Doomsday Scenario

by rencito on Nov 4, 2009 12:16 AM EST Comment 33 comments

Probably like many others out there, I am cheering for something ridiculous to happen so that the BCS is drawn up differently.

Here is what I think would cause the most chaos:

Texas wins out and gets to be in the championship game.
Iowa loses to Minnesota (okay, that one is just personal).
Cincinnati wins out and finishes undefeated.
TCU wins out and has some impressive wins en route to an undefeated season.
Boise State wins out and finishes undefeated.
Oregon wins out and finishes 11-1.
LSU wins out (beats Alabama and then Florida in the SEC title game).

Oregon and LSU would both be 11-1 and Alabama and maybe Florida would fall behind them in the rankings. Does anyone else think that the voters would vault LSU straight to #2 after beating both 'Bama and Florida (LSU is currently #9 in the BCS)? I think they would, I'm not sure what the computers would say. LSU would also have to jump over Oregon (who is currently ranked 8th in the BCS). Oregon only plays teams .500 and above from here on out, so their schedule wouldn't hurt them too much.

So here would be the mess: LSU would jump over a team it is currently behind, plus undefeated teams from Boise State (who beat Oregon), TCU (wins over BYU, Utah, Clemson, Virginia), and Cincinnati. Furthermore, BSU (probably) would get left out of the BCS all together and miss out on BCS money.

Any other good scenarios that could lead to too much controversy?

0 recs | Comment 33 comments

Story-email Email Printer Print

Comments

Yes.

Didn’t I just talk about this?

1) Oregon goes to the NCG and wins, TCU finishes above Boise State and goes to the Fiesta Bowl and undefeated Boise State gets left out of the BCS altogether.

2) Undefeated Iowa gets left out. Delaney’s response: Priceless.

3) 80,000 people show up for the BCS annual meeting in April for a tail-gate themed rally to show they would travel for a playoff.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 4, 2009 12:32 AM EST reply actions 0 recs

Not that it will happen this year, but I think one way where the BCS would have to be “blown up” is in a given year when say two non-BCS teams finish undefeated (take your pick from Boise St, BYU, TCU, Utah, Houston, MAC, etc) and every other BCS conference finishes with teams with 2 or more losses.

At that point, they would almost have to vote in the 2 non-BCS schools into the title game.

While utesfan rhetorical answer would hold true (See, look they can qualify for a title game), the BCS conferences would not be very happy seeing the $$ being siphoned off to the non-BCS schools and not having a shot at the title,

Personally, I am for any scenario that cause the BCS to be revamped to include some form of playoff. As I have posted before, a 16 team playoff can be done, but I suspect that it would most likely end up as an 8-12 team playoff.

by talonk on Nov 4, 2009 12:13 PM EST reply actions 0 recs

The revenue for BCS bowls does not depend on which bowls you go to.

The two NCG teams would still recieve the $9 and $4.5 alloted for the case two teams from outside the automatic qualifying conferences participate and the automatic qualifying conferneces would still get thier $18 million each. Two would get thier additional $4.5 million for qualifying a second team.

The BCs revunue would not be altered and all of the BCS bowls would have thier traditional tie-ins.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 4, 2009 3:56 PM EST up reply actions 0 recs

ok, sorry I goofed on that.

FYI, check Mandel’ Mailbag at SI today … he has a plus one concept that seems plausible. Unfortunately, it really is only a 4 team playoff whi ch I feel shuts out too many other teams.

And I posted my playoff scenario under your If I was King post.

by talonk on Nov 4, 2009 4:34 PM EST up reply actions 0 recs

If Iowa and Cincy go undefeated and make the NC

the guys heads at Fox would explode

"Polamalu’s lineage can be traced through several roots. Chuck Norris mated with an Amazon Queen, and on the other side, Tony Hawk mated with Mother Nature. The two children of these spawned and fused in a tantric love session to create Troy Polamalu. The mother however died as he tore through the birth canal with a spin move."
Mechem on the roots of Troy Polamalu

by WVPiratesfan on Nov 4, 2009 1:50 PM EST reply actions 0 recs

Would be awesome though

"It was almost like if Harry didn't call it, it wasn't real." - Jayson Stark

by Chris Haines on Nov 6, 2009 1:21 PM EST up reply actions 0 recs

I hate any +1 option

+1 solves nothing whatsoever. If you are going to go to a +1, why not just have an 8 team playoff? The problem with a +1 scenario is that the +1 often isn’t needed (or otherwise wouldn’t be conclusive since the +1 team is arguably MORE debatable than the NCG). It often isn’t needed because the #1 choice is fairly clear after the NCG and other bowl games. For instance, if #3 loses its bowl game, what is the point? Then you just have a new #3 from the rest who may have two losses while the the NCG winner may very well be undefeated.

+1 is arguably worse than we have now. It compounds the problem rather than solves it.

by rencito on Nov 4, 2009 7:17 PM EST reply actions 0 recs

Evidence

2008: Last year was a mess. Texas, USC, and Utah all would have had a legitimate case for the +1 scenario

2007: Ohio State and LSU were 1 and 2. Virginia Tech and Oklahoma were 3 and 4. They both lost.

2006: Ohio State and Florida 1 and 2. #3 Michigan lost and Ohio State was still ranked 2nd. If voters had dropped OSU further (which they undoubtedly would have done under a +1 pretense), then they would have had an LSU/Florida rematch—they played in the SEC championship game. That wouldn’t be fair to Florida at all. Boise State was the only undefeated team that year (the year they beat Oklahoma in a BCS game) and they were still only #5.

I reiterate that I HATE any +1 scenario. The bowl games are just an extension of the regular season. It doesn’t clear up the situation at all.

by rencito on Nov 4, 2009 7:32 PM EST up reply actions 0 recs

While I understand what you are getting at, but if the +1 is modified so the 1 seed plays the 4 seed, while the 3 seed plays the 2 seed, with the winners advancing, I think that is what most envision (fans anyway, not the BCS honchos themselves). Sort of a 4 team playoff more than an actual +1.

For example, in 2007, it would have been OSU vs Oklahoma with winner playing champ of LSU vs VA Tech.

But yes, if it is +1 with #1 playing #2 with the winner playing the highest ranked team left afterwards, I can see your point better.

by talonk on Nov 4, 2009 10:29 PM EST up reply actions 0 recs

I wouldn't call it a +1

Why not just call it a Final Four? Okay, that name is taken. So try something other than +1, since what you are describing is a 4 team playoff.

by rencito on Nov 4, 2009 11:55 PM EST up reply actions 0 recs

Valid point

I wish I read this before naming my tier based design. Final-four is generic and used in multiple leagues. Copyrights don’t apply to names like this that have entered public domain.

One of my goals here is to develop language that better describes ideas and terminology that provides a good and consistent description of the structures considered.

From now on I will reserve plus-one to refer to any four team design that determines the NCG participants after the BCS bowls or uses two of the BCS bowls as a semifinal game based on the top two conference tie-ins.

If a system adds qualifying games for any BCS slots these will be called play-in games.

Systems larger than 4 teams will be called a single elimination tournament of the number of teams invited. These can be modified by specifying a selection method from among: BCS ranking only, top N conference champions or all conference champions.

Technically my primary design is a variable plus-one with play-in options available.

My 10-2 additional bowl design would be a final four with play-in support.

talonk’s design would be a 16 team single elimination tournament with all conference champions.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 5, 2009 12:24 AM EST up reply actions 0 recs

I would like to see a flex system...

where you play what is necessary…by my calculations, UT, Utah, FL, and USC are still alive from 2008. Three more games were needed. But, in 2005, USC and UT were the clear frontrunners, faced each other, and Texas won…case closed.

"Stats are for losers. I like winning games." - Will Muschamp

by Mulliganville on Nov 4, 2009 7:24 PM EST reply actions 0 recs

Based on the BCS standings...

Penn State was closer to Texas in 2005 than Auburn was to Oklahoma in 2004.

Check out the ideas in the my proposal section.

I have spent the last year and a half honing a flex system of my own.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 4, 2009 10:39 PM EST up reply actions 0 recs

But Penn St. had a loss...

UT and USC were undefeated.

"Stats are for losers. I like winning games." - Will Muschamp
"Somebody will always break your records. It is how you live that counts." - Earl Campbell

by Mulliganville on Nov 6, 2009 7:38 PM EST up reply actions 0 recs

I'm just saying...

By my design three teams would have entered in 2005. Two for being undefeated and one more (Penn State) for being close enough in the BCS standings. This is the most arguable year against my design in the 11 year’s of the BCS.

Their is almost always a gap in the standings larger that that between Texas and Penn State in 2005, like that between Penn State and Ohio State in 2005. Both are near the boarder line I have set. I should research better the magnitude of the cutoff on the past 11 years.

Upon further review I was mistaken that is was closer than Auburn-Oklahoma. This error came from the fact that at this point last year my cutoff was defined differently to determine when a #3 team was worthy to play the #2 team and Auburn would not have qualified.

This first design called for an additional game that would either be a play-in game for the top two non-automatic qualifying teams to play in a BCS bowl or a #2 vs. #3 game to determine the national championship participant depending on the events of the year. If a semifinal was not needed and not enough non-automatic qualifying teams were eligible the highest ranked at-large not selected by a BCS bowl would participate.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 6, 2009 8:08 PM EST up reply actions 0 recs

Posted this on BON...

The Garbage Factor of the BCS…

…is IF the horns drop one, the polls are 2/3, and the later the loss, the greater the penalty which is the most absurd aspect of this moronic system we have in place. You hear that OU? No more at large title game berths for you.

That being said…let’s get out the crystal ball:

I think Cincy will lose in Pittsburgh.

I think Iowa will lose in the Shoe.

I think FL runs the table until the SEC championship.

I used to think Bama would absolutely run the table…but they have looked pedestrian for a few weeks (22 is highest point total over this span) against inferior competition. I am still inclined to give them the nod of being unbeaten going into the SEC championship game. I like FL right now in that matchup…regardless, one of them will lose.

I think TCU will run the table along with Boise.

I do think the Horns run the table which leaves us with 4 unbeatens.

Now, the aforementioned is what is looks like on paper. We could be looking at 6 undefeated teams after conference championship games. Moreover, the scariest team in the country right now, IMHO, is Oregon. Eugene will be the latest city to clamor for an 8 Team Playoff as I am sure they would love their chances this season. I for one am very thankful we do not have to line up against them this season. But, in an 8 Teamer, they would be a fitting and deserving champion were they to run the table on the assumption they finish top 8 in the rankings.

Once again, I personally believe that the BCS will fall short of its intentions all the while the Horns will benefit this season from a high pre-season ranking. Not that they would be undeserving, the Horns most certainly would be…but there will be controversy in Iowa, Cincy, Fort Worth, and Boise (weak argument here…but they did beat Oregon…and that is looking better by the week) as they feel snubbed by not having the opportunity to play for the title should they all run the table. Doubtful? Sure. Possible? That is why they play the games.

What a disaster of a system we have to try and settle this with annually.

"Stats are for losers. I like winning games." - Will Muschamp

by Mulliganville on Nov 4, 2009 7:43 PM EST reply actions 0 recs

The polls are the worst part

Take the power out of the hands of the pollsters. They are way too biased on teams, conferences, and recency. I trust computer rankings more than any pollster (some computer algorithms are better than others). It is absurd that the voters have as much power as they do.

I still say 16 team tournament. All 11 conference champs get an invite. 5 at large teams. Everyone gets a chance. This year:

Georgia Tech, Texas, Cincinnati, Iowa, Houston, Central Michigan, TCU, Oregon, Florida, Troy, Boise State as conference champs. At large: Alabama, Oregon, Penn State, USC, LSU.

Sure we could argue about the at large teams, but we would rather argue about that than who plays for the NC. This would be a great tournament.

by rencito on Nov 5, 2009 12:21 AM EST reply actions 0 recs

I like the computeres too.

But the MOV restriction needs to go.

With the current computers, the computer component is far less accurate than the polls at indicating the winners of games head to head. Computers that allow MOV equal or even exceed the polls in their predictive value.

The predictive value is what should drive which components are used for the BCS standings, not the philosophy of the algorithm designers.

The proof is in the pudding, not the ingredients.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 5, 2009 12:28 AM EST up reply actions 0 recs

Personally I dislike MOV because Oklahoma, Florida etal get wayyyy too much credit for pasting the Citadels of the world 70-7.

If the computers could incorporate a cap to the MOV, then that would be better. For example, any win over 21 points counts as +21. Those extra 28 points scored in blowouts wouldn’t mean better rankings just for scheduling a sacrifice school.

by talonk on Nov 5, 2009 1:48 AM EST up reply actions 0 recs

I like it

I really do believe MOV does matter. For instance, in baseball there is a Pythagorean Theorem used that is a pretty good predictor of what a team will regress to based on its past games.

For instance, I really believe Iowa is outperforming its expected winning percentage right now because it is barely beating opponents. You cannot expect a team with that small of a margin for error to keep winning.

Overall, I realize this probably isn’t as useful in football because there is far less of a sample size. So using MOV with some sort of cap could help fix it. Also, for purposes of MOV, perhaps it shouldn’t include games against FCS teams.

by rencito on Nov 6, 2009 8:53 AM EST up reply actions 0 recs

Interesting read on that Pathagorean Theorm of Baseball.

I don’t follow baseball stats much.

It looks like that exponent relates to the mean margin of victory, as wikipedia states without reference that 13-16 is used for basketball.

A more mathematical sound estimate would be:

%win = invnorm( ( total_points_for – total_points_against ) / ( league_average_MOV * sqrt( games_played ) )

If a least squares estimate is used to ranke teams based on MOV one finds the average difference between the predicted and actual values is roughly 7 points. Adding a three point home field adjustment knocks this down to 6 points.

As for including FCS teams, using those same least squares estimates of all 238 division 1 teams (or what ever it is since I last run one), one finds that generally teams like Florida are favored by 100+ points relative to teams like Charleston Southern. This would provide a solid incentive not to schedule such teams.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 6, 2009 1:26 PM EST up reply actions 0 recs

I do know all about Pythag and how incredibly terrible my Tribe has been because of it (although I’ll blame most of it on the cruddy bullpens, but that is a different topic altogether). However, Pythag, while interesting, isn’t the be all end all either. It isn’t super reliable, but more of interesting discussion point.

And you are correct that the extremely small sample size of 12 football games would negate the Pythag of a 162 game baseball season.

Like I tried to explain before, my fault with the MOV without a cap, really makes teams like Florida, Oklahoma, etc who will continually run and throw up until the final whistle against the many sacrifice schools on their schedules. But if a coach is a “good sport” and runs up the middle on every play in the 3rd quarter and wins only by 28, they somehow are a worse team than the one who won by 60.

This is why the MOV is currently thrown out. If they agreed to put a cap in, I would have no problem with it.

FYI, by the “Pythag”, I think you would claim a 5-3 Oklahoma with a +164 is a better school than 9-0 Iowa with +89 this year, which I don’t believe is true at all.

Even with a MOV cap of 28, Oklahoma comes in as +111 and Iowa is +85. I can’t see how Oklahoma really is “that much better” in this regard.

by talonk on Nov 6, 2009 1:26 PM EST up reply actions 0 recs

But don't the voters use MOV?

It would be hard to argue that teams that run up the score don’t fair much better in the human polls than the computers. It can also be shown that the polls coorelate much better with MOV rankings than thoise that dont use MOV.

Average MOV is the best predictor of bowl outcomes using any single stat. Better than W-L record. Adjusted for SOS it becomes a powerful ranking.

I can see the argument regarding Iowa and Oklahoma, but the idea that these teams are closer then they are currently ranked is not as absurd as you make it sound.

My argument will always be that the choice of rankings should be based on performance not methodlogy.

Do we want to eliminate teams for the BCS because they choose to use an unfashinable technique on offense? Those who hate the triple option might like that idea.

BCS Evolution -- Punctuating the Equilibrium - twitter

by utesfan100 on Nov 6, 2009 2:26 PM EST up reply actions 0 recs

but don’t the voters use MOV

Yes they do, that is why I like the computer polls much better. Except for Billinglsy who tries to approximate the human polls.

I think MOV should be included in a human poll, but to give Oklahoma more credit for stomping a bad FCS school by 60 than say VA Tech for beating Southern Mississippi by 14 is just plaing wrong. And unfortunately, that IS what the humans do, because 60 is prettier than 14 even though the competition is that much tougher.

As I have stated, I believe MOV can be used but only if it is capped AND the SOS of the opponent is factored in. Otherwise, you have to leave it out. Am sure the computers used in the BCS can come up with something along those lines. They have much better computational tools than I do.

While I agree Oklahoma and Iowa might be closer than 9-0 to 5-3, I would still expect it to be closer to 8-1 Iowa to 6-2 Oklahoma, not completely the other way around because Oklahoma has a higher differential.

by talonk on Nov 6, 2009 3:43 PM EST up reply actions 0 recs

For the record

I was actually agreeing with you above in my post titled, “I like it”. The part that I like is capping the MOV factor at 20 or whatever so that it isn’t skewed by one large outlier, particularly against out of conference patsies. Again, I think MOV should be left out of the equation for games against FCS foes. That would make the schools think twice about playing them since it wouldn’t help them as much in the computer rankings.

I don’t think MOV should be the ranking system, but just one of several factors (variables) in the equation. I recall putting together a ranking system of my own last year to see if I could tell how good my Golden Gophers really were after a 7-1 start. I included MOV as a variable and predicted them to finish at 8-4 (they actually finished at 7-5) despite the strong start.

My point is that I think it can be a useful factor in ranking teams, but definitely not the end all ranking. SOS, turnover margin (because I believe turnovers are somewhat random and that a defense “creating” turnovers isn’t really a skill, but more luck, so it should be normalized), and yards differential are also other important factors that can tell you how good a team really is.

by rencito on Nov 6, 2009 5:27 PM EST up reply actions 0 recs

If Iowa and Oklahoma played

…my money would be on Oklahoma, at a neutral site. I would take whoever was home if they played at either locale.

There is no way Iowa can keep up this streak of coming from behind in the 4th quarter. The margin for error is just so small. Everything has to go just right.

by rencito on Nov 6, 2009 5:30 PM EST up reply actions 0 recs

Sounds like you would like my scenario.

by talonk on Nov 5, 2009 1:49 AM EST up reply actions 0 recs

And right on cue

Iowa loses a game. They will most likely lose at least one more.

by rencito on Nov 7, 2009 5:21 PM EST reply actions 0 recs

yes they did, however, if their QB didnt go out early in that game, I think they runaway with it. But that is why you have to have depth. They didn’t at a key position, and it cost them.

by talonk on Nov 7, 2009 8:56 PM EST up reply actions 0 recs

Not exactly a key to their team

Stanzi wasn’t exactly having a bang up year. 15 TD and 14 INT. He was responsible for them always falling behind while their defense helped them get back into games.

I’m not sure what makes you think they would have run away with it.

by rencito on Nov 8, 2009 10:02 AM EST up reply actions 0 recs

While Stanzi’s overall numbers do look poor, 5 of those picks were from the week before.

As for why I think they run away that game? Before Stanzi got hurt, they had accumulated 154 yards of offense in about 18 minutes. After, with a redshirt freshman who hadnt taken a snap all season, they accumulated 132 yards of offense in 42 minutes. I doubt Stanzi is limited that much the rest of the game, whether he throws a pick or not.

by talonk on Nov 12, 2009 1:05 PM EST up reply actions 0 recs