a bit of a big deal, I think, but of course they're also the two weakest football programs in the league right now. They'll live. The in-state rivalries in both states would be maintained.
OK, on to competitive fairness. The table below shows annual Sagarin rankings (to allow comparisons with Nebraska) for each Big Ten team over the last decade, along with three different measures of a mean ranking: simple 5-year average, simple 10-year average, 10-year average with the most recent year weighted at twice the level of the first year in the decade (with proportional weights in between).
|Team||2000||2001||2002||2003||2004||2005||2006||2007||2008||2009||5-Yr Avg||10-Yr Avg||Wtd 10-Yr|
Proposed East Division teams are in bold. Looks pretty balanced visually, eh? Note that under none of the three "average" columns, do OSU, PSU, and U-M make up the top three. Doc Saturday notes that those three teams have accounted for the last eight conference championships, but that ignores the facts that (1) Iowa has been right in the mix in several years and (2) Nebraska would have been, too, had they, you know, been a part of the conference. It also ignores the issues of (1) Michigan not being very good since Lloyd Carr retired and being far from a rock-solid bet to return to national prominence in the next few years and (2) the big question mark that looms over Penn State's post-JoePa future.
Back to the numbers. Under all three measures of average ranking, the 12 teams can be cleanly grouped into four tiers:
- Ohio State, by itself.
- Iowa, Michigan, Nebraska, Penn State, Wisconsin
- Michigan State, Minnesota, Northwestern, Purdue
- Illinois, Indiana
(Single exception: Penn State moves into Ohio State's tier if you only look at the last 5 years.)
A simple East-West alignment splits the six teams in the first two tiers into two groups of three and splits each of the other two tiers neatly in half, as well. Tough to beat that. Doc Saturday attempts it by swapping out Wisconsin for Penn State. That's not really an improvement based on the data above--and if we're not using recent historical performance data to measure "competitive fairness," what is it we're using? Meanwhile, you'd be splitting those team off from both their current protected rivalries.
Ladies and gentlemen of the jury, I rest my case. Anything you do to place more emphasis on competitive fairness leads to a marginal improvement, while creating significant disruptions in the other two categories of consideration.
(Now let me offer potential rebuttals to my own argument, thinking from purely a Spartan perspective. Suppose Michigan does return to glory again in the near future and Penn State manages to maintain its current level of success under whoever succeeds Paterno. MSU could be related to permanent second-[intra]division status. But, you know what, until MSU can consistently compete with Michigan and Penn State, are we really going to have any real football success, anyway?
Also: This thing is problematic from a basketball perspective, assuming you're going to use to the same divisions in both revenue-producing sports. Three of the four teams that have been consistently in the hunt for the Big Ten title [MSU/Purdue/OSU/Wisconsin] in recent years are in the East. Illinois provides some counter weight in the West, but Indiana is the best bet of the current lower division teams to rise to contention and they're in the East. Further analysis is warranted here. Big Ten regular season championships aren't quite as big a deal as football championships, given the existence of an actual postseason championship event. Still, I'd really hate to give Bo Ryan a significant advantage before the first conference game is even played each season.)
P.S. Yes, you need divisions. Otherwise, you end up with a situation where a superior team can be excluded from the championship game because two teams had substantially easier schedules. With divisions, you've at least equalized 62.5%+ of any team's schedule, including a head-to-head match-up, relative to any other team in its division. Plus permanent divisions are preferable in terms of maintaining/establishing/building annual rivalry games.