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: Best #25

 (Read 10745 times)

FearlessF

  • Hall of Fame
  • *****
  • Posts: 45625
  • Liked:
Re: Best #25
« Reply #70 on: July 04, 2018, 01:33:48 PM »
You’re right.  BayareaBadger is right.  And I’m right.  
Let me see how to put this...we’ve got a player talent bell curve on top of a play-calling bell curve on top of a playing time bell curve.
But a simple thing to take away from it all would be something simple and direct - like if Darren McFadden had 1,200 Carrie in his career, his YPC avg would be lower.  Regardless of play calls or what quarter, etc.  it’s a statistical near-certainty.  

plus a thousand or ten thousand other variables
there are not simple takeaways
"Courage; Generosity; Fairness; Honor; In these are the true awards of manly sport."

bayareabadger

  • Legend
  • ****
  • Default Avatar
  • Posts: 9366
  • Liked:
Re: Best #25
« Reply #71 on: July 04, 2018, 02:31:23 PM »
plus a thousand or ten thousand other variables
there are not simple takeaways
We are splitting this hair right down to the smallest bit, and I'm honestly unsure if we're all not sort of on the same page, but pushing around little elements of minutiae.
There are system factors.
There are defensive focus factors.
There are talent factors.
There are surrounding talent factors. 
There are fatigue factors, the runner in question's first and foremost, but the opponent's as well. 
I'm sure I'm missing some.
I think we can say that, chances are on average, more carries make someone more tired, and tired people are, on average, worse at breaking long runs. This is not all the time. There are some freaks. There are some runners that get lathered up. There are some systems that funnel carries or certain ways or set up certain situations. (There's also a question of the outlier nature of long runs and efficiency, which would totally be fun to break down in a different sort of in the weeds discussion).
My small view. More carries often make it harder to keep a high YPC because of fatigue and situational factors. But not always, owing to talent differences and systems and other oddities. 
Vote Melvin. 
(I always like to look at the YPC stuff for modern triple teams because it's interesting to see how they fit or don't fit the patterns/roles. It's a little harder with old Nebraska because that FB role, and sometimes QB role is less well defined)

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 21807
  • Liked:
Re: Best #25
« Reply #72 on: July 04, 2018, 11:12:51 PM »
My small view. More carries often make it harder to keep a high YPC because of fatigue and situational factors. But not always, owing to talent differences and systems and other oddities.

This is exactly what I've been saying.  Never have I said "always" - hence the repeated citations of a bell curve.
“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: 21807
  • Liked:
Re: Best #25
« Reply #73 on: July 04, 2018, 11:16:44 PM »
Why did Tim Tebow, Chris Leak's "backup", get playing time as a freshman? Because he was used on playcalls that weren't well-suited to Leak's strengths.  Great, but has zero to do with this discussion.
Why did McFadden and Jones occasionally take direct snaps in the "Wild Hog" playcalls instead of snapping the ball to a QB? Because the playcalls called for some skill sets that Arkansas' QB didn't have.  Again, totally irrelevant.  
Why is it so hard to accept that there is no "mean" to be regressed to. Different players get different playcalls. If McFadden was running the same plays that Felix Jones was running, it's conceivable to think that his ypc would actually go UP rather than down.   Here is where you seem 'off'.....everyone has a mean to regress to - but everyone's mean is not the same.  
If McFadden was running the same plays that Felix Jones was running in addition to his workload, his YPC would absolutely have decreased.
If McFadden was running the same plays that Felix Jones was running instead of his workload (ie- they traded places), his YPC would absolutely have increased.
Both of these are most likely true, due to what we know about statistics - bell curves, sample size, and diminishing returns.  
“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: 14563
  • Liked:
Re: Best #25
« Reply #74 on: July 05, 2018, 11:54:46 AM »
If McFadden was running the same plays that Felix Jones was running in addition to his workload, his YPC would absolutely have decreased.
If McFadden was running the same plays that Felix Jones was running instead of his workload (ie- they traded places), his YPC would absolutely have increased.
Both of these are most likely true, due to what we know about statistics - bell curves, sample size, and diminishing returns.  
It isn't a question of statistics. 
You are making the assertion that more carries leads to a back being more tired and worn down, and thus their ypc will go down as carried go up.
I am making the assertion that a back's ypc are more due to individual talent level and usage in a system. I am also making the assertion that your typical "workhorse" back is used in a way that will naturally have more carries and lower ypc than a change of pace / 3rd down back. 
I am not denying that it's possible to over-work a player to the point that fatigue may make that player less effective. But I am denying that this is the reason that those backs with really high numbers of carries have lower ypc. 

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 21807
  • Liked:
Re: Best #25
« Reply #75 on: July 05, 2018, 02:40:39 PM »
Here, I'll do a little study, posting the stats here, not knowing the result beforehand, ok?  
I'll look at the collegefootballreference all-time leading rushers, by carries.  I thought this was shown by the previous study I did about highest YPC at each 100-carry interval level, but I guess not.  

Anyway, it lists the top 253 players by carries in a career.  I'll take 3 groups (high on the list, middle, low) and compare their YPC averages.  Keeping in mind the better a RB is, the more likely he is to get more carries, I believe the lowest group will have the highest average YPC, because they had many fewer carries.

I will also do the reverse - look at the top 250 players by YPC in a career.  A high, middle, and low group - predicting the high group, by contrast, will have fewer carries on average than the low group.

Hopefully this will be deemed fair and persuasive.  And hopefully it shows what I think it will, lol.
“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: 14563
  • Liked:
Re: Best #25
« Reply #76 on: July 05, 2018, 02:45:39 PM »
Hopefully this will be deemed fair and persuasive.  And hopefully it shows what I think it will, lol.
I think this is a great idea.
One thing I would do to change it. Instead of breaking it down into "high, medium, low" groups, just put it in Excel and do a scatter plot. And then see if you do a scatter plot, if you can get Excel to actually perform the statistical analysis, give you a linear regression line, and actually come up with the R-value of the correlation strength.

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 21807
  • Liked:
Re: Best #25
« Reply #77 on: July 05, 2018, 02:48:14 PM »
Uhh, you can do that.  I'm just going to create groups and find averages.  I believe the averages will be different enough to show I'm right.

I'll link the pages the lists come from if you want to do all that.
« Last Edit: July 05, 2018, 02:50:52 PM by OrangeAfroMan »
“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: 21807
  • Liked:
Re: Best #25
« Reply #78 on: July 05, 2018, 02:50:12 PM »
I'm going to avoid the top 10 on either list, as they're the tip of the bell curve.  The bottom end is not, however, as it's only the top 250 out of thousands of rushers college football has had.  I hope you acknowledge this as prudent.

https://www.sports-reference.com/cfb/leaders/rush-yds-per-att-player-career.html

https://www.sports-reference.com/cfb/leaders/rush-att-player-career.html

« Last Edit: July 05, 2018, 02:51:59 PM by OrangeAfroMan »
“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: 14563
  • Liked:
Re: Best #25
« Reply #79 on: July 05, 2018, 02:56:05 PM »
The top rushing attempts list doesn't also contain the yards gained, and the top ypc list doesn't actually contain the number of carries...

How are you going to get all that data together?

If you can do the cross-reference and get the raw data into a spreadsheet, I can do the other calculations/graphs I mentioned. 

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 21807
  • Liked:
Re: Best #25
« Reply #80 on: July 05, 2018, 03:02:08 PM »
The top rushing attempts list doesn't also contain the yards gained, and the top ypc list doesn't actually contain the number of carries...

How are you going to get all that data together?

If you can do the cross-reference and get the raw data into a spreadsheet, I can do the other calculations/graphs I mentioned.
I'm going to click their individual pages - you can do that when you're doing a quick 'n dirty study like I am.  :13:
“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: 14563
  • Liked:
Re: Best #25
« Reply #81 on: July 05, 2018, 03:04:42 PM »
I'm going to click their individual pages - you can do that when you're doing a quick 'n dirty study like I am.  :13:
Any chance your quick 'n dirty study will involve you clicking those pages and then putting the numbers into an Excel spreadsheet? ;-)

OrangeAfroMan

  • Stats Porn
  • Hall of Fame
  • *****
  • Posts: 21807
  • Liked:
Re: Best #25
« Reply #82 on: July 05, 2018, 03:07:23 PM »
I'll do three 20-player groups, with midpoints at rankings 240, 130, and 20.  
“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: 21807
  • Liked:
Re: Best #25
« Reply #83 on: July 05, 2018, 03:08:16 PM »
Any chance your quick 'n dirty study will involve you clicking those pages and then putting the numbers into an Excel spreadsheet? ;-)
I'll type them out here: YPC - player - Car for the one and then Car - player - YPC for the other and conclusions.
“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.