Friday, January 28, 2011

2010 - 2011: Final Rankings -- RBA

Full rankings after the jump.

2010 - 2011: Final Top 25 -- RBA

Better late than never, right?  Presenting the final top 25 for Regression-Based Analysis:

Rank+/-TeamWinPctSoSOff PtsDef PtsPace
004--Boise St.0.98290.460826.68.684.6
006--Ohio St.0.95800.529126.18.179.8
009--Virginia Tech0.93160.530124.98.479.1
015--Oklahoma St.0.91890.548124.412.785.9
017--Florida St.0.90180.553217.910.683.2
018--West Virginia0.86730.537217.59.582.8
019--South Carolina0.85710.546324.112.478.9
022--Mississippi St.0.82570.559220.510.981.1
025--Arizona St.0.81420.526020.09.287.0

RBA agrees with Justin that Auburn is not a particularly memorable national champion, placing them 7th overall.  The same reasons apply:  close games and flaky defense.  RBA declares a split title between Stanford and TCU.  The two teams would likely play to a tie on a neutral field.

The "conference title" is effectively wrapped up by the SEC, placing a ridiculous eight teams in the top 25.  The biggest WTF comes from Georgia's inclusion.  This is a direct result of their flakiness.  When the Bulldogs show up, they are a powerful team.  When they don't, well... you saw what happened against UCF.

Wednesday, January 26, 2011

2010 - 2011: Final Rankings -- TFG

Biggest jumps: Tulsa (+9, 57 to 48); North Carolina State (+7, 33 to 26); Mississippi State (+7, 39 to 32); Syracuse (+7, 83 to 76).

Biggest drops: Michigan State (-8, 35 to 43); Navy (-6, 38 to 44); Connecticut (-6, 49 to 55); Hawaii (-6, 52 to 58).

Full rankings after the jump

2010 - 2011: Final Top 25 -- TFG

The season is over, the bowl games have been played, Auburn defeated Oregon 22-19, and your TFG National Champion is ...

Rank +/- Team WinPct SoS Off Pts Def Pts Pace
001-- TCU 0.9342 0.4520 28.0 9.1 80.7
002-- Boise St. 0.9333 0.4287 28.6 9.3 82.8
003+1 Alabama 0.9251 0.6464 27.8 9.5 77.7
004-1 Ohio St. 0.9103 0.5275 25.4 9.5 80.5
005+1 Oregon 0.8690 0.5691 26.0 11.8 92.3
006+2 Stanford 0.8616 0.5987 29.6 13.7 80.9
007-- Oklahoma 0.8530 0.6022 23.4 11.2 91.8
008-3 Virginia Tech 0.8484 0.5858 27.7 13.3 78.7
009-- LSU 0.8368 0.6407 24.4 12.2 78.3
010-- Auburn 0.8330 0.6619 29.1 14.9 83.4
011+2 Florida 0.8075 0.6583 23.9 13.2 81.7
012+3 Arkansas 0.7995 0.6583 27.2 15.2 84.2
013-2 Iowa 0.7977 0.5639 21.1 11.7 80.8
014-- Wisconsin 0.7844 0.5468 28.8 16.7 78.9
015-3 Nebraska 0.7802 0.5073 22.2 13.0 80.1
016+2 Florida St. 0.7670 0.6519 25.0 15.2 81.4
017+4 Oklahoma St. 0.7668 0.5335 24.4 14.8 89.6
018-1 Missouri 0.7592 0.5450 20.2 12.3 85.2
019-- South Carolina 0.7554 0.6837 24.1 15.0 79.2
020-4 West Virginia 0.7500 0.5367 18.9 11.9 81.4
021+3 Notre Dame 0.7326 0.5989 19.6 12.8 84.7
022-2 Utah 0.7299 0.4878 21.9 14.5 81.0
023-- Georgia 0.7162 0.6161 25.0 16.9 77.2
024+1 USC 0.7101 0.5908 21.9 15.1 84.0
025-3 Miami-FL 0.7045 0.6326 19.3 13.4 85.8

the TCU Horned Frogs?

New entries: none.

Dropped out: none.

Not only is TCU at the top of the TFG heap, but BCS champion Auburn is barely in the top 10 and behind three teams it defeated (Alabama, Oregon, and LSU).  How can that be?

Simple: TFG only looks at points, it doesn't actually look at victories. There's a disconnect between offense and defense, and given that Auburn's offense only averaged 24.7 PPG against the three teams ahead of it -- and gave up 21.0 PPG in those games -- it's not hard to see why the Tigers didn't overwhelm the computers. On top of that, the Tigers needed freak play after freak play to even get where they were, including
  • a dropped TD pass by Clemson in OT;
  • a last-minute field goal to beat Kentucky;
  • three Alabama turnovers inside the Auburn 5;
  • four failed attempts by the Ducks to punch it in from the 1 (yet they went 2-for-2 in 2pt conversions);
  • a failure of the Ducks to get the easy 3 on a 4th-and-goal from the 1; and
  • a tackle that was (correctly) called not-a-tackle in the closing minute of the title game.
I'm not pointing these out because I'm anti-Auburn or anti-SEC. The SEC was the best conference this year, but not for the main reason the pundits claimed ("Five titles in five years!!!!!!!!"). The SEC placed five teams in the top 12 and seven in the top 25. Auburn was the last team standing because (a) they were very good, and (b) they were amazingly lucky, in part thanks to some of the reasons listed above. But their defense remained suspect -- 7th-worst in the top 25 -- even as their offense placed just behind Stanford at 29.1 PPH.

So in the end it was entirely fitting that it was not Cam Newton punching through for the winning touchdown on a QB sneak (a play that would have sent the ESPN university into a rabid frenzy), but rather a freak failed tackle and a last-second field goal that gave Auburn the title. To the bitter end, they did it their way.

TCU, on the other hand, only had a close call against Wisconsin. Otherwise the Horned Frogs basically did what they needed to do to by winning and winning convincingly. The Wisconsin offense had been prolific -- albeit against inferior opposition -- and the TCU defense stopped it cold.

What would happen if 14-0 TCU played 14-0 Auburn? TFG says that (1) TCU would defeat (10) Auburn, 35-31, with 74% likelihood. That might be the debate of the offseason, but with TCU going to the Big East that will blunt some of the arguments over how non-AQ teams will get access to BCS bowls and the national championship game.

Let the debates continue, and we'll see you next year.

Monday, January 10, 2011

What's up with the ACC?

Part II in our trivia series examines the crazy world that is ACC football. Unpredictable and frustrating, even ESPN has given up on providing anything but head-scratching bewilderment at what's going on there. But is there any actual data to support the gut feeling that the ACC is an anything-goes-who-knows-who'll-win-this-one collection of teams? That is the question we examine today.

The first part of this is how we define "unpredictable". While it might be difficult to predict a single game correctly, given a large enough sample size and an accurate enough predictor we should be able to know how many we should get right (see: expected value). As a simple example, think of flipping a coin and trying to predict its behavior. If we always guess "heads" then we'll be wrong roughly half the time. But if I tell you I'm going to flip a coin 1,000 times and I want you to tell me how many heads will come up, 500 would be a good guess. If the coin is fair the answer should be somewhere in the neighborhood of 500.

Similarly, if over time our college football predictor is correct 75% of the time and I tell you there are 400 ACC conference games in our set, then we should be able to get roughly 300 correct. There might be stretches of bad luck where the predictor is wrong, and there will be some big upsets. But over time we should get nearly 75% correct.

With that in mind, let's see how our predictor holds up for each of the "Big Six" conferences. For each conference we'll look at the
  • total number of games,
  • number we expected to get right,
  • number we actually got right; and
  • ratio of actual to expected.
These results are from the 2003-2004 season until the 2010-2011 season.

Conference Games Played Predictions
# Expected correct % Expected correct # Actually correct % Actually correct Actual / Expected
ACC 383274.871.825566.60.928
Big 10 352255.872.726274.41.024
Big XII 392287.173.329475.01.024
Big East 195135.169.313669.71.007
Pac-10 345253.773.525373.30.997
SEC 392290.274.028873.50.992

Over the course of eight years the ACC is (a) the second-most-difficult to predict, and (b) even then we don't get close to the expected value. For most conferences the predictions are within 2-3% of what we expect to get correct; even the Big 10 and Big XII are within 10, but for those we're under-estimating the number we should get correct.

Let's put this in perspective: the Big XII has 49 conference games per year (48 regular season plus the championship game) and the predictions miss about 2.4% of the time. That means every 6 years it gets a total of 7 games wrong. The ACC, on the other hand, has the same number of regular season games but hits the 7-wrong-games mark in less than two seasons. This 7-game deficit is compounded by the fact that the computer is already less confident about ACC games to start with. The raw data tells that tale well enough; the computer gets 3 out of every 4 Big XII games correct, while it struggles to get 2 out of every 3 ACC games correct.

Why does this happen? We're not entirely sure. The ACC tends to chew through coaches somewhat regularly, so a lack of consistency may lead to fewer predictable results. Similarly the ACC is offensively challenged, so a favored team falling behind may have difficulty catching up. Small, improbable events such as pick-6s, interceptions, fumbles, and other odd plays could have a greater-than-usual effect on the result.

Until we figure it out, we'll just have to rely on the ACC Wheel of Destiny.