Why the B1G Ten Isn’t Likely to Do Well in the Bowls (Again) and Why That Might Not Mean What You Think

simpsons-college-drawings-big-ten-conference-2013There are few certainties in life.  You must pay your taxes.  You will of course die one day.  And it’s December, so the B1G Ten is about to get pummeled in bowl games (FA-LA-LA-LA-LA, LA, LA LA LA).   And I am here to tell you that, yep, once again, the B1G Ten is on the precipice of a less than stellar bowl season (I will make some very loose predictions based on advanced metrics below–SPOILER ALERT–it ain’t good for the B one G).  But I am also here to tell you that there is a very good reason for this (that you will not hear from the talking heads on the Four Letter Network), and that, while far from a dominating conference, the B1G Ten likely isn’t as bad as they would have you believe.  Here is why.

I mean, the B1G Ten is really really bad, aren’t they?  Just google “the B1G Ten is weak” and pull out your buttered popcorn and peanuts, kick back in the easy chair, and enjoy the “literature.”  Truer words have never been spoken (we will not concern ourselves here with the difference between fact and opinion).  For instance, this article asks, What Happened to B1G Ten:  Once Powerful Conference Looks Weak from some prestigious source called the State Journal Register (a story like that has got to be true).  Here, their chief concern is a few early out of conference losses suffered by the B1G Ten early this season, a premise that made this author look rather foolish, detailed just below.  In this awe-inspiring piece, we are told that part of Maryland’s master plan was to escape the Almost Competitive Conference (“ACC”) and join the B1G Ten to “fatten up” on those poor corn fed mid-west boys–early returns have thus far been mixed (and it would seem no one likes their chances against Stanford, but that is neither here nor there).  Well, here’s the funny take-away from that piece, dated September 6, 2014, and I quote:

Two Big Ten teams were supposed to have a realistic chance to reach the new college football playoff come December. You can pretty much forget that now. In fact, even though the season is only two weeks old, it isn’t going out on a limb to say the Big Ten will almost certainly be the one conference not represented in the playoff.

Oh really.  Oops.  Looks like that Pulitzer Prize is going to have to wait, the hazards of running your gums with a mere 15 weeks of football left (I was five when I stopped calling my friends and “sticking it” to them five minutes into the first quarter).  And the prestigious State Journal Register was hardly alone as captured by this piece.

And they say journalism is dead.

But this isn’t a piece about rubbing it in that Ohio State was in fact selected for the College Football Playoffs over the reactionary statements of a few clueless pundits internet trolls, as much fun as that would be.  After all, that could just mean that Ohio State is really really (did I say really) good but the rest of the conference, not so much, and one team a good conference does not make (Or does it?  From that same Journal Register article, “[e]ven the ACC, which is made up of Florida State and 11 potential Buffalo Wild Wings Bowl participants, does have the Seminoles”).  Sigh.  More of that award winning journalism right there.

imagesBut, but, but, the B1G Ten really does suck, just look at how they perform in the bowl games University of SEC* fan says.  And to that, I say, exactly what does that mean?  Yes, the B1G Ten hasn’t been getting over on anybody lately, that’s for sure.  As a for instance, just last season, the B1G Ten managed to go just 2-6 in bowl games while the SEC registered a 7-3 mark.  And last year was hardly an outlier, with the 2012-13 season producing similar results.  But what is frustrating to this guy is that no one is willing to look behind the numbers and analyze the data, you know, ask the tough questions.  So I am going to do what the so called journalists and talking heads refuse to do–I am going to gather data, analyze it, then draw reasonable conclusions from said data in a way that conveys information to the reader about the bowl games that will soon transpire.  I know, a novel approach indeed.

The B1G Ten is sporting a hefty 10 bowl teams in this year’s festivities, just behind the mighty SEC (11), a statistic that sounds more impressive than it really is when one considers that some Power 5 conferences are now stocked with 14 members.  Thus, it stands to reason, especially with a number of less than stellar out of conference games, that some pretty mediocre teams will find their way into bowl games that you never heard of (and probably won’t watch).  But the difference between the two conferences is that the B1G Ten teams aren’t inspiring much confidence out there in the business world, to include those comfortably on their couch draining cans of Schlitz and pounding cheese balls, desperately trying to carve out a living off the efforts of some very gifted teenagers.  Feast your eyes on these betting lines from the good folks at Bovada (as of 12/17):

Team Spread Team Spread
Illinois +7 vs. Louisiana Tech
Rutgers +3 vs. North Carolina
Boston College vs. Penn State +2.5
Nebraska +7 vs. U.S.C.
Maryland +14 vs. Stanford
Minnesota +5 vs. Missouri
Wisconsin + 6.5 vs. Auburn
Michigan State + 3 vs. Baylor
Alabama vs. Ohio State +9.5
Iowa +3.5 vs. Tennessee

In case you are part of the uninitiated, the plus numbers above are not some exponential powers possessed by the B1G Ten Teams teams like some nefarious Dungeons and Dragons characters.  No, those numbers represent the bribe Vegas is offering prospective bettors in order to get them to wager on a particluar B1G Ten team this bowl season.  And yes, you are right, that doesn’t sound good.  And yep, your eyes do not deceive you–not one single B1G Ten team is favored to a win a bowl game this season.  And for some, that’s all the ammunition they need to support their conclusion that the B1G Ten is once again continuing its precipitous decline.  But understanding the conclusions that could be drawn from another sub-par bowl season requires looking behind the numbers.  As such, here are the same match-ups with some additional metrics that I will explain below:

FO Rank Team Spread FO Rank Team Spread
78 Illinois +7 vs. 56 Louisiana Tech
88 Rutgers +3 vs. 60 North Carolina
28 Boston College vs. 51 Penn State +2.5
29 Nebraska +7 vs. 13 U.S.C.
58 Maryland +14 vs. 20 Stanford
34 Minnesota +5 vs. 23 Missouri
30 Wisconsin + 6.5 vs. 11 Auburn
19 Michigan State + 3 vs. 9 Baylor
2 Alabama vs. 5 Ohio State +9.5
45 Iowa +3.5 vs. 42 Tennessee

The FO Rank represents a team’s ranking from 1 to 128 from football outsiders.com, an advanced metric that I have referenced multiple times on this site.  This site does a weighted ranking of every team broken down by each drive based on the strength of opponent played.  Though I will not delve into the nitty-gritty of their methodology, I will tell you that this has been my secret go to source in the past when drawing conclusions (for entertainment purposes only of course) and making bold predictions that have all too often come true.  For instance, a few years ago, I did a bowl pick ’em at a college football site for fanatics, and mostly using this data, finished second from a group of over forty some devoted fans (I was in first with a few bowl games left before getting screwed by some team like San Diego State).  I have spent hours at their site reading through their methodology and would recommend you do the same if interested in learning how the sausage is made so to speak.  That site, with their full rankings, can be found here.

What do those numbers tell us?  Well, it confirms, at least for this year, what I have always suspected–because of the popularity of the B1G Ten and the fact that their fans travel very well, B1G Ten teams are often over-slotted in Bowl games.  What do I mean by this?  I mean that statistically weaker B1G Ten teams are regularly paired with stronger opponents.   And that is exactly what has happened for each and everyone of the ten bowl games the B1G Ten will participate in this year, and why I wouldn’t expect a lot of December and January wins for the conference that has  perhaps erroneously become the butt of everyone’s joke.

Let’s take this analysis a little bit further, shall we?  If these metrics were 100% accurate, and I will be the first to tell you that this is not an exact science (and that there are certainly other metrics you could use), pairing teams ranked roughly equal (e.g. # 12 against # 13) would likely produce the best bowl games, at least on the statistical average (in any one game of football, statistical anomalies like a significantly higher than average turnover margin can result in skewed outcomes.  This is why, by the way, you want to look at as much data as possible**).  But even if we accept that these rankings are merely indicators of relative strength (i.e. approximations), it might also be that # 12 might still be a good match for say # 17 (this is akin to what scientists call margin of error, an acknowledgement that the rankings might be accurate within a certain +/- factor or range).  Even assuming we make such allowances, the disparities in match-ups above far exceed such margins as you will see in a moment.

635483976877728073-AP-Ohio-St-Nebraska-Football

Though the Conference finally disposed of “Legends” and “Leaders,” some arrogance still remains.  Fans of other conferences are all to willing to point out exactly where the Big Ten is not first rate.

First off, before I go through some numbers, I want to acknowledge some basic truths lest one believes that I am some sort of delusional utopian: there are many factors that go into deciding bowl match-ups, and that, unfortunately, a quality match-up of evenly matched teams often is the least of the considerations.  Some of those reasons include teams with fans that travel, historical powers selected over up and coming teams, and geographical rivalries to name but a few (though the Committee rankings were suppose to eliminate these type of considerations this year, I think it would be a bit naive to believe such concerns are not factored in.  Another possibility, of course, is that the Committee simply got the rankings wrong.  Either way, the B1G Ten teams were matched consistently against superior competiton).  I also will acknowledge that when the bowl pairings were done, they weren’t using this metric, but instead, the rankings from the committee, which primarily tabulated basic statistics like wins and losses and more mainstream strength of schedule metrics.  Thus, it isn’t expected that the pairings would mirror the data at football outsiders.com.  That it does not is merely an ancillary purpose of this piece.  My point is that, if these advanced statistics actually measure the relative strengths of the teams better than conventional statistics, then almost immediately it becomes apparent that the B1G Ten Teams are vastly outmatched in this year’s bowl games.  And I don’t think I am suggesting anything earth shattering when I predict that, in all likelihood, the stronger team will prevail in a larger percentage of the contests–translation, a drawer full of second place metals for the B1G Ten once again this year.

Now, I am going to extrapolate further from these numbers so we can draw some conclusions regarding the disparity of the match-ups.  Let me explain my methodology.  From the outsiders.com rankings, as discussed above, all 128 teams are ranked from 1 through 128, with one the strongest team and 128 the weakest.  To establish the value of each positional ranking, I divided 1 by 128, yielding a statistical advantage of .0078125 for each spot a team moves up in the rankings (again, we are assuming the relative accuracy of these rankings for the purpose of this discussion).  I then multiplied that number by the number of spots separating the teams, and converted that decimal into a percentage.  That is captured in the table below in the column marked Rank Adv.  This percentage represents how much “better” the higher ranked team is based on the football outsiders.com final rankings.  So, one more chart is illustrative, this time with the percentage of the “better” team rounded to nearest tenth of a percent:

FO Rank Team Rank Adv. Spread FO Rank Team Rank Adv Spread
78 Illinois +7 vs. 56 Louisiana Tech 17.1%
88 Rutgers +3 vs. 60 North Carolina 21.8%
28 Boston College 18.0% vs. 51 Penn State +2.5
29 Nebraska +7 vs. 13 U.S.C. 12.5%
58 Maryland +14 vs. 20 Stanford 29.7%
34 Minnesota +5 vs. 23 Missouri 8.6%
30 Wisconsin + 6.5 vs. 11 Auburn 14.8%
19 Michigan State + 3 vs. 9 Baylor 7.8%
2 Alabama 2.3% vs. 5 Ohio State +9.5
45 Iowa +3.5 vs. 42 Tennessee 2.3%

Just looking at the numbers, in the ten bowl games based on football outsiders.com’s advanced metrics, in half the games, the underdog B1G Ten team is more  than 10% worse than the favored team in its bowl pairing.  The poorest statistical chance of victory based on the disparity of rankings occurs for the State Journal Recorder’s darling Maryland Terrapins, who are 29.7% worse than the Stanford Cardinal.  And if one adopted a 5% margin of error (this is a pretty standard margin of error, though I could not find an actual margin of error on their site), only two teams are within 5% and thus are relative statistical matches (regardless of actual margin of error, it is still a basis of comparison of roughly equivalent teams).  Interestingly enough, perhaps the best bang for the buck would appear to be Ohio State, who is statistically only 2.3% worse than Alabama but is receiving 9.5 points at Bovada based on wagers placed by voters to date, though of course one should be mindful that much of Ohio State’s statistical advantage occurred with a quarterback that won 11 games, and not the one that will be making his second start against Alabama in the semifinals (it is remarkable how often the final Vegas line is on the mark).  If I were picking winners from this bunch, and I’m not seeing that many, I would pick the teams that have the smallest ranking disparity using these advanced metrics.  In order of most likely to win:  Iowa/Ohio State, Michigan State, Minnesota, and Nebraska.  When you factor in that Ohio State is on their third quarterback, or that Nebraska for example has a new coach, well, the B1G Ten in this guy’s estimation would be lucky to win 3 bowl games from this lot of mismatches.

No one is arguing that the B1G Ten is a dominate conference.  And of course, there is no way to know how the conference would do if the match-ups were a better statistical match (e.g. # 28 Boston College against # 29 Nebraska instead of # 51 Penn State.  That Boston College is ranked that much better than Penn State is an entirely different problem).  But, when the B1G Ten once again takes it on the chin, and they will, take solace from this article, because you won’t hear such statistical analysis from those purportedly paid for their analytical abilities.  After all, we can’t let the facts get in the way of a good narrative, can we?

* University of SEC fan refers to that guy who roots collectively for the whole SEC.  It is important to him that his conference can beat up your conference.  Though this is just likely a coincidence, he is usually a Kentucky, Arkansas, or South Carolina fan.

** To illustrate this point, and because the misinformation swirling around amongst the masses annoys me to no end, care to guess the B1G Ten’s record against the mighty SEC in the 38 previous bowl meetings through the 2012 season?  We all know that the SEC has been the most dominate conference in all of the interstellar galaxies and that the B1G has basically been their whipping boy during that duration, do we not?  Would you be surprised to learn that those meetings were split 19-19?  I recommend you check out one of my favorite articles on supposed SEC dominance entitled, Why The SEC Isn’t As Great in Football As You Think.  It’s nice to know that there are others out there actually looking behind the numbers as well.

As a note, the author of this piece has a B.A. in Political Science with an emphasis on statistical analysis.

Advertisements


Categories: College Football

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: