BCS Implications: Preseason


We are less than a week away from the start of another college football season and the chase for the BCS title has already begun, before the first snap is even taken. The Coaches’ Poll has been released and we can already get a look at who has an inside track towards the BCS title game. As I said last year, I think the importance of-and therefore the debate surrounding-the preseason polls is overblown. Especially in recent years, poll voters have shown a willingness to be fluid from week to week when a top team has an impressive or unimpressive performance. The only way that I see preseason polls being truly influential in the national title discussion is if 3 or more teams from AQ conferences go undefeated in equally impressive fashion. Nevertheless, the poll does give the season a framework to start with that, sometimes more and sometimes less, is always relevant as the season goes on.

Thus, to start the season, only 3 teams truly control their destiny. LSU, Alabama, and USC have a sizable lead on the entire rest of the field right now and if any one (or two) of those teams run the table in relatively impressive fashion, they will be in the National Championship Game, without question. Oregon also probably controls their own destiny, but that is nowhere near as much of a guarantee as the top 3. After that, though, every other team needs to strive for perfection and hope for losses by teams in front of them.

The non-AQ picture looks pretty much the same this season as any. Remember, a non-AQ conference champion gets an automatic bid to the BCS if they are in the top 12 or if they are in the top 16 and ahead of an AQ conference champion (and it looks pretty unlikely that the Big East champion will be ranked this year). Boise State will make a BCS game if they run the table. They have a decent schedule this year, so they’d probably get into the National Championship Game before most 1-loss AQ teams, but definitely not all. BYU could finish in the top 16 if they run the table and pprobably even with 1 loss, but cannot get an automatic BCS bid (unless they are in the top 2) because they are an independent. An undefeated Conference USA champ might be able reach the top 16, but an undefeated Southern Miss or UCF would definitely be ranked high enough. I don’t think that an undefeated Sun Belt champion would make the top 16. Ohio is an interesting case. When Ball State won their first 12 games in 2008, they did reach #12 in the BCS. And while this Ohio team is more talented than that Ball State team, the MAC is not as strong this year and Ohio’s schedule is weaker. It will be very interesting to see if Ohio can reach the top 16 if they run the table-which is quite possible due to their talent and a very easy schedule.

The Polls:

The Harris Poll will not be published until after Week 6, so it is impossible to make any judgments off of it until then. It is usually more similar to the AP Poll than to the Coaches’, but it is far from identical to the AP. So we can make educated guesses as to what it will look like, but it’s impossible to predict precisely how the Harris voters will lean. Also, Ohio State can be ranked in the AP Poll but they cannot in the Harris, so that would confuse trying to figure out any vote-share totals.

Always remember, the BCS formula does not care about what place a team is in the human polls. It only cares about the number of votes they get. Thus, the difference between LSU and USC in the preseason Coaches’ Poll is mostly irrelevant. The difference between LSU and Alabama is basically meaningless. Both gaps are so small that they could easily be made up by just a few differing opinions in the Harris Poll or by the computers.

The things to look for when looking at the human polls, especially early in the season, are significant gaps between teams. Any small gap, especially this early in the season, can pretty much be written off as a tie for now. But large gaps will take a bit more to make up for. The first significant gap in the Coaches’ Poll is between #3 USC and #4 Oklahoma. The 112-vote gap between those two teams represents almost a two-place gap in the rankings. That would be worth a .025 difference between the teams in the actual BCS standings. And while that may not sound like a lot, that is actually a pretty significant difference for just one of the three rankings criteria to make.

The other major significant gap to note is between #5 Oregon and #6 Georgia. That massive gap between the two teams (197 votes, more than 3 entire places-worth of votes) means that voters were a lot less sure of who precisely to put in the 6-9 spots (give or take) than the 4-5 spots. While this is fairly normal-to see big bunching the farther we get away from the top of the rankings-it also means that any of those teams will have to separate themselves from the others around them before they can even really begin to gain on those ahead.

The Computers:

While several of the computers publish preseason polls, only 1 is actually relevant to the rest of the season. The Billingsley Report uses a preseason ranking, which is based solely on each team’s finishing position the previous year. This ranking then affects the entire season, as each week teams gain or lose points based on the rating of their opponent each week. Also, with the exception that an undefeated team must remain ranked ahead of every team that it beats, the system does not correct itself if a team is shown to have been grossly underrated or overrated by year’s end.

Thus, early-season rankings of teams in Billingsley can be very important, especially in teams that should show significant improvement or decline over this coming year. For example, Boise State starts out the season at #8. If they end up this year around 20-30, as many expect them to, the first team to beat them will get credit for a top 10 win. And even if Michigan State beats them in Week 1, a loss to the #13 team will not hurt Boise State’s rating so much. Thus, the first real team to upset them (quite possibly Billingsley Preseason #29 BYU) will get a disproportionate rating boost for beating them. And that, in turn, will affect every other team to play Brigham Young after that this year (for an in-depth look at this problem with the Billingsley Report, see my article here).

So Billingsley is the one computer that we need to keep an eye on early in the year. Of course, it also helps that it’s the only one whose rankings matter this early in the season. Sagarin has a preseason bias in his rankings as well, but that bias is completely removed once enough games have been played and therefore it has no actual bearing on the rankings later in the season.

Week 1 Implications:

Aside from the obvious games involving top teams, some of these early-season games can have a lasting impact on the season-both in the eyes of the voters and the computers. The first type of games to keep an eye on are those between good AQ teams and top non-AQ teams. A non-AQ team that ends up 11-1 or 10-2 can really be a good SOS boost in the computers and a ranked win in the eyes of the voters. So keep an eye out in the first few weeks for those non-AQs. The few non-AQs that probably have the potential for 10-win seasons are probably Boise State, Ohio, Southern Miss, Brigham Young, and Arkansas State. Boise State visits Michigan State on Friday night, which could be a big factor later in the season as Michigan State is expected to be a contender in the Big Ten race. And a 12-1 Big Ten champion would definitely be in the national title hunt, and an OOC win over a ranked team like Boise State can only help that. Conversely, this game will really show us if Boise State has a chance to make a BCS game this year.

Southern Miss travels to another Big Ten team, Nebraska. Nebraska is in the same situation that Michigan State is, and Southern Miss has the potential to be a good OOC win (Southern Miss is also #20 to start the year in Billingsley, so a win over them would give a nice boost there too). Southern Miss probably has too tough of a schedule to end up with 10 wins and ranked, but they are a solid nonconference opponent nonetheless. Arkansas State travels to Oregon and we will see what Gus Malzahn’s offense can do with Sun Belt talent. Arkansas State, being in the Sun Belt, will not carry the same weight as Boise or Southern Miss in the eyes of the voters and computers, but wins over 10-win teams still do look good on resumes, and the Red Wolves definitely do have the potential to win 10 games. And BYU hosts Washington State. WSU will almost definitely not be part of the title picture, but BYU is a team to keep an eye on anyway, even though they will probably not face any title contenders (nor be one themselves) over the course of the season.

But, of course, the biggest Week 1 game is the obvious one. Alabama and Michigan will play in JerryWorld. It’s the only Week 1 meeting between two top 10 preseason teams. Alabama honestly stands far less to gain from this game than Michigan does. With the amount of respect that Alabama (and the SEC) receives in the polls right now, an undefeated SEC champion would not be left out unless they had an absolutely horrid SOS. And certainly no team from the SEC West. But Alabama can make a statement in this game against a team that will likely be a valuable win later in the season, especially if the Tide lose a game and have to fight it out with other 1-loss teams for a spot in the NCG. But for Michigan, this is a chance to really enter the national picture for the first time since 2006 and it would be a great early positive on a national resume.

Leave a Comment