2 "Real Games" and 11 Cupcakes? How Does Boise State's Schedule Stack Up?

I.  Introduction:

Every year Boise State fans are treated to the same arguments.  Boise State only schedules one or two "real" teams and plays "cupcakes" the rest of the time.  They would be a mid-level team in "any" BCS conference.  They do not deserve their high ranking. 

This year, these arguments have become even more pervasive as Boise State is currently ranked number 3 and the pre-season hype suggested Boise has a shot to play in the National Championship Game (NCG).  Defenders of the BCS are quick to claim that Boise's schedule is so weak they do not deserve a shot at the NCG and they probably should not be invited to play in any BCS bowl except, perhaps, as an act of charity as CBS sportswriter Greg Doyel suggests.

Fans of Boise State, myself included, feel the BCS is a system designed to exclude all teams from non-AQ conferences.  From my personal perspective, the BCS functions like a cartel and the best an AQ team can hope for is an invite to one of the big games every now and again.  Boise State is fortunate that they have at least have had two such invites in 4 undefeated seasons.  And even our critics admit they have done well on the big stage.

So whose fans are right?  In this article, and I apologize in advance for its moderately technical nature (I am a math teacher), I examine the Strength of Schedule (SOS) argument of the critics to see if Boise really is excluded for their "weak" schedule or if failure to make the title game (assuming they go undefeated this year) is due to the BCS system itself.  In general, much as it pains me to admit this, the critics are correct.  But not to the degree they think.  There is also some truth to the argument that the BCS is designed to exclude schools regardless of SOS. 

II.  Method

In order to analyze the strength of schedule argument, I looked at the schedules of all 31 currently unbeaten teams.  I ranked each team's opponents using the CBS Sportline 120 poll.  This poll has an advantage over the two more commonly cited polls (AP, Coaches) in that it lists all 120 teams and is quickly updated following games.  With every college team now having played 2-3 games, and many games available for comparison, it is the most accurate gage of a team's strength of schedule. 

The use of the CBS poll however has two problems.  First, it is a work in progress.  As more games are played, there will be more movement in the poll.  However, those who have followed this poll over the years know that after week 3, which conveniently enough is when I am doing this article, the more dramatic movements tend to decline.  In other words, a team's position will continue to change week to week, but jumps of 20 to 30 places after a week of results decline considerably.

The second problem is that many teams schedule FCS opponents.  The CBS poll does not rank these teams at all.  In order to get around this problem, I used the Sagarin ratings.  This raises a 2cd problem, however.  Sagarin's algorithms (the formulas he uses to calculate a team's correct position) are not really accurate until week 6.  After that point, they are extremely accurate.  Indeed, his computer poll is probably the best of the computer components of the BCS.  But as of the time of this writing, his poll is at best just a proxy for information that is otherwise not readily available.

Finally, in order to address the "cupcake" argument I felt obliged to actually define what a "cupcake" is.  Most critics of Boise State think a cupcake can be defined as, "any team Boise State plays," but for purposes of this article, a cupcake is a team that is ranked in the bottom quarter of the CBS 120, or holds an equivalent (or lower) ranking from Sagarin if an FCS team.  (Only 4 FCS teams are not "cupcakes" by this definition:  in order, they are Richmond-84, Mass-72, Villanova-69, and James Madison-37.)

In this article, I will total the number of "cupcakes" each team plays, and I will calculate an average SOS for each of the teams examined.  I use a simple mean to calculate the average:  add the rankings of each team and divide by the total number of teams played.  A number of 60 would indicate a team plays an exactly median schedule (60 is 1/2 of 120).  Above 60 is comparatively easier.  Below 60 is comparatively harder.  Readers should not be surprised to find that none of the top teams really play "tough" schedules by this measure.  The reason is fairly obvious.  They cannot play themselves.  Similarly, bottom teams will often have a harder SOS, just by virtue of the fact that they play the top teams.  For example, New Mexico State (rank 115) has a strong schedule just by virtue of the fact they play Boise (rank 3).  Boise by contrast, cannot play the n.3 team, but they do play NMSU.  This same scenario applies to every top team.

III.  Results

A.  For Boise State the results are as follows.  The team plays 3 cupcakes.  They are New Mexico State University (rank 115), San Jose State University (105), and Louisiana Tech (rank 91).  I think all 3 of these teams fully deserve their rankings and no major conference has 3 teams that fall into this category.  Several have 2 however, and some major conference teams schedule additional cupcakes besides those found in their conference.

Boise's mean SOS is 71.33.  This means that the "average" team Boise plays is currently ranked just above 71, which is in the bottom half of FBS football.  Boise has been severely hurt by Virginia Tech's fall in ranking.  As of Week 3, VT is ranked 58.  On the other hand, the fall of VT has been partially off set by the rise of both Fresno and Nevada, both much stronger teams than the critics assume.  Nonetheless, Boise's schedule, while not as weak as some suppose, is not very strong either and as we shall see, it does not stack up well against many of their potential competitors for a national title game.

B.  One competitor that Boise State does do well against is TCU.  Many in football assume that TCU, because it plays in the Mountain West, must have a stronger schedule than Boise.  They do not.  TCU, currently ranked number 4, plays 5 cupcakes:  Tennessee Tech, Colorado State, UNLV, San Diego State, and New Mexico. They share two common opponents with Boise:  Wyoming and Oregon State (currently 28), but their average SOS is 76.7.  Even excluding their loss to essentially the same Boise State team last year in the Fiesta Bowl, there  is no way undefeated TCU team should be ranked ahead of an undefeated Boise State team this year if you believe the SOS argument.  Those who believe that TCU's schedule is stronger simply do not know the facts.

C.  Alabama.  The Crimson Tide play only one cupcake, San Jose (105).  Their mean schedule rank is 45.1 which, considering that they are the number 1 ranked team, is truly amazing.  (As noted above, they cannot play themselves, and that in and of itself hurts their SOS!).  Put as bluntly as possible.  If both Boise State and the Tide should find themselves undefeated at the year's end, there is no way Boise should get into the title game ahead of the defending champions.  I will examine the question of what should happen to a one or two loss Tide later, but to preview a little, the results are not good for Boise State fans like myself.

One might wonder if the above analysis is true for all SEC teams.  After all, the SEC is the strongest conference in football.  Shouldn't any team that runs the table from the SEC (especially when you consider the title game, which can't be factored into my analysis at present) be ranked ahead of Boise State.  Here the answer, however, is a resounding no.  Alabama's schedule is strong not merely because they play in the SEC, but also because they are able to schedule out of conference opponents.  Other SEC teams, however, are notorious for not doing so.  Consider the case of Kentucky. 

Kentucky is currently 3-0, and it is still mathematically possible they could run the table.  If they were to do so, would they not be more deserving than Boise?  No.  Here is why.  Kentucky also plays 3 cupcakes: Western Kentucky, which at a rank of 120 is the worst school in FBS, Akron (114), and Charleston Southern (203!).  Moreover, the rest of their SEC schedule simply is not as strong as Alabama's.  Their average SOS is 76.1, barely better than TCU. 

The sad thing is, Kentucky is far from the exception to the rule in the SEC.  Teams there schedule very easy non conference games, and not all the conference games are all that impressive.  Vanderbilt is 84.  Ole Miss is 89!.  But for purposes of this analysis, it simply means that Alabama's schedule is all that much more impressive.  The Tide are not resting on their accomplishments last year, nor on the the rank of their conference.  They are ranked n. 1 and deservedly so.

D.  Ohio State.  The Buckeyes have nowhere near as strong a schedule as Alabama.  They play 3 cupcakes:  Eastern Michigan (119), Marshall (108), and Minnesota (100).  Sorry Gopher fans, but all these teams have fully earned their rankings too.  However, the Big 10 is strong this year, and the Buckeyes have an average SOS of 57.6.  This is still significantly better than Boise State. 

Unlike the SEC, all teams in the Big 10 with the potential to go undefeated also have stronger schedules than Boise.  The least of these, Northwestern, currently 3-0 has an average SOS of 63.8, below median strength, but still ahead of Boise's 71.3. 

E.  Oregon is the top rated PAC 10 team, and they have put up big offensive numbers over the opponents they have faced.  But their situation is a little different from the other top teams discussed thus far.  Like Boise State, they face 3 cupcakes:  New Mexico (118), Portland State (145), and Washington State (99).  If anything, I think the final ranking is generous!  Their average SOS is 62.4.  This is better than Boise's, but I think two factors mitigate against an undefeated Oregon team going ahead of Boise.  First, they lost to Boise, badly, in both of the previous years.  This is, it bears repeating, essentially the same Boise State team as the previous year.  So, as with TCU, Oregon does not deserve a title ahead of Boise. 

A different PAC 10 team might, however.  Both Stanford and Arizona are ranked higher, and each has a stronger SOS than Boise.  PAC 10 fans often complain that the round robin schedule of the league (soon to change!) hurts them, but if you really believe the SOS argument, it helps.  Unlike Kentucky, which can schedule cupcakes and with a little luck avoid most of the SEC powerhouses, no team in the PAC 10 can escape any other team.  So if SOS is all the critics of Boise claim it is, then the PAC 10 must be regarded as the second best conference, even with the penalty USC faces this season.  (The Trojans will remain unranked in the BCS this year, a fact that makes the CBS poll yet more valuable.)

F.  Nebraska.  As a Boise State fan, I have a special hatred for the Cornhuskers.  They turned down a home and home with Boise, turned down a 1 and done with roughly the same price tag that they are willing to pay to Idaho, and then had the gall to offer an insulting 2 for 1 deal that did not even match Boise's open dates.  And their fans, after all that, blamed Boise for not scheduling up!  But based on this year's schedule, the truth is that the Huskers do not need to schedule Boise.  They have only two cupcakes:  Western Kentucky (120) and South Dakota (132).  Their rank is exactly 61, or about 10 ahead of Boise.  If both teams are undefeated (sob) then Nebraska, as of the current numbers should go ahead of Boise to the NCG.

G.  West Virginia.  Here is where Boise's case is much stronger.  West Virginia is 3-0, and they play in an automatic qualifying conference, the Big East.  History suggests that an undefeated team from the Big East will be ranked ahead of any non AQ school.  (And indeed, this is precisely what happened last year when Cincinnati was ranked 3 at the end of the regular season, ahead of TCU.)  Similarly, the winner of this conference, undefeated or not, gets into a BCS game ahead of any AQ team!  I think we all remember how Cincinnati struggled to get by a Hawaii team that Boise had manhandled two years ago.  Of course, Boise did not go to the BCS that year, but the Bearcats did!)

Bottom line:  Based solely on the SOS argument, should WV go ahead of Boise?  No.  Like Boise, they play 3 cupcakes:  Coastal Carolina (198!), Marshall (108) and UNLV (110) and their overall SOS is slightly lower:  72.58.  They also don't have the "history" of Boise State recently, meaning they were not undefeated and ranked last year with essentially the same team.

H.  The situation is even worse for the ACC champ, should they go undefeated by some miracle.  Like Boise, their SOS has taken a big hit, and not just from Virginia Tech.  The entire conference is down.  The highest ranked undefeated ACC team is Boston College, currently 2-0 and ranked 30.  They play two cupcakes:  Weber State (149) and Kent State (96) and no team in the top 30!  They also are not coming off a spectacular season with virtually all their players returning.  Even with an SOS of 66.4, they do not deserve a title game invite over Boise State. 

IV:  Conclusions:

What is one to make of all this?  In general, the SOS argument has more merit than I would normally like to admit.  It is true that Boise has a difficult time scheduling quality opponents, but in terms of this analysis, it does not matter.  I am looking at the schedules as they are, not as I would like them to be.  Worse yet, this analysis suggests that the move to the Mountain West will actually not help Boise much.  The Mountain West is slightly stronger right now, with TCU and Utah, but as those teams cannot play themselves, the schedules they actually play are weaker.  Utah leaves at the end of this season, along with an overrated BYU, to be replaced by Boise, with Nevada and Fresno to follow soon.  What this means is that, at best, the move will only marginally improve SOS, if at all.

On the other hand, there is some merit to the argument that Boise will not be allowed to play for a national title in any event.  History suggests that even a Big East Team or an ACC team, with an undefeated record, will be given the shot ahead of BSU, and anyone who claims this is based on merit is either ignorant or lying.  Pending the BCS defenders modifying a few of their claims, beginning with the argument that Boise is at best a mid level team in "any" BCS conference, the fact remains that Boise should be in a BCS game every year they go undefeated and a system that rewards the Boston Colleges and West Virginia's of the world is in fact a cartel and not a merit based system.

Now, one can make the argument that the Big East and ACC aren't always this bad, and that is true.  But if you look back over the years, an undefeated Boise is always better than the best of one of the so called Big 6 conferences.  So the claim that the BCS is a wonderful system that really does match the "best" teams simply does not hold up.  It really is a cartel to protect money and power from all comers.  Even if you buy the SOS argument, and I now think it has more merit than I once did, the fact remains the BCS has to go if you want to see truly great college football on an even playing field.

Finally, what about the argument that a BCS team with one (or even two) losses deserves the title more than an undefeated Boise State team?  Here the math is a little more difficult.  The basic problem is, it depends on who the loss(es) are to.  But in general, a loss is worth about 10-15 spots on an SOS chart (based on how 1 loss teams are ranked v. no loss teams on CBS Sportsline poll in the past.)  And here is where the bias of the BCS defenders is really apparent.  The fact is, of all the teams examined above, only a 2 loss Alabama has even a chance of being more worthy than an undefeated Boise (again, depending on who the two losses are to).  One really cannot praise the Tide's tough schedule enough.  A one loss Tide is still a shoe in over Boise.

What about the rest of the title contenders?  Well, no ACC or Big East Team should get in ahead of Boise, even if undefeated, much less if they have a loss.  The fact is, they will still have a better shot at a BCS game, even with 3 or 4 losses, than an undefeated Boise, but that's just the way the system is.  But no team from either conference, at this point, and based on the SOS theory, deserves a title.

As for the others, a one loss team from the Big 12 will always fall ahead of Boise.  The same cannot be said for the Big 10, though in the case of Ohio State, they deserve to go before Boise.  Despite the overall parity in the PAC 10, no PAC 10 team with one loss deserves to go ahead of Boise State.

V.  Final Notes:

It goes without saying that as the season progresses, the SOS rankings used here will change.  It is likely that Boise's will rise (I still think Fresno and Nevada are slightly underrated, and VT will not stay at 58!) but it is unlikely Boise will see an SOS ranking of better than 65 at any point this season, which puts them at a disadvantage.  Both Big 10 and Big 12 rankings are likely to get slightly worse, but not significantly, and I think the ACC rankings will improve, but not enough to offset their disappointing season thus far.

The bottom line is that Boise State's schedule is weaker than other top teams.  It is not as weak as many assume, and they demonstrably do not play "11" cupcakes.  But it is still weak enough that several teams with one loss could reasonably claim to deserve a title shot more than Boise.  One team, Alabama, could legitimately make that claim with two losses. 

However, the fact remains, the BCS functions like a cartel and the inability of SOS to account for putting both the Big East and the ACC ahead of Boise suggests it is a cartel.  Those who claim the BCS is based on merit are simply wrong.  Because of this fact, I maintain that, for Boise, TCU, and any other non BCS power that may come along, SOS ultimately does not matter.  Even if they were to schedule all the top teams outside themselves (and those teams were willing to play them) it is unlikely they will ever be invited to a title game.  The SOS argument has merit, but that merit is devalued by a BCS system that is inherently unfair.  It needs to go.