The college football season is finally over. Though the BCS championship game between LSU and Alabama wasn't very exciting, I thought the contestants were the best teams and deserved to be matched up in the final. The SEC, home to both schools, appears to have distanced itself from its BCS football conference peers over the last 10 years.
For Alabama, it's the second BCS title since 1999 and the 14th football championship overall. It's also the third BCS title for peripatetic Alabama coach Nick Saban – two at Alabama and one at LSU. If Saban remains at Alabama for a few more years, don't be surprised if the faithful start mentioning his name in the same breath as the god-like Paul “Bear” Bryant.
Alabamans are experts at both college football and analytical designs, it seems. Many years ago, a Crimson Tide fan remarked of Bryant's coaching prowess: “He'll take his'n and beat your'n, then take your'n and beat his'n.” In philosophical/statistical terms, “his'n against your'n” is the observed factual, while “your'n against his'n” is the unobserved counterfactual. Methodological nirvana compares factual against counterfactual outcomes to determine causality. Brilliant proposal!
Alas, the post-season outcomes weren't so good for my beloved Wisconsin Badgers and their eleven Big 10 brethren. It's been a rough last few years of bowl games for the Big 10, conference teams losing a lot more than they've won. Indeed, The Wall Street Journal has grown inpatient with their performance, recently publishing an unflattering critique, “The Big 10 has Seen Better Days.” Maybe former President Clinton aid, LSU alum and SEC football fanatic James Carville had it right in a hoot of a video denouncing Big 10 football several years ago.
Wisconsin lost a very close and exciting Rose Bowl game for the second year in a row, this time to Oregon. As was the case last year against TCU, the Badgers moved the ball up and down the field at will. At crunch time, though, Oregon, like TCU, made the big plays. And beat Wisconsin with superior speed.
The Badgers are still chasing Oregon freshman De’Anthony Thomas, who had two carries, one for 91 yards, the other 64 – and both for touchdowns. In fact if I were Thomas, I'd complain that reaching the goal line artificially compressed my rushing average. I guess you could say his TD runs were “censored” observations, akin to cases in “time to failure” studies that don't fail, because he never was tackled. Maybe his true average would have been 99 yards. And maybe Wisconsin should be “censored” for their porous defense.
If the Big 10's currently no better than the third or fourth best football conference, it still has elite academic and marketing brands. The rousing success of its cable TV Big Ten Network has spawned a host of imitators and led to conference gerrymandering that's now upheaving big-time college sports.
As I channel-surfed to the BTN the other night, I came across a promotion for conference schools, showing highlights of sports accomplishments by member institutions. The clip of Nebraska was one of their many football successes. The problem is that Nebraska has only been in the Big 10 for a year, and its football program was less than noteworthy this season. The footage shown was actually from Nebraska's tenure in the rival Big 12 conference years ago.
BI analysts might recognize this as a changing dimension problem. The dimension in this case is, of course, conference affiliation, while the attribute is school. How is such change handled in BI? The two most common techniques, as articulated by data warehousing guru Ralph Kimball, are: 1) simply overwriting the current dimension, and 2) preserving the past with the addition of date and time stamps. For the commercial I saw, the Big 10 appears to have chosen the former, which would attribute all of Nebraska's past to the Big 10. My guess is that an experienced BI practitioner would go for the more flexible date and time stamp solution, preserving Nebraska's Big 12 and Big 10 legacies.
In an era of ever-less government financial support of higher education, it's comforting to see increasing collaboration between our great universities and the private sector. And that marriage doesn't get much better than the one between Yale University and Las Vegas sports oddsmakers. Featured in the Wall Street Journal, the Massey-Peabody ratings of NFL teams are a measure of “how much better or worse a team would fare against an average team on a neutral field.” Factoring in a home team advantage of 2.42 points, the MP scores be used to “compare” the strength of teams for wagering.
At the end of the regular season, New Orleans led in MP, followed by Pittsburgh, Green Bay and New England. None of the top three made it to the conference championships. And the Steelers lost in the first round to Denver. I bet the MP models don't include a Tim Tebow factor!