As I understand it, according to the “human” polls, USC is ranked #1. Then along comes this new-fangled “computer” poll which puts them at #3, meaning they won’t get a shot at the national title this weekend in the Sugar Bowl.
So, what gives? Am I mistaken in my assumption that these polls, whether “human” or “computer” generated are based on cold hard facts…like, oh I don’t know, ummm, lessee, ummm…the team that wins the most games is #1 ???
Obviously, I’m missing something…
The variables about which team is better are extremely complex. For instance, if it’s your alma mater, then obviously it’s the national champion and no other rating matters. 
The BCS is an attempt to computerize factors to come up with a definitive champion. Like all computerized rankings, you can get wildly different results by small changes in ratings of factors (computerized rankings are an inherently chaotic system). In this case, there were some games between teams the final week in the season that were utterly meaningless under any rational ranking, but which had important ramifications. It led to a silly result this year.
The only real “objective” champion would be in some sort of playoff system.
What if you have multiple teams with only one loss? What if the team with one loss played an immensely more difficult schedule than a team that played in the WAC and scheduled a few I-AA teams?
You couldn’t use the factor of the team with the most wins being the champion.
That isn’t the case in any sport with a postseason.
And college football teams don’t play an equal number of games. Most teams this year had 12 regular season games. But some had 13 or 14. You can get an extra game if you play in a few specific “preseason” games. You also get an extra game if you play a game against Hawai’i (to help defray the extra travel costs of having to go out there) or if you are in one of the conferences with a championship game (Big 12, SEC, or MAC).
Kansas State will be playing its 15th game of the year. The Wildcats played a “presason” game against Cal and in the Big 12 championship game.
Also keep in mind that there are over 100 Division I-A college football teams and there is no way to ensure that all of the good teams play each other. USC played 13 games and won 12 of them. They beat teams from the SEC, Big 10, WAC, MWC, and an independent (Notre Dame) in addition to their Pac-10 schedule.
But there’s no way to guarantee that you will play the best teams from those conferences with the lead time required to make a college football schedule.
Are you answering my question with a question or asking for clarification? (that’s an honest question, BTW - I’m not trying to be sarcastic)
While the “multiple teams with only one loss” question seems like it could be solved through a system of playoffs, I have no idea what the WAC or I-AA are.
Sorry, I’m really clueless on this whole thing…hence the OP.
Well, I wrote this, so I might as well post it.
The big problem is that there are what, like 100 teams in Division 1A football. They don’t always play the same number of games, the rankings change from week to week, and the top two teams don’t often get to play one another in the regular season. Some teams play very easy schedules, while others face multiple top-25 teams in a season.
What the BCS is supposed to do is combine a team’s (human-generated) national ranking with a statistical ranking generated by seven separate methods. The computer rankings track all sorts of stuff, from total team yardage to strength of schedule.
The problem this year is that Oklahoma was unbeatable until the very last game of the season, when they completely fell apart. They fell to number three in the human polls. But USC and LSU have also taken a single loss each, and the computers think that Oklahoma had the tougher schedule. So despite their most recent loss, Oklahoma is at the top of the list while the sportswriters and coaches think they should be at #3.
The entire purpose of the Bowl Championship Series was to ensure that the two best teams in college ball play one another. Now that USC has won the Rose Bowl over #4 Michigan, everything is all skeewumpus, because the #2 team is playing the #3 team for the championship as far as we non-silicon judges see it.
Many argue that the only real way to decide who is the best is to have a playoff series. The results of this season are going to give those folks a lot of ammunition, because what we’re going to have is going to look a whole lot like an unfinished season with two semi-finalists and no games left to play.
One other factor to throw into the equation. While USC won the PAC-10 conference (a conference on the west coast including teams like Washington (and State), Oregon (and State) among others), Oklahoma did not win their conference. So, thearetically you could have a national champion of college football who couldn’t even win their own conference.
What many people seem to forget however, is that the same thing happens in Pro sports as well.
In almost every NCAA sport winning your conference championship (in fact I think it is every sport), winning your conference has no bearing on your ability to win the championship.
Even team sports with very small playoff fields (water polo, men’s volleyball), there is a slot reserved for a “wildcard” team.
If this didn’t happen, every year the final of men’s water polo would be something like UCLA or Cal versus the school from the East that managed to make its way to the final.
The last two teams to win MLB’s World Series did not win their division.
What I don’t understand is how come every year people comment on who “should” be playing in the final game and complaining that the computer system didn’t put the “correct” team in the national championship. If people have this ability to determine who should be in the final game, why don’t we scrap the BCS and just take the top two teams in the human polls? Everyone gets so indignant that the computer screwed up, so why don’t we just go with what the polls say? Obviously having a playoff system would be ideal, but in lieu of that, I say go with the polls. As for the scenario where the ESPN/Coaches poll disagreeing with the AP poll, I say combine the two polls into a single one, because having two different polls is retarded.
There are different polls because different organizations wish to promote them. It’s difficult to say if the football polls exist to determine how everyone ranks or whether they are just around to sell newspapers.
The AP started its poll in 1936 as a marketing ploy to drum up interest in college football.
The computer polls are supposed to be objective and not prone to human biases.
Supposed to.
And what do you do when the polls look like this:
AP
1 USC
2 LSU
3 Okla
Coaches
1 USC
2 Okla
3 LSU
Or worse:
AP
1 USC
2 LSU
3 Okla
Coaches
1 Okla
2 LSU
3 USC
Such things happen with regularity. You need either one poll, or a tiebreaker. Since there are major trophies associated with each of the major polls, you can’t ditch one of them. That leaves a tiebreaker - which is more or less all the BCS is intended as. Just this year it came out wacky and inverted the major polls’ top 3.
I say bring on an 8 or 16 team playoff with an NCAA selection commitee a la basketball. Better the #7 or #15 team in the polls bitch about the #9 or #17 team getting in ahead of them than the #1 team in the polls being the one left out. The “bubble” teams left out of March Madness may complain, but nobody thinks they had anything but the longest of shots at the title anyways.
This is getting into IMHO territory but the BCS is doing exactly what it was intended to do: break the death grip the traditional bowls had on certain teams and conferences, and take a first step towards a true playoff system.
I myself am delighted that the BCS is not delivering a clear cut national champion this year. This just puts that much pressure on the handful of people (mostly school presidents and some of the bowls) who don’t want a playoff system.
BobT,
It is true that in every sport there are wild cards that are eligible for a championship, but there is no sport at any level where you can go straight from losing your league title straight to the national championship without winning some other game in between. Those wild cards earn there way to the final.In college basketball, it happens all the time. Teams lose their conference tournaments and then go on to the NCAA tournament.
The difference between a football conference championship and one in basketball isn’t all that great when you look at it.
Good for the BCS, weighing the relevant variables properly and making the tough call. It’s better than the emotional humans can do sometimes… (this coming from a UM fan!)
Ok, at worst, the BCS needs to add a “falls apart in the last game” adjustment variable for future years, if everyone feels that that is so important.
I don’t really follow college football, but it seems to me that most colleges are interested in the bowl game $ than in a good way to determine #1. And there seems to be an awful lot of bowl games. At the current rate I predict every team in the country will be in a bowl game by 2020 (no, I haven’t done the math)
Brian
Scrap the computers and simply insure that the top four teams play in two of the major bowls. Then play one extra game for the National Championship (a dead week before the Super Bowl would be ideal for the fans). There may be argument over who should be included in the top four between the 3rd, 4th, 5th, 6th etc. but it will be a certainty that the deserving number one will be included.
For example:
Poll A Poll B
It would be easy to select the top one or two teams. In this case USC and Notre Dame would be included, Florida State would most likely be included and the other high ranking teams will have to suffer a selection process for the fourth spot. But they certainly won’t be able to claim they were denied the number one spot if they are not included.
Now play 1 against 3 in a major bowl on New Year’s Day and 2 against 4. Then play the one extra game later that month. Rotate the major bowls as to the host of the two games. This will preserve the bowl system and determine an absolute winner with only one additional game.
On rare occasions a 5th or 6th ranked team may come out on top but there still won’t be a controversy at the end. In other words the team that lost out on the fourth spot can’t reasonably say that they would have defeated number one and number two.
Is this correct? I can’t find any hint of this at the linked site. Surely individual game stats would be a horrible distortion for a method trying to calculate the “winningest” teams. Team stats only matter in the contribution they make to victory. I would think any team would be happy with a win and less total yards than the reverse.
Florida was being blamed for running up the score against opponents under Spurrier, but he was playing under the rules of the BCS, that rewarded point spread. They changed the weighting the next year. They change the way the BCS scores teams EVERY DAMNED YEAR. It has not been “right” very often.
I’m not sure any formula can be right. Why the hell does anyone think it’s even possible to choose the best two teams out of 117 (I think) when they only play 12 games in a season with vastly different schedule strengths? The human polls are of course no better, tainted as they are by subjective factors (Michigan and Notre Dame get the benefit of the doubt while Bowling Green and Tulane get overlooked), preseason hype, and short memory spans (otherwise known as the “if you’re going to lose, lose early” rule). Choosing the best 8 or 16 is of course subject to the same difficulties, but the top half of your picks are likely to be largely uncontroversial, and the bottom half don’t matter as much in the larger pools.