Sunday, December 7, 2014

Final SORTACUS Rankings

Saturday was an eventful day for the first ever college football playoffs, with all six teams under consideration playing.  Alabama, Oregon, and Florida State were all likely in with a win, but TCU, Ohio State, and Baylor were wrapped in a fierce battler for the fourth and final slot.

All six teams won their games, but the Buckeye's 59-0 demolition of a good Wisconsin team stood out as an Ohio Statement win.  It impressed the playoff committee enough to jump TCU when the final playoff bracket was revealed today. 

Was this a just decision?  Did Ohio State really deserve a spot in the playoffs, or did a big name team in a big name conference sneak their way in?  SORTACUS is a computerized ranking system that only uses wins and losses, therefore effectively ranking the most worthy teams as opposed to the best. 

Top 25

For the fourth consecutive week I am applying the Bayesian statistics of SORTACUS to rank college football teams.  These are the final rankings, using every regular season game involving two FBS teams.  Below are the Top 25.


Rank Team Record Rating
1  Florida St 12-0 25.31
2  Alabama 11-1 23.66
3  Oregon 11-1 22.37
4  TCU 10-1 20.77
5  Ohio St 12-1 19.93
6  Baylor 10-1 19.40
7  Mississippi St 9-2 17.77
8  Mississippi 8-3 16.39
9  Michigan St 9-2 15.83
10  Arizona 10-3 15.81
11  UCLA 9-3 15.31
12  Boise St 11-2 14.84
13  Kansas St 8-3 13.92
14  Auburn 7-4 13.85
15  Georgia Tech 9-3 13.42
16  Marshall 11-1 12.93
17  Missouri 9-3 12.67
18  Clemson 8-3 12.66
19  Georgia 8-3 12.49
20  Wisconsin 9-3 12.20
21  LSU 7-4 11.91
22  Arizona St 8-3 11.80
23  Colorado St 9-2 10.81
24  Louisville 8-3 10.75
25  Nebraska 8-3 10.59

It's extremely close, but SORTACUS has TCU emerging from the pack of three.  Why?  Largely because the Big 12 had a better year then the Big 10.  Both teams had weak out of conference schedules (although Minnesota was a nice win for TCU), but Oklahoma, West Virginia, and Texas form a better middle of the pack than Maryland, Rutgers, and Penn State.  Baylor's horrible non-conference schedule dropped them to number 6.

It's possible that the committee used a different criteria than I have.  In fact, it appears pretty likely that Ohio State's margin of victory yesterday was a big factor in their trip to the inaugural playoffs.  It should be noted however that Team Ranking's college football rankings have TCU at #3 and Baylor at #4, both ahead of Ohio State. 

Overall, the playoff committee seems to have done a fine job though.  Florida State, Alabama, and Oregon are the three teams that definitely deserve to be in the playoffs, and they are.  The committee then had to pick one from a tier of three, and although I believe the wrong team was picked it's clear there was only small separation between them.

How Is My Team Doing?

Schools I'm associated with don't show up in the top 25.  Not much has changed since last week, but here's where a few notable ones land:
Stanford:  35th, 6.51
Washington:  42, 4.94
Maryland:  47, 4.45
Iowa:  58, 1.66
Northwestern:  73, -1.43
Iowa St:  93, -7.62
New Mexico:  96, -8.28
Purdue:  100, -9.19
SMU:  125, -20.83 (Sorry Uncle Mark)

Tuesday, December 2, 2014

SORTACUS Penultimate Rankings

It's late, so I'll do this quick.  SORTACUS is back again to rank the top 25 (really all, I only post 25) teams in college football.  These rankings use only win loss results, and answer which teams are the most deserving of a spot in the playoffs, not which teams are the best.

Top 25

Below are the top 25 in the SORTACUS rankings through the games of November 29.  With only a few games left the rankings won't change much before the end of the regular season.


Rank Team Record Rating
1  Florida St 11-0 24.13
2  Alabama 10-1 22.68
3  Oregon 10-1 21.04
4  TCU 9-1 20.58
5  Ohio St 11-1 18.56
6  Mississippi St 9-2 17.71
7  Baylor 9-1 17.56
8  Arizona 10-2 17.04
9  Mississippi 8-3 16.29
10  Kansas St 8-2 15.60
11  Michigan St 9-2 15.56
12  UCLA 9-3 15.35
13  Georgia Tech 9-2 14.45
14  Boise St 10-2 14.44
15  Auburn 7-4 14.08
16  Missouri 9-2 13.63
17  Wisconsin 9-2 13.51
18  Oklahoma 8-3 13.10
19  Clemson 8-3 12.80
20  Georgia 8-3 12.78
21  LSU 7-4 11.99
22  Arizona St 8-3 11.96
23  Marshall 10-1 11.23
24  Colorado St 9-2 10.80
25  Nebraska 8-3 10.77

As expected Mississippi State drops out of the top 4 after a loss to Ole Miss.  SORTACUS still recognizes the strength of the SEC West (or has a HUGE SEC BIAS!) and keeps the Bulldogs at number 6.  Once again however, SORTACUS's 4 most deserving playoff teams match the Playoff Committee's top 4.  Other notable features of the rankings include:

-TCU is well ahead of Baylor, again agreeing with the committee. Baylor's head to head win may have some value that SORTACUS doesn't see, but there is a pretty substantial difference.

-Ohio State is a clear number 5 team, and probably the most likely to sneak in if one of the top 4 teams loses next week.

-Wisconsin, Missouri, Georgia Tech, and Arizona all appear out of striking distance, even if an upset wins them a conference title.

All of these pale to one point the committee needs to see however:

Florida State Is Number One!

They have gone undefeated in a power conference, and played solid out of conference teams like Notre Dame, Florida, and Oklahoma State.  SORTACUS doesn't even know they're defending national champions, and still has them a clear number one!  Games are about winning and losing, not how pretty you look along the way, and the playoff committee is dancing on the edge of disaster with the Seminoles ranked number 4.


How Is My Team Doing?

Schools I'm associated with don't show up in the top 25.  Here's where a few notable ones land:
Stanford:  34th, 6.51     (Nice win over UCLA!)
Maryland:  46, 4.45     (Bad loss to Rutgers)
Iowa:  57, 1.80
Northwestern:  73, -1.37
Iowa State:  93, -7.58
New Mexico:  96, -8.20
Purdue:  99, -9.12

Monday, November 24, 2014

SORTACUS playoff update

Last week The Taste returned for the first time in a couple years to help determine which teams most deserve to be in the four team college football playoffs.  I'm running the rankings anyway, so I'll likely continue to post the results.  For now I won't give details of the system (someday maybe) but last week has some insight.  Or you can ask Andy Meisner.

Top 25

SORTACUS creates rankings using Bayesian statistics and FBS wins and losses.  Score doesn't matter, PPG doesn't matter, beating UNI doesn't matter, location doesn't matter.  Those last two are not necessarily on purpose, but just because I'm lazy.  Below are the rankings through November 22nd, along with record against FBS teams and SORTACUS rating.


Rank Team Record Rating
1  Florida St 10-0 23.32
2  Alabama 9-1 21.71
3  Mississippi St 9-1 20.88
4  Oregon 9-1 20.55
5  TCU 8-1 19.70
6  UCLA 9-2 18.18
7  Ohio St 10-1 17.52
8  Marshall 10-0 17.48
9  Baylor 8-1 16.86
10  Auburn 7-3 15.75
11  Arizona 9-2 15.43
12  Kansas St 7-2 15.28
13  Georgia 8-2 15.11
14  Michigan St 8-2 14.54
15  Mississippi 7-3 14.49
16  Colorado St 9-1 14.18
17  Arizona St 8-2 13.35
18  Missouri 8-2 12.90
19  Oklahoma 8-3 12.81
20  Boise St 9-2 12.66
21  Georgia Tech 8-2 12.24
22  Wisconsin 8-2 12.22
23  Clemson 7-3 11.19
24  Texas A&M 6-4 10.93
25  LSU 6-4 10.77

The top few teams held strong this past week, with Mississippi's loss to Arkansas the only upset impacting potential playoff teams.  The top 5 from last week hold strong, and match (with a slight rearrangement in the top 4) the committee's top 5.  Things diverge a little from there though.

How Much Does Record Matter?

UCLA, Ohio State, Marshall, and to some extent Auburn are an interesting set of teams to compare.  They all have a different number of losses, but their disparate schedule strengths make evaluation difficult.  Auburn has fought through the weekly gauntlet in the SEC West.  UCLA has been waging through a tough PAC-12 that has knocked out quality teams such as Stanford and Utah.  Ohio State is in the Big Ten, but this version of the conference is one of the weakest of all time.  Finally, Marshall has been dominating a string of nobodies.

What does the committee find valuable?  Marshall's unjustifiable absence from the top 25 would indicate it's strength of schedule.  But if that's the case, UCLA and maybe even Auburn should be ahead of an Ohio State team that is somehow sneaking it's way into the playoff discussion.  The way it's currently playing out the Buckeyes might reach the top 4 largely on their reputation.

How Is My Team Doing?

Maryland:  35th, rating of 6.34
Stanford:  47th, 3.59
Iowa:  52nd, 2.88
Northwestern:  64th, 0.46
Iowa State:  93rd, -6.73
Purdue:  94th, -7.18
New Mexico:  109th, -11.11

Sunday, November 16, 2014

Who Should Make The Playoffs? (feat. SORTACUS)

It's now been over two years since I made a blog post, but the process behind the college football playoff selection was too much for me to resist.  Does Ohio State deserve a shot?  Is the SEC overrated (or underrated)?  Should we be talking about Marshall?  I brushed off an old skill set in an attempt to answer these questions and more.

College Football Playoffs

I have long had a fascination with the process used to select college football's champion.  It a sport uniquely unsuited for a large playoff bracket, the kind the selects appropriate champions in the NBA, or creates a month of excitement in college basketball.  I wrote about my philosophy on NCAA football champions extensively a few years ago during what I termed BCS Week

For now I'll skip the details on who deserves to be in the playoffs, and instead hit upon the main points here.

-Championships, and therefore playoffs, should reward performance, not potential.  Luck happens in sports, and sometimes Northwestern beats Notre Dame.  The only goal in a game is to win, and playoff selection should reflect that.

-Not all wins are equal.  In other sports nothing besides record determines seeding, but with the disparity in college football that is often unreasonable.

-There should be an algorithm for determining qualifiers.  This is true in most other sports, the algorithm is just wins and tie breakers.  A committee of people, all with ties to teams, has inherent biases a computer doesn't.  Coming up with an algorithm is a challenge.  Fortunately, I have a proposal.

SORTACUS

SORTACUS is a computer ranking system that I created with Andy Meisner to rate items based off head to head comparisons.  It's currently used only to rank Andy's 100 favorite movies, and occasionally help my family select where to go to dinner.  Because games are just head-to-head comparisons however, it is easily applied to college football.

If these posts continue I'll describe the algorithm in detail at some point, but a summary will have to satisfy the math nerds for now.  It uses Bayesian theory to adjust the odds that a team has any given rating.  Each game is another observation to adjust the probability that team A has rating X.  It's fun and exciting math, if you're into that sort of thing.  For now though, I'll post a few summary points.

-Only wins and losses matter.  Margin of victory is ignored.  This makes the ratings worse at predicting future results, but better determining the most deserving playoff qualifiers.  If you want predictions and projections, check out my work for TeamRankings.

-Home and away are not implemented currently.  I think it would be reasonable to adjust and include a game's locations, it would just require some additions to the code.

-Only games against FBS teams matter.  A loss to Northern Iowa would slip through the cracks (for now).

-A win never hurts a team.  Many ratings systems include average opponent strength in some way.  A win against a bad team could hurt a good team, even if it was a lopsided game.  With SORTACUS, even a win against a junior high team can't hurt.

Top 25

The experienced readers likely just jumped straight to these ratings.  Below is the SORTACUS top 25 using games through Saturday.  In theory, the top 4 teams are the most worthy of a playoff spot so far.


Rank Team Record Rating
1  Florida St 9-0 22.62
2  Alabama 9-1 21.90
3  Mississippi St 8-1 20.39
4  Oregon 8-1 19.95
5  TCU 8-1 19.26
6  Mississippi 7-2 17.49
7  Ohio St 9-1 17.41
8  UCLA 8-2 16.68
9  Marshall 9-0 16.29
10  Baylor 7-1 16.20
11  Auburn 7-3 15.63
12  Georgia 8-2 14.38
13  Colorado St 8-1 14.36
14  Arizona 8-2 14.06
15  Michigan St 7-2 13.70
16  Kansas St 6-2 13.70
17  Boise St 8-2 13.44
18  Arizona St 7-2 12.90
19  Oklahoma 7-3 12.72
20  Georgia Tech 8-2 12.53
21  Nebraska 7-2 11.90
22  Missouri 7-2 11.23
23  Wisconsin 7-2 11.20
24  Louisiana Tech 7-2 10.53
25  LSU 6-4 10.46

Through Novermber 16 the playoffs should contain Florida State, Alabama, Mississippi State, and Oregon with TCU on the outside looking in.  The playoff rankings this week may actually agree with that, which is a good check for both SORTACUS and the committee.

What else stands out here?

The Big Ten is probably being overrated by the committee.  Ohio State is probably worthy of a fair shot at the playoffs, but Michigan State, Nebraska, and Wisconsin don't fare well with SORTACUS.

The SEC is as good as midseason polls said.  Schools in the West have beaten up on each other, potentially jeopardizing an SEC spot in the playoff.  SORTACUS puts five teams in the top twelve though, almost guaranteeing at least one deserves a final spot.

Marshall is number nine?  That seems high and probably is.  There aren't enough inter-conference data points for SORTACUS to work with.  Still, for Marshall to be undefeated but not among the 25 most deserving playoff teams is a silly mistake on the committee's part.

How is my team doing?

Schools I'm associated with don't show up in the top 25.  Here's where a few notable ones land:
Maryland:  41st, rating of 5.18
Iowa:  49, 3.51
Stanford:  55, 1.96
Northwestern:  68, -0.26
Iowa State:  85, -4.92
Purdue:  91, -6.15
New Mexico:  106, -10.68