David Shaw: CFP should look at full schedules

[4down20]

Are you going to try and tell us that playing games at home makes Bama's schedule more difficult???

Didn't want to touch that question huh?

Oregon has played in 15 games in a single season once, how many times has Washington done it?

Home games make money.

 

[7Samurai13]

Mizzou has two SEC East titles and an SEC bankroll for coaches, facilities, etc..... not to mention it's a fvking home & away series. Better yet why are you posting in the CFF if you don't have a clue.

And Baylor has two actual conference titles in the same time period, does that mean that in 2018 Baylor is a quality win? Keep focusing on the past it doesn’t change the present though. I don’t think any of the players on this team were a part of those division titles. What is relevant is this team is the one that dropped their ankles and bent over whenever they faced a team with a pulse.

 

[Deep Creek]

If they went to 9 conference games they would still play an FCS team, just like nearly every Big12 team and 2/3s of the PAC does.

Accurate statement. 17 of 22 play FCS teams.

Big 12 - Abilene Christian, South Dakota State, Nichols, South Dakota, Missouri State, Southern, Lamar, Younstown State
PAC - Idaho State, Portland State, Southern Utah (2), UC Davis, North Dakota, Eastern Washington, New Hampshire, Weber State

Big 12 and PAC schools NOT playing an FCS school. Oklahoma, Texas, Arizona State, UCLA, USC. (West Virginia plays Youngstown but also plays 2 P5s in NC State and Tennessee.)

Southern Utah cashing some checks.

 

[theboardref]

So playing a bunch of 7-6 teams = greater SOS... interesting.

 

[theboardref]

Accurate statement. 17 of 22 play FCS teams.

Big 12 - Abilene Christian, South Dakota State, Nichols, South Dakota, Missouri State, Southern, Lamar, Younstown State
PAC - Idaho State, Portland State, Southern Utah (2), UC Davis, North Dakota, Eastern Washington, New Hampshire, Weber State

Big 12 and PAC schools NOT playing an FCS school. Oklahoma, Texas, Arizona State, UCLA, USC. (West Virginia plays Youngstown but also plays 2 P5s in NC State and Tennessee.)

Southern Utah cashing some checks.

Their baseball team came out to Alex Box a couple years back, scrappy squad.

 

[nddulac]

LOL

Take away Duke's highest ranked FBS OOC game last year and you replace Northwestern. A game they won.

Instead they get Clemson, a game they would have lost.

Now Duke is 5-7 in the regular season instead of 6-6 and they don't get the chance to beat NIU in the Quick Lane Bowl to finish 7-6 and their ranking is lower in your "rankings" which lowers Clemson's ranking.

Again, this is apples and oranges. Of course out of conference competition can have an effect on average conference rating (in fact, it is the only factor that does.) The conference game would have no effect on the average conference rating or strength of schedule.

Now - I agree with you that the place that 9th conference game hurts is the at the bottom of the bowl-eligible list of teams. So, I don't dispute your analysis on the effect that that extra conference game can have on a Duke. However, that was not the point I was making (or refuting) in the post to which you responded. The point of that post was specifically that playing an extra conference game, as an isolated variable, would have the effect of raising the conference's average strength of schedule.

 

[nddulac]

Except the majority of those teams stay on the schedule every fucking year anyway dipshit. The majority of the strength in the SEC is in the SEC West. The majority of the strength of the Big10 is in the Big10 East.

Jimmie status: rustled

If you are too stupid to understand that the conference average is different from division to division in each conference as a result of the teams being rotated being different, then you are even dumber than I thought.

The conference average is different from division to division? Dude - do you even English? This is perhaps the most nonsensical comment i have ever seen you make on the Hoop.

The division average is different from division to division. The conference average includes both divisions. It's ridiculous that that even needs to be said.

There is no such thing as "conference average" that applies when the rotations is only a few teams while the majority of the schedule is set with the same teams every year.

You don't get to redefine things on the fly. As I defined the values I used in my calculation, everyting is self-consistent and correct. Your argument is now reducing to tell me that 2 + 2 = 5 . . because that second 2 is actually a 3!

Now - if you want to dispute my calculation, using a different methodology, be my guest. Show me your calculation. (You won't do that, because unless you fuck it up, it is going to produce the same result, and you'll end up looking stupid.)

Indeed.

 

[4down20]

Jimmie status: rustled


The conference average is different from division to division? Dude - do you even English? This is perhaps the most nonsensical comment i have ever seen you make on the Hoop.

The division average is different from division to division. The conference average includes both divisions. It's ridiculous that that even needs to be said.

You don't get to redefine things on the fly. As I defined the values I used in my calculation, everyting is self-consistent and correct. Your argument is now reducing to tell me that 2 + 2 = 5 . . because that second 2 is actually a 3!

Now - if you want to dispute my calculation, using a different methodology, be my guest. Show me your calculation. (You won't do that, because unless you fuck it up, it is going to produce the same result, and you'll end up looking stupid.)

Indeed.

The simple fact of the matter is the average strength applied to the extra game is going to be higher for Alabama than Georgia. Your method just flat out ignores this fact and acts like it exists.

Kind of like shitty SoS formulas that pretend all 8-5 teams are equal. The math may be "correct", but it's an ignorant way of doing it. At least the people who know the methods have the balls to admit such methods are extremely flawed, you on other hand refuse to acknowledge such flaws.

And you knew that I meant division averages, if you had to deal with the amount of alerts I deal with on a daily basis you'd make similar mistakes. Yet the fact you choose to hang on to that says a lot.

And I won't do it because - I have other shit to do, and I don't have rankings and shit to get data from to start with and I'm not wasting all that time. But I notice you did only the ACC, is there a reason you won't do the Pac12 with the same method?

 

[droider]

David Shaw is a moron and one of the most overrated coaches in the game. I always LMFAO when NFL beat writers try hyping him as a potential great coach in the league. All because he's black, it's a joke. The guy hasn't won anything important and hasn't innovated the game in any way.

 

[AlaskaGuy]

David Shaw is a moron and one of the most overrated coaches in the game. I always LMFAO when NFL beat writers try hyping him as a potential great coach in the league. All because he's black, it's a joke. The guy hasn't won anything important and hasn't innovated the game in any way.

 

[AlaskaGuy]

 

[outofyourmind]

Wait
All 69 offers are real offers?
Did I get that right

 

[AlaskaGuy]

Wait
All 69 offers are real offers?
Did I get that right

A ranking of the hardest scholarship offers to get among Power 5 schools - FootballScoop

 

[outofyourmind]

A ranking of the hardest scholarship offers to get among Power 5 schools - FootballScoop

That is quite a range there.

 

[nddulac]

The simple fact of the matter is the average strength applied to the extra game is going to be higher for Alabama than Georgia. Your method just flat out ignores this fact and acts like it exists.

Two things. I did not run my analysis on the SEC. Second, I did not report any team's changed individual strength of schedule. (Hint: I didn't do that because it wouldn't be accurate.) What I did report was the change in the conference average strength of schedule.

So let me demonstrate why my method doesn't change the result for the average. Suppose I am taking the average of four numbers: 3, 5, 7, and 13. The average is 7. Now = suppose I take the average of a different set of four numbers: 7, 7, 7, and 7. Guess what the average is! If you said "7", you get a cookie.

So sure, the individual numbers moved, but the average I calculate at the end remains exactly the same.

Now - If you disagree, or think I have pulled some crazy mathematical slight of hand, please feel free to show the results of your own calculation. Substitute whichever team you wish for a conference opponent and calculate how the strength of schedule changes. Then take the average over the entire conference. Then, just for fun, do the calculation again substituting the same team as before with the average conference rating. then calculate the average conference strength of schedule and compare. You may use any rating system you wish (since you don't trust my numbers) and any conference you wish. Hell - use the ordinals off the Massey composite if you like. You can assign every FCS team the same rating - your pick! The only hard requirement is that you define strength of schedule as the average of the ratings a team plays.

Go!

And you knew that I meant division averages, if you had to deal with the amount of alerts I deal with on a daily basis you'd make similar mistakes. Yet the fact you choose to hang on to that says a lot.

I kinda figured something was up. That was, after all, an imbecilic comment to begin with. Next time I will try to take into account that I can't expect what you wrote to represent what you meant.

And I won't do it because - I have other shit to do, and I don't have rankings and shit to get data from to start with and I'm not wasting all that time. But I notice you did only the ACC, is there a reason you won't do the Pac12 with the same method?

I only did the ACC because it took some time to pull and analyze the data. But if you wish to save time, use a fictional set of data and a smaller "conference". It doesn't matter. The math will be the same if you substitute the conference average for a team, or a unique conference opponent for each team.

I still don't believe you will do the calculation because you know that it will show that I am correct.

 

[4down20]

Two things. I did not run my analysis on the SEC. Second, I did not report any team's changed individual strength of schedule. (Hint: I didn't do that because it wouldn't be accurate.) What I did report was the change in the conference average strength of schedule.

So let me demonstrate why my method doesn't change the result for the average. Suppose I am taking the average of four numbers: 3, 5, 7, and 13. The average is 7. Now = suppose I take the average of a different set of four numbers: 7, 7, 7, and 7. Guess what the average is! If you said "7", you get a cookie.

So sure, the individual numbers moved, but the average I calculate at the end remains exactly the same.

Now - If you disagree, or think I have pulled some crazy mathematical slight of hand, please feel free to show the results of your own calculation. Substitute whichever team you wish for a conference opponent and calculate how the strength of schedule changes. Then take the average over the entire conference. Then, just for fun, do the calculation again substituting the same team as before with the average conference rating. then calculate the average conference strength of schedule and compare. You may use any rating system you wish (since you don't trust my numbers) and any conference you wish. Hell - use the ordinals off the Massey composite if you like. You can assign every FCS team the same rating - your pick! The only hard requirement is that you define strength of schedule as the average of the ratings a team plays.

Go!


I kinda figured something was up. That was, after all, an imbecilic comment to begin with. Next time I will try to take into account that I can't expect what you wrote to represent what you meant.


I only did the ACC because it took some time to pull and analyze the data. But if you wish to save time, use a fictional set of data and a smaller "conference". It doesn't matter. The math will be the same if you substitute the conference average for a team, or a unique conference opponent for each team.

I still don't believe you will do the calculation because you know that it will show that I am correct.

Do you understand that by default, Alabama will have a tougher conference schedule than Georgia on average even though they are in the same conference?

Also, you won't do the Pac12, but you say 1 conference is the proof. You should at minimum do a 2nd conference to see if the results differ. According to you, the results would be the same apparently.

 

[4down20]

A ranking of the hardest scholarship offers to get among Power 5 schools - FootballScoop

A ranking of the laziest college football staffs is what that is.

 

[nddulac]

Do you understand that by default, Alabama will have a tougher conference schedule than Georgia on average even though they are in the same conference?

Yes. Do you understand that it would not affect the conference average (which is the only thing I reported)?

Once again, you don't have to trust my math. You're a smarty pants - you can do the math for yourself. But you continue to choose not to. Gutsy gambit.

Also, you won't do the Pac12, but you say 1 conference is the proof. You should at minimum do a 2nd conference to see if the results differ. According to you, the results would be the same apparently.

Oh - so now you are going from "your math is incorrect" to "your conclusion is incorrect." Maybe I should restate my conclusion: For any conference whose average strength of schedule for out-of-conference games is higher than the average rating of teams in the conference, the average SOS will increase if they substitute an out-of-conference opponent for an additional conference game.

The reason it would make little sense to apply the analysis to the Pac-12 is because it would require moving them from 9 conference games to 10 - a move for which no one has advocated.

 

[4down20]

Yes. Do you understand that it would not affect the conference average (which is the only thing I reported)?

Once again, you don't have to trust my math. You're a smarty pants - you can do the math for yourself. But you continue to choose not to. Gutsy gambit.

Oh - so now you are going from "your math is incorrect" to "your conclusion is incorrect." Maybe I should restate my conclusion: For any conference whose average strength of schedule for out-of-conference games is higher than the average rating of teams in the conference, the average SOS will increase if they substitute an out-of-conference opponent for an additional conference game.

The reason it would make little sense to apply the analysis to the Pac-12 is because it would require moving them from 9 conference games to 10 - a move for which no one has advocated.

It shows that the conference average doesn't matter because it's only 1 division that each team gets rotated in. Nevermind the fact that Alabama doesn't play Alabama.

The entire point is the conference average doesn't actually apply, and yet all you want me to do is find a conference average.

And I'm not doing the math because I'm too busy working on stuff I care about. Which right now is AI, and getting my damn units to retreat when they are outnumbered. Apparently gonna have to add in a little memory because it retreats to 1 point, moves towards that point, decides well there is better point over there in the other direction, and moves back to the original point and guess what happens? Yep, decides the previous point is best again and moves back and forth until it dies. And the best part is in the end it's all for nothing because eventually a neural net will be making the decisions, but right now I'm focusing on training an economic advisor and they need somewhat competent units for learning what to do.

Want to help me with that math? It's quite fun until they get caught in an infinite loop. I'm thinking a bias towards the previous decision will solve the problem.

Also, since you are merely doing conference averages there is no reason you can't see the difference from 8 to 9 game schedules inside that conference. Come on now.

 

[nddulac]

It shows that the conference average doesn't matter because it's only 1 division that each team gets rotated in. Nevermind the fact that Alabama doesn't play Alabama.

True - Alabama does not play Alabama. But someone plays Alabama, so their rating will figure into the conference strength of schedule for that game. Get it yet?

The entire point is the conference average doesn't actually apply, and yet all you want me to do is find a conference average.

Yes - because that's what you said I did incorrectly.