Author
Topic: Will the CFP Committee be influenced by past year results?
(Read 6224 times)

Kris60 · « **Reply #42 on:** February 06, 2020, 09:17:42 PM »

There may be an element of truth to that but I also think there’s an element of truth to the randomness of turnovers. If there’s a way to coach it even arguably the greatest coach in CFB history seemingly hasn’t found the secret to be consistently successful doing it. Since 2009, this is Bama’s ranking in turnovers caused:

2009- 10th
2010- 30th
2011- 77th
2012- 20th
2013- 81st
2014- 65th
2015- 13th
2016- 5th
2017- 29th
2018- 45th
2019- 4th

If I had the time to do it for all schools my guess is we would see very similar results in turnovers. Since Saban has been there Bama has been much more consistent in where they finish in the polls, in total offense, and in total defense. Turnovers has been all over the board.

OrangeAfroMan · « **Reply #43 on:** February 06, 2020, 09:53:54 PM »

Sure, possibly. But you've, I assume, inadvertently chosen the team with the most year-to-year turnover in defensive players. It's easy to see your list and just chalk it up to randomness, but I'm simply saying "not so fast, my friend".
.
Shouldn't we study something before we give up on it?
.
That being said, 'turnovers' is too vague. There's fumbles and interceptions. Those could (and should, but aren't) be further broken down into forced fumbles (on a tackle), unforced fumbles (guy just goofs the ball out of his own hands), tipped INTs, wayward throw INTs, miscommunication-on-the-route INTs, and then traditional INTs.
.
If we had those stats handy and have for decades that we bandied about, then the turnovers idea becomes a lot less likely to be random. Or if it was random, we'd know. But we simply don't have the stats breakdown like we do for yards and points and completion %, etc.
.
If early fans and stat-keepers were obsessed with turnovers from the start, we'd have a handle on it.

OrangeAfroMan · « **Reply #44 on:** February 06, 2020, 09:58:31 PM »

And it's important to keep in mind that statistically, if something truly is random, then there's a sort of universal baseline for it and it can be set aside. We can then focus on what isn't random and study that. Teams and coaches could try to improve on the portions that weren't random.
.
This all reminds me of the problem of quantifying defense in baseball. You start with x% of baseball defense being pitching and y% being fielding. Then you break it down further to catcher pitch framing, double plays, and range factor. Many throw their hands up and give up on that task as well, but I don't.

Kris60 · « **Reply #45 on:** February 07, 2020, 06:48:47 AM »

But there’s even a randomness to something like a forced fumble. I see ball carriers get rocked all the time and not fumble the ball. Then they are running through a crowd and a defender just happens to throw his hand out and inadvertently punch the ball at the exact split second the ball carrier doesn’t have it tucked away just right.

You couldn’t replicate that happening again if you tried 100 times. So even if you had statistics on something like which teams actually “force” more fumbles I’m not sure what you could take away from it.

bayareabadger · « **Reply #46 on:** February 07, 2020, 09:07:13 AM »

Quote from: OrangeAfroMan on February 06, 2020, 08:24:22 PM

Well of course, you're more confident in predicting the overall quality of a team more than it's turnover ability because you've been predicting and paying attention to the former since you were a kid and you haven't really studied the latter at all.
.
When it comes to turnovers, are any of us an expert? Is the DC the same as last year? How many starters returning? Good pass rush? Great CB or Safety? Is the CB good enough to play really well but not so elite that QBs avoid him? Do the players go for big hits or wrap up? Do they have unique blitz packages? Are the 4 new starters influencing what the DC feels he should call in certain downs and distances? Do the DBs tend to get their head around? How's their cardio - do they trail plays full-go or barely jogging?
.
Just because these specific aspects aren't in all the preseason mags, I'm not going to shrug and say we can't know these things. And to be honest, we don't have to know them, just acknowledge they exist and have some confidence that these aspects add up to something tangible and quite the opposite of random.

I've made up for this by reading other people who actually study it. And generally they say teams with high turnover numbers struggle to maintain and all those number tend to regress.

I don't know what the middle graph is. It's basically an explanation that the minutia of football is random, which it is. But we're trying to look past stacking up teams by loss just a little, so we can't let the hyper-subjective details (and very possibly meaningless ones) get in the way.

bayareabadger · « **Reply #47 on:** February 07, 2020, 09:09:17 AM »

Quote from: OrangeAfroMan on February 06, 2020, 09:53:54 PM

Sure, possibly. But you've, I assume, inadvertently chosen the team with the most year-to-year turnover in defensive players. It's easy to see your list and just chalk it up to randomness, but I'm simply saying "not so fast, my friend".
.

If "year-to-year turnover in defensive players" is a factor ... this is college football, and thus the the year-to-year turnover is by nature high everywhere. And most places don't have the same defensive architect in place throughout.

bayareabadger · « **Reply #48 on:** February 07, 2020, 09:11:27 AM »

Quote from: OrangeAfroMan on February 06, 2020, 06:49:31 PM

Who recovers fumbles tends to be random, but no, turnovers themselves (or the number of them) are certainly not random.
Utah State is a good example - they've been among the national leaders in turnovers the past 3 years. Syracuse, UCF, among others seem to emphasize causing turnovers.
.
The idea that all turnovers are simply random makes you wonder why coaches spend time on tip drills and stripping ball-carriers or even bother with zone blitzes and disguised coverages.
.

I like this site because we're good at all syncing into asides and we generally enjoy doing it.

But to get back to the start for a moment. If something is important on defense, so important it should factor into playoff choice, and the best examples are defenses from Utah State, Syracuse and UCF (natty champs!!) then I think we can safely say it doesn't need to be in there.

Cincydawg · « **Reply #49 on:** February 07, 2020, 11:48:55 AM »

There are random turnovers, and caused turnovers. The random kind occur out of the blue and are not caused intentionally and any trend cannot be predicted.

Kris60 · « **Reply #50 on:** February 07, 2020, 12:20:03 PM »

Quote from: Cincydawg on February 07, 2020, 11:48:55 AM

There are random turnovers, and caused turnovers. The random kind occur out of the blue and are not caused intentionally and any trend cannot be predicted.

Well, and like I mentioned in another post, I think even “caused” turnovers have an aspect of randomness about them.

OrangeAfroMan · « **Reply #51 on:** February 07, 2020, 12:45:46 PM »

Quote from: bayareabadger on February 07, 2020, 09:09:17 AM

If "year-to-year turnover in defensive players" is a factor ... this is college football, and thus the the year-to-year turnover is by nature high everywhere. And most places don't have the same defensive architect in place throughout.

Yes, there is a baseline of player turnover, but programs that have an inordinate number of players leaving early are going to have the highest turnover.

OrangeAfroMan · « **Reply #52 on:** February 07, 2020, 12:48:52 PM »

Quote from: Kris60 on February 07, 2020, 06:48:47 AM

But there’s even a randomness to something like a forced fumble. I see ball carriers get rocked all the time and not fumble the ball. Then they are running through a crowd and a defender just happens to throw his hand out and inadvertently punch the ball at the exact split second the ball carrier doesn’t have it tucked away just right.

This is absolutely correct, but you need to look at the process in order to gain anything from the outcome.
.
Teams that "rock the ball-carrier" more often are going to cause more fumbles, even with your post being accurate. If an outcome only happens 3% of the time and that's a set number, when is it more likely to happen more times - when you do the action 15 times or 50?
Now, the 3% chance thing that occurs MIGHT happen more times in the sample of 15 tries, but it's far more likely to happen more times in the sample of 50 tries.

OrangeAfroMan · « **Reply #53 on:** February 07, 2020, 12:50:55 PM »

Quote from: bayareabadger on February 07, 2020, 09:11:27 AM

I like this site because we're good at all syncing into asides and we generally enjoy doing it.

But to get back to the start for a moment. If something is important on defense, so important it should factor into playoff choice, and the best examples are defenses from Utah State, Syracuse and UCF (natty champs!!) then I think we can safely say it doesn't need to be in there.

I never claimed it was a major thing, but yeah, if I'm picking between 2 teams and one causes 3 turnovers per game and the other causes 1 per game, it's simply a check in a box of the 3 turnover team. Not a deciding factor, but one of many items that comprise the percieved quality of a team.

FearlessF · « **Reply #54 on:** February 07, 2020, 01:02:44 PM »

I'm jumping into this discussion only randomly because of the Huskers in ability to cause/force turnovers has been statistically poor through Pelini, Rielly, and now Frost.

and obviously there have been many more Defensive coordinators over this span

randomly a statistical anomaly to be statistically poor this long

Kris60 · « **Reply #55 on:** February 08, 2020, 09:24:08 AM »

Quote from: OrangeAfroMan on February 07, 2020, 12:48:52 PM

This is absolutely correct, but you need to look at the process in order to gain anything from the outcome.
.
Teams that "rock the ball-carrier" more often are going to cause more fumbles, even with your post being accurate. If an outcome only happens 3% of the time and that's a set number, when is it more likely to happen more times - when you do the action 15 times or 50?
Now, the 3% chance thing that occurs MIGHT happen more times in the sample of 15 tries, but it's far more likely to happen more times in the sample of 50 tries.

Lol. Fair enough. I’ll roll over a little. Turnovers may not be quite as random as I’ve always perceived...

...but they are still pretty random.

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.

Topic: Will the CFP Committee be influenced by past year results?

Kris60

Re: Will the CFP Committee be influenced by past year results?

OrangeAfroMan

Re: Will the CFP Committee be influenced by past year results?

OrangeAfroMan

Re: Will the CFP Committee be influenced by past year results?

Kris60

Re: Will the CFP Committee be influenced by past year results?

bayareabadger

Re: Will the CFP Committee be influenced by past year results?

bayareabadger

Re: Will the CFP Committee be influenced by past year results?

bayareabadger

Re: Will the CFP Committee be influenced by past year results?

Cincydawg

Re: Will the CFP Committee be influenced by past year results?

Kris60

Re: Will the CFP Committee be influenced by past year results?

OrangeAfroMan

Re: Will the CFP Committee be influenced by past year results?

OrangeAfroMan

Re: Will the CFP Committee be influenced by past year results?

OrangeAfroMan

Re: Will the CFP Committee be influenced by past year results?

FearlessF

Re: Will the CFP Committee be influenced by past year results?

Kris60

Re: Will the CFP Committee be influenced by past year results?