An Interesting Take on 2020

User avatar
hapapp
Posts: 14648
Joined: Tue Jun 27, 2000 12:48 pm
School: Appalachian State
Location: Rocky Mount, VA
Has thanked: 1600 times
Been thanked: 1505 times

An Interesting Take on 2020

Unread post by hapapp » Sat Feb 08, 2020 7:30 pm




Appstate88
Posts: 1486
Joined: Mon Dec 24, 2012 9:38 am
School: Appalachian State
Has thanked: 284 times
Been thanked: 636 times

Re: An Interesting Take on 2020

Unread post by Appstate88 » Sat Feb 08, 2020 7:51 pm

Beach Chickens could be a threat in the East Division.



User avatar
Rekdiver
Posts: 4915
Joined: Mon Aug 13, 2012 1:14 pm
School: Appalachian State
Has thanked: 723 times
Been thanked: 1952 times

Re: An Interesting Take on 2020

Unread post by Rekdiver » Sat Feb 08, 2020 8:43 pm

Worthless stats....



User avatar
hapapp
Posts: 14648
Joined: Tue Jun 27, 2000 12:48 pm
School: Appalachian State
Location: Rocky Mount, VA
Has thanked: 1600 times
Been thanked: 1505 times

Re: An Interesting Take on 2020

Unread post by hapapp » Sun Feb 09, 2020 6:09 am

They are not necessarily predictive but they do speak of team's level of experience which is not worthless. However, while a team can be losing a number of starters it doesn't speak of the depth at those positions. The way we rotate on D, hopefully, mitigates the loss of so many starters. Though, it may be unrealistic to think initially we can be as strong as this year.



Yosef84
Posts: 2362
Joined: Thu Aug 09, 2012 11:27 am
Has thanked: 307 times
Been thanked: 1071 times

Re: An Interesting Take on 2020

Unread post by Yosef84 » Sun Feb 09, 2020 9:03 am

Lots of returning production from a 5-7 team is probably an indicator that they have a good shot at making a bowl this year but I think it's a jump to say they are a threat just yet. Never know though. Any team is a threat if we don't show up and play our game.



EastHallApp
Posts: 5672
Joined: Wed Nov 06, 2013 4:34 pm
School: Appalachian State
Location: Raleigh
Has thanked: 2548 times
Been thanked: 2063 times

Re: An Interesting Take on 2020

Unread post by EastHallApp » Sun Feb 09, 2020 9:18 am

hapapp wrote:
Sun Feb 09, 2020 6:09 am
They are not necessarily predictive but they do speak of team's level of experience which is not worthless. However, while a team can be losing a number of starters it doesn't speak of the depth at those positions. The way we rotate on D, hopefully, mitigates the loss of so many starters. Though, it may be unrealistic to think initially we can be as strong as this year.
I think this is a good take. I think we have plenty of experienced talent to fill in for those departures (though I have some concern about the NT spot), but it’s not fair to expect there to be no drop off from guys like ADG and Fehr right away. OTOH we return all our corners and DEs plus Chris Willis, so that helps.

Then of course having a new DC and entirely new defensive staff is another variable we can’t account for.

In any case, the flip side is that the offense should hit the ground running, so hopefully we can out score a couple teams early if we need to.



AppFan11
Posts: 543
Joined: Mon Sep 03, 2012 10:22 pm
School: Appalachian State
Has thanked: 111 times
Been thanked: 126 times

Re: An Interesting Take on 2020

Unread post by AppFan11 » Sun Feb 09, 2020 10:23 am

Actually that’s what Bill Connolly has done for years, to make stats that are more predictive than the traditional stats. He fully acknowledges that some are better than others. Certainly Phil Steele uses returning production numbers as well. I agree with Hap and East that in our unique situation on defense where we are constantly rotating players so that younger players are getting regular snaps, that Bill’s stats will underestimate our performance. As we know, predictive models generalize predictions by whatever formula is being used and does not take into account specific/unique situations. All that being said, I would expect that 2020 defense to be good overall but probably start slow and get better and better every game.



AppStFan1
Posts: 2035
Joined: Mon Dec 03, 2018 10:37 am
School: Appalachian State
Has thanked: 70 times
Been thanked: 389 times

Re: An Interesting Take on 2020

Unread post by AppStFan1 » Sun Feb 09, 2020 12:16 pm

AppFan11 wrote:
Sun Feb 09, 2020 10:23 am
Actually that’s what Bill Connolly has done for years, to make stats that are more predictive than the traditional stats. He fully acknowledges that some are better than others. Certainly Phil Steele uses returning production numbers as well. I agree with Hap and East that in our unique situation on defense where we are constantly rotating players so that younger players are getting regular snaps, that Bill’s stats will underestimate our performance. As we know, predictive models generalize predictions by whatever formula is being used and does not take into account specific/unique situations. All that being said, I would expect that 2020 defense to be good overall but probably start slow and get better and better every game.
Agree with you guys. That chart is interesting and it matches what you expect that CCU should get better. Chadwell is a good coach so they should be a bowl team and I would not be shocked if they finish in 2nd or 3rd place in the division. I'm with you that early on the defense may start slow but I expect them to be dominant by week 5. Just like we said last year when new coaches came in. We have a lot of talent so the expectation is a 2020 SBC Championship. My worry is 2021 when we have a new QB and replace a ton of production on offense.



User avatar
AtlAppMan
Posts: 1102
Joined: Fri Aug 10, 2012 3:23 pm
School: Appalachian State
Location: ATL
Has thanked: 4 times
Been thanked: 404 times

Re: An Interesting Take on 2020

Unread post by AtlAppMan » Sun Feb 09, 2020 12:59 pm

The idea and some value for returning productivity is interesting. However is 70% returning from a13-1 team better or worse than 90% from a 5-7 team? And what is the relative new value of the players that will fill in the lost production? Are you replacing it with less, equal or better talent? Look at Zac replacing Lamb two years ago, we got better day one. One can’t automatically assume replacing a starter is net negative productivity.

Just some points to ponder.



EastHallApp
Posts: 5672
Joined: Wed Nov 06, 2013 4:34 pm
School: Appalachian State
Location: Raleigh
Has thanked: 2548 times
Been thanked: 2063 times

Re: An Interesting Take on 2020

Unread post by EastHallApp » Sun Feb 09, 2020 1:52 pm

AtlAppMan wrote:
Sun Feb 09, 2020 12:59 pm
The idea and some value for returning productivity is interesting. However is 70% returning from a13-1 team better or worse than 90% from a 5-7 team? And what is the relative new value of the players that will fill in the lost production? Are you replacing it with less, equal or better talent? Look at Zac replacing Lamb two years ago, we got better day one. One can’t automatically assume replacing a starter is net negative productivity.

Just some points to ponder.
Exactly. There are some spots where I think you have to assume some dropoff, at least in the short run. But there are also a couple places (I won’t name names) where I think the new starter could be an upgrade.



User avatar
hapapp
Posts: 14648
Joined: Tue Jun 27, 2000 12:48 pm
School: Appalachian State
Location: Rocky Mount, VA
Has thanked: 1600 times
Been thanked: 1505 times

Re: An Interesting Take on 2020

Unread post by hapapp » Sun Feb 09, 2020 3:24 pm

AppStFan1 wrote:
Sun Feb 09, 2020 12:16 pm
AppFan11 wrote:
Sun Feb 09, 2020 10:23 am
Actually that’s what Bill Connolly has done for years, to make stats that are more predictive than the traditional stats. He fully acknowledges that some are better than others. Certainly Phil Steele uses returning production numbers as well. I agree with Hap and East that in our unique situation on defense where we are constantly rotating players so that younger players are getting regular snaps, that Bill’s stats will underestimate our performance. As we know, predictive models generalize predictions by whatever formula is being used and does not take into account specific/unique situations. All that being said, I would expect that 2020 defense to be good overall but probably start slow and get better and better every game.
Agree with you guys. That chart is interesting and it matches what you expect that CCU should get better. Chadwell is a good coach so they should be a bowl team and I would not be shocked if they finish in 2nd or 3rd place in the division. I'm with you that early on the defense may start slow but I expect them to be dominant by week 5. Just like we said last year when new coaches came in. We have a lot of talent so the expectation is a 2020 SBC Championship. My worry is 2021 when we have a new QB and replace a ton of production on offense.
I wonder if it takes into account the number of folks who have left the Coastal program. I'm not clued in enough to know if their losses were backups or starters. But, I believe they had a significant number in the portal.



User avatar
hapapp
Posts: 14648
Joined: Tue Jun 27, 2000 12:48 pm
School: Appalachian State
Location: Rocky Mount, VA
Has thanked: 1600 times
Been thanked: 1505 times

Re: An Interesting Take on 2020

Unread post by hapapp » Sun Feb 09, 2020 3:33 pm

AtlAppMan wrote:
Sun Feb 09, 2020 12:59 pm
The idea and some value for returning productivity is interesting. However is 70% returning from a13-1 team better or worse than 90% from a 5-7 team? And what is the relative new value of the players that will fill in the lost production? Are you replacing it with less, equal or better talent? Look at Zac replacing Lamb two years ago, we got better day one. One can’t automatically assume replacing a starter is net negative productivity.

Just some points to ponder.
Of course, if that 30% was the key to your success it could be significant. While those were very experienced, key players on D, we did have some excellent athletes behind them.



User avatar
Rekdiver
Posts: 4915
Joined: Mon Aug 13, 2012 1:14 pm
School: Appalachian State
Has thanked: 723 times
Been thanked: 1952 times

Re: An Interesting Take on 2020

Unread post by Rekdiver » Sun Feb 09, 2020 5:26 pm

Ok I read all the comments and then relooked at the chart. I’m sorry but There are just too many variables to consider to make this any type of predictor. For instance you lose DE but you whole line returns along with Peoples and other. Your best receiver is back too missing most of the season. Raw numbers mean nothing like this. Returning production is relative and no predictor of improvement or failure.



EastHallApp
Posts: 5672
Joined: Wed Nov 06, 2013 4:34 pm
School: Appalachian State
Location: Raleigh
Has thanked: 2548 times
Been thanked: 2063 times

Re: An Interesting Take on 2020

Unread post by EastHallApp » Sun Feb 09, 2020 6:08 pm

Every team has injured guys coming back, every team has some promising young players who may or may not turn out to be great as first-time starters.

On balance, returning production/# of starters has proven to have decent predictive value, I believe. Of course it’s all fodder for discussion, nothing’s guaranteed, etc.

One thing that I think favors us is that a mature program typically is more likely to have the next group of guys ready to step in than one that’s struggled or been up and down. We’ve replaced a lot of key players each of the past couple seasons and done great.



citroknight
Posts: 92
Joined: Thu Oct 31, 2019 2:23 am
School: UCF
Has thanked: 145 times
Been thanked: 71 times

Re: An Interesting Take on 2020

Unread post by citroknight » Sun Feb 09, 2020 8:35 pm

Rekdiver wrote:
Sun Feb 09, 2020 5:26 pm
Ok I read all the comments and then relooked at the chart. I’m sorry but There are just too many variables to consider to make this any type of predictor. For instance you lose DE but you whole line returns along with Peoples and other. Your best receiver is back too missing most of the season. Raw numbers mean nothing like this. Returning production is relative and no predictor of improvement or failure.
Just the raw numbers do miss a lot of the intricate details. But it's not too bad as a quick and dirty method to see what to expect from a team. All relative of course though. A 5 win team returning 90% likely isn't better than a 12+ win team returning 70%. But that 5 win team, at least on paper with this quick and dirty method, is less likely to get worse since they're returning most of the team. Of course that could just mean the 5 win team improves to 6 wins and the 12+ win team "regresses" to 11 wins.



Yosef84
Posts: 2362
Joined: Thu Aug 09, 2012 11:27 am
Has thanked: 307 times
Been thanked: 1071 times

Re: An Interesting Take on 2020

Unread post by Yosef84 » Mon Feb 10, 2020 8:28 am

It's the off season so this is the type of thing fans do...diving into stats and metrics to attempt to predict the next year. None of the models are particularly accurate at this stage because there simply isn't enough data. Attempting to impose context on all the teams in the FBS is a huge undertaking, even if these columnists WANTED to do it....and the fact is they don't. They just want to spark conversation and drive interest (or attract interest to be more specific).

Statistics simply aren't meaningful without context. "Returning productivity" in this is calculated by which players are returning having calculated certain metrics (yards, tackles, etc.) Those productivity metrics aren't always meaningful by themselves. We know that Duck's interceptions dropped his last season at App, but it isn't because he wasn't having an impact on the game. Offenses avoided throwing in his direction so his stats dropped. Defenders make more tackles if the offense is sputtering and leaving them on the field longer, but is that "productivity" a predictor for the future? Well, if the offense continues to sputter, it might be but it has to be taken in context. Stats also vary depending on opponents and the systems they run.

I don't see anybody getting bent out of shape about this kind of stuff and that's a good thing. Games aren't played on paper. Is it almost September yet?

GO Apps!



AppStFan1
Posts: 2035
Joined: Mon Dec 03, 2018 10:37 am
School: Appalachian State
Has thanked: 70 times
Been thanked: 389 times

Re: An Interesting Take on 2020

Unread post by AppStFan1 » Mon Feb 10, 2020 4:53 pm

hapapp wrote:
Sun Feb 09, 2020 3:24 pm
AppStFan1 wrote:
Sun Feb 09, 2020 12:16 pm
AppFan11 wrote:
Sun Feb 09, 2020 10:23 am
Actually that’s what Bill Connolly has done for years, to make stats that are more predictive than the traditional stats. He fully acknowledges that some are better than others. Certainly Phil Steele uses returning production numbers as well. I agree with Hap and East that in our unique situation on defense where we are constantly rotating players so that younger players are getting regular snaps, that Bill’s stats will underestimate our performance. As we know, predictive models generalize predictions by whatever formula is being used and does not take into account specific/unique situations. All that being said, I would expect that 2020 defense to be good overall but probably start slow and get better and better every game.
Agree with you guys. That chart is interesting and it matches what you expect that CCU should get better. Chadwell is a good coach so they should be a bowl team and I would not be shocked if they finish in 2nd or 3rd place in the division. I'm with you that early on the defense may start slow but I expect them to be dominant by week 5. Just like we said last year when new coaches came in. We have a lot of talent so the expectation is a 2020 SBC Championship. My worry is 2021 when we have a new QB and replace a ton of production on offense.
I wonder if it takes into account the number of folks who have left the Coastal program. I'm not clued in enough to know if their losses were backups or starters. But, I believe they had a significant number in the portal.
Good question. I have not looked that closely at it. I just remember they seemed to not have many seniors and I feel like Jamey is a good coach who would have them as a contender in the next year or two.



Post Reply Previous topicNext topic

Return to “Appalachian Football”

  • Who is online
  • In total there are 191 users online :: 1 registered, 0 hidden and 190 guests

    Most users ever online was 1927 on Tue Mar 17, 2020 3:27 am

    Users browsing this forum: appstate24 and 190 guests