header pic

Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

The 'Old' CFN/Scout Crowd- Enjoy Civil discussion, game analytics, in depth player and coaching 'takes' and discussing topics surrounding the game. You can even have your own free board, all you have to do is ask!!!

Anyone is welcomed and encouraged to join our FREE site and to take part in our community- a community with you- the user, the fan, -and the person- will be protected from intrusive actions and with a clean place to interact.


Author

Topic: The Helmets (and near helmets) rolling 10-year winning percentages

 (Read 13726 times)

betarhoalphadelta

  • Global Moderator
  • Hall of Fame
  • *****
  • Posts: 12188
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #154 on: May 27, 2020, 04:48:14 PM »
But even that "objective" assigning of a champion isn't possible today in college football.  The inequities among conferences and schedules are too vast. 
You'd need the P5 to split off from the rest.
You'd need the P5 to become the P4 (basically an ACC/Big12 merger)
You'd need the remaining 4 conferences to have the same number of teams and scheduling guidelines (yes/no vs FCS, if so how many, yes/no vs G5, if so how many, etc).

Even then, you SHOULD go a step further and include:
every team has 6 home/6 away games
every team plays every other team in its conference/if not that, then division/if not that, then pod
and then you'd still have co-champions, which would be determined by h2h, but statistically that's erroneous...

So in lieu of all that happening, I don't see the point of pretending it exists or will exist.
You need [almost] none of that if you just admit each P5 champion to the playoff.

I'll admit, the existence of G5 causes a major problem. You simply can't take "all conference champions" like you do in the basketball tournament because that gets you to a 10+ team playoff (12 if you give 4 byes, 16 if you don't, so you get 2 or 6 at-large teams). So you make a deal with the G5 that unless they want the P5 plus a handful of independents to formally break off, that they live with their top team being included. 

But the key about including conference champions is that it puts ALL of that extra minutiae off the table. It's incumbent on each individual conference to determine their champion. I don't care if their tiebreaker is the coaches of any tied teams play tiddlywinks. I don't care if the SEC plays 8 conference games with a November FCS tune-up while the B1G plays 9 conference games. I don't care if the ACC and PAC are weaker, top-to-bottom, than the other P5 leagues so their champions have lesser SOS. Win your conference, AS YOUR CONFERENCE decides it, and that's your representative in the playoff.

All of the concern about the rest of it (scheduling parity, # of conference games, OOC scheduling guidelines) only comes into play for the at-large slots. 

medinabuckeye1

  • Legend
  • ****
  • Default Avatar
  • Posts: 8906
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #155 on: May 27, 2020, 05:15:14 PM »
This made me think about 1984, and then 1983.

What happened in 1984 forced a group of 38 people (out of the 60 voters) to knowingly lie.  Everyone knew BYU wasn't the best team in the country, but it won the NC.  A 2-team or 4-team playoff would have fixed this.  Either BYU would have earned their #1 ranking or they would have been exposed.

In 1983, Miami wouldn't have even been in a 4-team playoff.  And while we all would probably agree that Nebraska was better than Miami, Miami won and earned the NC.  Unless you're an Auburn fan - 1 and 2 lost, #3 Auburn won their bowl, and didn't win the NC. 
Auburn would have been able to earn the NC in a 4-team playoff, but would have been excluded by the BCS.  I think the BCS exclusion is less important, as either 1 or 2 would have necessarily won.
Many would argue Auburn did earn the NC by finishing their season with wins over 5,7,4,19, and 8.  Yes, consecutively. 

You could just take my last 2 posts and take it as an opportunity to shit on the voters.  And I would join you.
I completely agree that BYU's 1984 championship was a complete fraud.  They just got REALLY, REALLY lucky.  It started in week #1:
  • On September 1, 1984 BYU played preseason #3 Pitt at Pitt and won 20-14.  That win catapulted BYU all the way from unranked to #13 but in retrospect, it shouldn't have.  In their next game Pitt (now down to #17) lost 42-10 to Oklahoma in Norman.  Pitt's preseason ranking turned out to be ridiculous.  They went 3-7-1 for the year and the wins came by a TD over an ECU team that finished 2-9, by 11 over Tulane (3-8), and by 20 over PSU (6-5).  
  • On September 8 they beat Baylor 47-13.  That shot them up into the top-10 at #8 and I just don't get why.  Baylor wasn't ranked and finished 5-6 so this shouldn't have been seen as impressive for a top-15 team which BYU was by then.  
Based on final rankings, BYU's performance against teams that were ranked or receiving votes was:
  • Won 30-25 at Air Force.  Air Force didn't finish ranked but they were 4th among ORV (only 20 ranked then)

That is it.  

Their Holiday Bowl win over Michigan was only by a TD against the worst Michigan team of the entire Schembechler/Moeller/Carr era.  That same Michigan team lost:
  • by 2 at Purdue (finished 7-5 and ORV)
  • by 9 at home to Washington (finished 11-1 and #2)
  • by 12 at home to MSU (finished 6-6 and nr)
  • by 15 at Ohio State (finished 9-3 and #13
  • by 26 at Iowa (finished 8-4-1 and #16
That BYU team was no better than high-teens.  They just got really lucky because everybody else lost.  BYU would have lost as well if they had played an opponent with a pulse but they didn't.  



OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 18847
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #156 on: May 27, 2020, 05:44:48 PM »
I think the voters would have put OU #1 if they had won their bowl.  And Florida possibly, but they were on probation.
“The Swamp is where Gators live.  We feel comfortable there, but we hope our opponents feel tentative. A swamp is hot and sticky and can be dangerous." - Steve Spurrier

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 18847
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #157 on: May 27, 2020, 05:47:13 PM »
You need [almost] none of that if you just admit each P5 champion to the playoff.

I'll admit, the existence of G5 causes a major problem. You simply can't take "all conference champions" like you do in the basketball tournament because that gets you to a 10+ team playoff (12 if you give 4 byes, 16 if you don't, so you get 2 or 6 at-large teams). So you make a deal with the G5 that unless they want the P5 plus a handful of independents to formally break off, that they live with their top team being included.

But the key about including conference champions is that it puts ALL of that extra minutiae off the table. It's incumbent on each individual conference to determine their champion. I don't care if their tiebreaker is the coaches of any tied teams play tiddlywinks. I don't care if the SEC plays 8 conference games with a November FCS tune-up while the B1G plays 9 conference games. I don't care if the ACC and PAC are weaker, top-to-bottom, than the other P5 leagues so their champions have lesser SOS. Win your conference, AS YOUR CONFERENCE decides it, and that's your representative in the playoff.

All of the concern about the rest of it (scheduling parity, # of conference games, OOC scheduling guidelines) only comes into play for the at-large slots.
Yeah, screw this - it penalizes the strong conferences.  Be like Clemson/FSU.  Punch your ticket every year you're not down.  
If you want objectivity, then you need to have equity.
“The Swamp is where Gators live.  We feel comfortable there, but we hope our opponents feel tentative. A swamp is hot and sticky and can be dangerous." - Steve Spurrier

betarhoalphadelta

  • Global Moderator
  • Hall of Fame
  • *****
  • Posts: 12188
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #158 on: May 27, 2020, 06:04:53 PM »
Yeah, screw this - it penalizes the strong conferences.  Be like Clemson/FSU.  Punch your ticket every year you're not down. 
If you want objectivity, then you need to have equity.
Eh. Not really. In every sport there are tough divisions and easy divisions. 

If the criteria for acceptance is 100% a "committee" decision, then yes it penalizes strong conferences because it's harder to get out of a strong conference running the table or with only 1 loss. That's why a strong conference like the SEC only plays 8 conference games and has a November FCS patsy weekend--because they don't want the chance of picking up another loss. 

Heck, Ohio State in 2017 had Iowa as a cross-division game, so it was just chance they played Iowa. In 2018, they had Purdue as a cross-division game, so it was just chance they played Purdue. If the B1G had an 8-game conference season, there's a 1 in 3 chance each year that they might have replaced that cross-division loss with a win over a MAC or FCS team and in both cases they'd likely have been CFP-bound. 2018 is a certainty, because they'd be undefeated at 13-0. 2017, as a 1-loss team with their only loss being to Oklahoma, they'd get in as a 1-loss conference champ over 11-1 non-champ Alabama.

But alas, in 2017 Ohio State was losing to cross-division Iowa in Iowa City in mid-November when they could have been playing at home vs Youngstown State, much like Alabama was playing at home vs Mercer in mid-November.  Things might have been different if they were playing in Gainesville or Athens on that mid-November Saturday.

But if the criteria for inclusion is "win your conference", then you just need to win your conference. A strong conference will not be penalized, because their champ is auto-bid. If the criteria for at-large selection includes strength of schedule, then tough conferences will have a better chance of getting their worthy non-champs selected than weak conferences. 

Cincydawg

  • Oracle of Piedmont Park
  • Global Moderator
  • Hall of Fame
  • *****
  • Default Avatar
  • Posts: 71545
  • Oracle of Piedmont Park
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #159 on: May 27, 2020, 06:12:28 PM »
The SEC late patsy is usually before a rivalry game, and is slated for that reason.

Not every team does it of course, and if you have one, it means fewer "tune ups" in September.

I'd like to see a max of two patsies a year.

medinabuckeye1

  • Legend
  • ****
  • Default Avatar
  • Posts: 8906
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #160 on: May 27, 2020, 06:33:59 PM »
I'd like to see a max of two patsies a year.
In theory this would be great.  In practice I think it would be VERY difficult to come up with a rule that worked.  

Who is the patsy, P5 Rutgers or G5 Memphis or UCF?  

Kris60

  • All Star
  • ******
  • Default Avatar
  • Posts: 2514
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #161 on: May 27, 2020, 07:23:10 PM »




If you want objectivity, then you need to have equity.
I agree. It’s why all pro sports leagues have objective measures and all college sports (at least that I know of) have some degree of subjectivity.  In the NFL this season teams in the same division will play 14 of their 16 games against common opponents or each other.  

Now take Nebraska and Michigan. Only 6 of their 12 games are against common opponents and they  don’t play each other, and these are teams in the same conference. 

medinabuckeye1

  • Legend
  • ****
  • Default Avatar
  • Posts: 8906
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #162 on: May 28, 2020, 11:50:27 AM »
I agree. It’s why all pro sports leagues have objective measures and all college sports (at least that I know of) have some degree of subjectivity.  In the NFL this season teams in the same division will play 14 of their 16 games against common opponents or each other. 

Now take Nebraska and Michigan. Only 6 of their 12 games are against common opponents and they  don’t play each other, and these are teams in the same conference.
Fundamentally, this was why I was so strongly opposed to the two-SEC BCSNCG back in 2011.  

We all know that Bama won the rematch with LSU and did it in fairly convincing fashion but in the regular season LSU was clearly a LOT more deserving.  Not only did LSU win the H2H with Bama but, in addition to that:
  • The H2H was AT Bama so it isn't like LSU skated by on HFA, and
  • LSU played a MUCH tougher OOC with Oregon (finished 12-2 winning the PAC and the RoseBowl); @WVU (finished 10-3 winning the Orange Bowl) and two scrubs.  Meanwhile Bama played a mediocre PSU (finished 9-4, 6-2 in the B1G) and three scrubs, and  
  • LSU played in the SECCG while Bama relaxed at home.  In that game LSU absolutely throttled a pretty solid UGA team that finished 10-4

In the final pre-bowl poll LSU was a unanimous #1 at 13-0 including wins over:
  • #2 Bama on the road
  • #6 Oregon, neutral site
  • #18 UGA, neutral site
  • #23 WVU, on the road

After LSU it was messy.  There were five one-loss teams:
  • #2 Bama lost at home to LSU by a FG, best win was #24 PSU
  • #3 OkSU lost by 6 at ISU, best win was #19 OU
  • #4 Stanford lost badly at home to Oregon and were NOT PAC Champs because of that tiebreaker, best win was #5 USC
  • #8 Boise lost at home by a point to TCU, best win was #18 UGA
  • #20 Houston lost CUSCCG to USM.  

Stanford was out because they didn't win the PAC.  It was silly that they ended up ranked ahead of Oregon but whatever.  Boise and Houston were out because they didn't have P5 SoS.  

I think that most of us agree that Bama was probably better than OkSU but none of us KNOW that Bama was better than OkSU because the competition was too divergent.  Bama and OkSU played a grand total of zero common opponents.  


Cincydawg

  • Oracle of Piedmont Park
  • Global Moderator
  • Hall of Fame
  • *****
  • Default Avatar
  • Posts: 71545
  • Oracle of Piedmont Park
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #163 on: May 28, 2020, 11:55:18 AM »
In theory this would be great.  In practice I think it would be VERY difficult to come up with a rule that worked. 

Who is the patsy, P5 Rutgers or G5 Memphis or UCF? 
I would term any conference opponent as not being a patsy by definition, even if they are.  It would mean playing 10 P5 level teams a year, and yes, some G5 teams are better than Rutgers.

medinabuckeye1

  • Legend
  • ****
  • Default Avatar
  • Posts: 8906
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #164 on: May 28, 2020, 03:24:53 PM »
I would term any conference opponent as not being a patsy by definition, even if they are.  It would mean playing 10 P5 level teams a year, and yes, some G5 teams are better than Rutgers.
I just don't see that accomplishing much.  Suppose the two teams competing for the fifth spot are two P5 Champions who went 11-2.  We'll call one of them Ohio State and say that they played a nine game conference schedule and two P5 teams OOC while we'll call the other Bama and say that they played an eight game conference schedule and four non-P5 teams OOC.  

Just from that, it sounds like Ohio State played, or at least tried to play a much tougher schedule.  Fine, but what if Ohio State's two P5 OOC opponents were NCST who went 1-7 in the ACC and 4-8 overall and Kansas who went 1-8 in the B12 and 3-9 overall and that Ohio State's three B1G-W cross-overs were Northwestern (1-8/3-9), UNL (3-6/5-7), and Purdue (3-6/4-8) while Bama's four non-P5 OOC were Cincy, UCF, ApSt, and Memphis and their two SEC-E cross-overs were UGA (7-1/12-2) and Florida (6-2/11-2)?  

By your definition Ohio State played a tougher schedule because they played 11 P5 opponents while Bama "only" played seven but Ohio State's 11 P5 opponents included these nine teams that finished sub .500 in their league and were probably all worse than any of Bama's hypothetical G5 OOC opponents:
  • 0-9/2-10 Rutgers
  • 1-8/3-9 Maryland
  • 1-8/3-9 Kansas
  • 1-8/3-9 Northwestern
  • 1-7/4-8 NCST
  • 3-6/4-8 Purdue
  • 3-6/5-7 UNL
  • 4-5/7-6 MSU

I think a rule that required teams to play nine or 10 or whatever P5 opponents would be great for Rutgers, Kansas, and a few other perennial P5 doormats because everybody would be knocking on their door.  

Cincydawg

  • Oracle of Piedmont Park
  • Global Moderator
  • Hall of Fame
  • *****
  • Default Avatar
  • Posts: 71545
  • Oracle of Piedmont Park
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #165 on: May 28, 2020, 03:27:50 PM »
I'd rather a team play 10 P5 level opponents each year than 9, with the 3 being typical pastries, UL-Monroe or whatever.

Schedules are made years in advance and you don't know how good a Rutgers might be in 10 years, but you can be fairly sure East Savannah State will be crap.

It would up the scheduling of a fair number of SEC and ACC programs perforce.

Cincydawg

  • Oracle of Piedmont Park
  • Global Moderator
  • Hall of Fame
  • *****
  • Default Avatar
  • Posts: 71545
  • Oracle of Piedmont Park
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #166 on: May 28, 2020, 03:33:48 PM »
So, right now, today, who are the Helmet teams, by which I mean a team that gets attention because of its name.  I suppose this is a bit like the dooks and barons and kings etc.

I think some are clearly on the list due to a combination of tradition and current performance:

Alabama
Ohio State
Oklahoma
Notre Dame

Then some historic helmet teams who may be slipping a bit, but their tradition aspect keeps them there:

Texas
USC
Michigan

The come a rash of near helmets like Penn State that are holding, perhaps you include Auburn/LSU/UGA, with Tennessee/FSU/Miami slipping, and Clemson who has been climbing steadily but from a lower position.  Anyone else?  Wisconsin?  Washington?  Oregon?  Texas A&M?  

I also think of the list as being programs you'd be excited to see scheduled in the future.  If your team schedules say Kansas, you probably think "Well, meh, at least they are P5.".

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 18847
  • Liked:
Re: The Helmets (and near helmets) rolling 10-year winning percentages
« Reply #167 on: May 28, 2020, 04:12:22 PM »
All of this messy stuff can be bypassed with an equitable framework.  Uniformity in conference size, structure, and scheduling. 
THEN you could have objectivity.
The closest we could get without resorting back to smaller conferences is actually 16-team conferences with 4 pods in each.  You'd play every other team in your conference every 2 years.  So while single-season equity is impossible, it would exist each pair of years.  

Yes, the 24 ACC/Big12 teams would have to pare down to 16.
Notre Dame has to join a conference.
The PAC adds 4, the B10 and SEC add 2 apiece.

If you do the above math, 8 teams will be left out of the ACC/Big12 conglomerate and 8 teams need to be added elsewhere.  While they're certainly not going to be the same 8, it all works out, except ND is bumping someone out.  Plus some of the 'homeless 8' may be replaced by the BYUs and Cincinnatis of the world.  

But yes, a 64-team, 4-conference group of schools would separate from the rest, become its own entity, and produce the most objective national champion possible.
“The Swamp is where Gators live.  We feel comfortable there, but we hope our opponents feel tentative. A swamp is hot and sticky and can be dangerous." - Steve Spurrier

 

Support the Site!
Purchase of every item listed here DIRECTLY supports the site.