Hello There, Guest! (LoginRegister)

Post Reply 
Lunardi early 2022 bracketology
Author Message
Atlanta Offline
Heisman
*

Posts: 9,360
Joined: Nov 2009
Reputation: 526
I Root For: Memphis Tigers
Location: Metro Atlanta
Post: #21
RE: Lunardi early 2022 bracketology
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  
(04-20-2021 03:31 PM)Atlanta Wrote:  Ah yes, start the pre-season with the designated two-team conference slot. Let's get the bias rolling so conference wins have no upside.

Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

Exactly. It's the bias that allows the B1G & others to schedule weak OOC to build wins that would mean nothing to us then build a resume against similar programs in-conference where they get 8-10 or more Q1-Q2 opportunities which allows them to play .500 in-conference & be virtually guaranteed an at- large bid. It's bias, pure & simple.
04-24-2021 01:42 PM
Find all posts by this user Quote this message in a reply
jedclampett Offline
All American
*

Posts: 3,312
Joined: Jul 2019
Reputation: 145
I Root For: Temple
Location:
Post: #22
RE: Lunardi early 2022 bracketology
(04-24-2021 01:42 PM)Atlanta Wrote:  
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  
(04-20-2021 03:31 PM)Atlanta Wrote:  Ah yes, start the pre-season with the designated two-team conference slot. Let's get the bias rolling so conference wins have no upside.

Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

Exactly. It's the bias that allows the B1G & others to schedule weak OOC to build wins that would mean nothing to us then build a resume against similar programs in-conference where they get 8-10 or more Q1-Q2 opportunities which allows them to play .500 in-conference & be virtually guaranteed an at- large bid. It's bias, pure & simple.

True, and they don't even have to play .500 in conference - - 9 of the 14 Big 10 teams made the NCAA in 2021.

.
04-24-2021 04:51 PM
Find all posts by this user Quote this message in a reply
macgar32 Offline
Legend
*

Posts: 29,927
Joined: Dec 2007
Reputation: 649
I Root For: Memphis
Location: Bartlett
Post: #23
RE: Lunardi early 2022 bracketology
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  
(04-20-2021 03:31 PM)Atlanta Wrote:  Ah yes, start the pre-season with the designated two-team conference slot. Let's get the bias rolling so conference wins have no upside.

Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

No the point is the computer doesn't care about pre-season human polls.

So in order to get Q1 wins this conference cant crap the bed out of conference.

Don't crap the bed out of conference and teams would have many more opportunities to get Q1 wins...And the all important ability to get Q1 wins at home.
04-24-2021 10:48 PM
Find all posts by this user Quote this message in a reply
jedclampett Offline
All American
*

Posts: 3,312
Joined: Jul 2019
Reputation: 145
I Root For: Temple
Location:
Post: #24
RE: Lunardi early 2022 bracketology
(04-24-2021 10:48 PM)macgar32 Wrote:  
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  
(04-20-2021 03:31 PM)Atlanta Wrote:  Ah yes, start the pre-season with the designated two-team conference slot. Let's get the bias rolling so conference wins have no upside.

Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

No the point is the computer doesn't care about pre-season human polls.

So in order to get Q1 wins this conference cant crap the bed out of conference.

Don't crap the bed out of conference and teams would have many more opportunities to get Q1 wins...And the all important ability to get Q1 wins at home.

You haven't interpreted the data in the table yet. Instead, you've brushed it aside as if it's impertinent.

Based on the way you've chosen not to delve into the table, I assume that you haven't "grokked" it yet.

If you did "grok" it, you would know that the data show a tremendous and growing lack of correspondence between the pre-season and post-season polls over the past 8 years, with the lack of correspondence showing a consistent bias in favor of the P5 teams.

If you have - - please do share your thoughts on what the data show.

.

Now, on to the points you're trying to make.

Are you right in pointing out that winning more OOC games would help boost the AAC's reputation? Undeniably correct.

However, is that all there is to it? I don't think so.

Why not? Because, while the lower half of the AAC has struggled against P5 teams, the lower echelon teams in the P5 conferences have also struggled. The struggles of lower-tier conference teams are by no means not unique to the AAC or the non-P5 conferences.

Here are the Big Ten standings from the 2020 season (all P5 vs. P5 games):

East:

Ohio State 7-1 (.875)
Indiana 6-2 (.750)
Maryland 2-3 (.400) ("crapped the bed" vs. P5 teams)
Penn State 4-5 (.444) ("crapped the bed" vs. P5 teams)
Michigan 2-4 (.333) ("crapped the bed" vs. P5 teams)
Rutgers 3-6 (.333) ("crapped the bed" vs. P5 teams)
Michigan St. 2-5 (.286) ("crapped the bed" vs. P5 teams)

West:

Iowa 6-2 (.750)
Northwestern 6-2 (.750)
Wisconsin 3-3 (.500)
Minnesota 3-4 (.429) ("crapped the bed" vs. P5 teams)
Nebraska 3-5 (.375) ("crapped the bed" vs. P5 teams)
Purdue 2-4 (.333) ("crapped the bed" vs. P5 teams)
Illinois 2-6 (.250) ("crapped the bed" vs. P5 teams)

SUMMARY: 9 (64.3%) of 14 Big Ten teams played poorly vs. P5 teams.

.
04-25-2021 12:38 AM
Find all posts by this user Quote this message in a reply
Hurricane Drummer Offline
All American
*

Posts: 4,621
Joined: Jul 2015
Reputation: 223
I Root For: Tulsa
Location:
Post: #25
RE: Lunardi early 2022 bracketology
There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.
04-25-2021 07:15 AM
Find all posts by this user Quote this message in a reply
Atlanta Offline
Heisman
*

Posts: 9,360
Joined: Nov 2009
Reputation: 526
I Root For: Memphis Tigers
Location: Metro Atlanta
Post: #26
RE: Lunardi early 2022 bracketology
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......
04-25-2021 07:43 AM
Find all posts by this user Quote this message in a reply
Hurricane Drummer Offline
All American
*

Posts: 4,621
Joined: Jul 2015
Reputation: 223
I Root For: Tulsa
Location:
Post: #27
RE: Lunardi early 2022 bracketology
(04-25-2021 07:43 AM)Atlanta Wrote:  
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......

Insane
04-25-2021 08:09 AM
Find all posts by this user Quote this message in a reply
macgar32 Offline
Legend
*

Posts: 29,927
Joined: Dec 2007
Reputation: 649
I Root For: Memphis
Location: Bartlett
Post: #28
RE: Lunardi early 2022 bracketology
(04-25-2021 12:38 AM)jedclampett Wrote:  
(04-24-2021 10:48 PM)macgar32 Wrote:  
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  
(04-20-2021 03:31 PM)Atlanta Wrote:  Ah yes, start the pre-season with the designated two-team conference slot. Let's get the bias rolling so conference wins have no upside.

Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

No the point is the computer doesn't care about pre-season human polls.

So in order to get Q1 wins this conference cant crap the bed out of conference.

Don't crap the bed out of conference and teams would have many more opportunities to get Q1 wins...And the all important ability to get Q1 wins at home.

You haven't interpreted the data in the table yet. Instead, you've brushed it aside as if it's impertinent.

Based on the way you've chosen not to delve into the table, I assume that you haven't "grokked" it yet.

If you did "grok" it, you would know that the data show a tremendous and growing lack of correspondence between the pre-season and post-season polls over the past 8 years, with the lack of correspondence showing a consistent bias in favor of the P5 teams.

If you have - - please do share your thoughts on what the data show.

.

Now, on to the points you're trying to make.

Are you right in pointing out that winning more OOC games would help boost the AAC's reputation? Undeniably correct.

However, is that all there is to it? I don't think so.

Why not? Because, while the lower half of the AAC has struggled against P5 teams, the lower echelon teams in the P5 conferences have also struggled. The struggles of lower-tier conference teams are by no means not unique to the AAC or the non-P5 conferences.

Here are the Big Ten standings from the 2020 season (all P5 vs. P5 games):

East:

Ohio State 7-1 (.875)
Indiana 6-2 (.750)
Maryland 2-3 (.400) ("crapped the bed" vs. P5 teams)
Penn State 4-5 (.444) ("crapped the bed" vs. P5 teams)
Michigan 2-4 (.333) ("crapped the bed" vs. P5 teams)
Rutgers 3-6 (.333) ("crapped the bed" vs. P5 teams)
Michigan St. 2-5 (.286) ("crapped the bed" vs. P5 teams)

West:

Iowa 6-2 (.750)
Northwestern 6-2 (.750)
Wisconsin 3-3 (.500)
Minnesota 3-4 (.429) ("crapped the bed" vs. P5 teams)
Nebraska 3-5 (.375) ("crapped the bed" vs. P5 teams)
Purdue 2-4 (.333) ("crapped the bed" vs. P5 teams)
Illinois 2-6 (.250) ("crapped the bed" vs. P5 teams)

SUMMARY: 9 (64.3%) of 14 Big Ten teams played poorly vs. P5 teams.

.

The post I replied to...

Get context...That is my point. We are rated poorly by the computers as well.

He said in league games didn't matter...They don't matter because we didn't perform well out of league. Perform well out of conference and pre season polls mean nothing.

You replied I missed the point...Which is totally inaccurate. My point is exactly true and your data is irrelevant to this conversation.

Pre season polls do not stop teams from winning games OOC...AAC teams are not only losing to P5 teams. There are formulas that have nothing to do with the Human polls you posted. The reason in conference games don't matter is because they are not considered Q1 games (Entirely formula based)...That is due to the conferences OOC performance.
(This post was last modified: 04-25-2021 08:21 AM by macgar32.)
04-25-2021 08:20 AM
Find all posts by this user Quote this message in a reply
jedclampett Offline
All American
*

Posts: 3,312
Joined: Jul 2019
Reputation: 145
I Root For: Temple
Location:
Post: #29
RE: Lunardi early 2022 bracketology
(04-25-2021 08:20 AM)macgar32 Wrote:  
(04-25-2021 12:38 AM)jedclampett Wrote:  
(04-24-2021 10:48 PM)macgar32 Wrote:  
(04-24-2021 01:14 PM)jedclampett Wrote:  
(04-23-2021 09:14 PM)macgar32 Wrote:  Lol...

If this league would quit crapping the bed OOC (including Memphis) then in league games would matter. We cant blame ESPN for that

You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

No the point is the computer doesn't care about pre-season human polls.

So in order to get Q1 wins this conference cant crap the bed out of conference.

Don't crap the bed out of conference and teams would have many more opportunities to get Q1 wins...And the all important ability to get Q1 wins at home.

You haven't interpreted the data in the table yet. Instead, you've brushed it aside as if it's impertinent.

Based on the way you've chosen not to delve into the table, I assume that you haven't "grokked" it yet.

If you did "grok" it, you would know that the data show a tremendous and growing lack of correspondence between the pre-season and post-season polls over the past 8 years, with the lack of correspondence showing a consistent bias in favor of the P5 teams.

If you have - - please do share your thoughts on what the data show.

.

Now, on to the points you're trying to make.

Are you right in pointing out that winning more OOC games would help boost the AAC's reputation? Undeniably correct.

However, is that all there is to it? I don't think so.

Why not? Because, while the lower half of the AAC has struggled against P5 teams, the lower echelon teams in the P5 conferences have also struggled. The struggles of lower-tier conference teams are by no means not unique to the AAC or the non-P5 conferences.

Here are the Big Ten standings from the 2020 season (all P5 vs. P5 games):

East:

Ohio State 7-1 (.875)
Indiana 6-2 (.750)
Maryland 2-3 (.400) ("crapped the bed" vs. P5 teams)
Penn State 4-5 (.444) ("crapped the bed" vs. P5 teams)
Michigan 2-4 (.333) ("crapped the bed" vs. P5 teams)
Rutgers 3-6 (.333) ("crapped the bed" vs. P5 teams)
Michigan St. 2-5 (.286) ("crapped the bed" vs. P5 teams)

West:

Iowa 6-2 (.750)
Northwestern 6-2 (.750)
Wisconsin 3-3 (.500)
Minnesota 3-4 (.429) ("crapped the bed" vs. P5 teams)
Nebraska 3-5 (.375) ("crapped the bed" vs. P5 teams)
Purdue 2-4 (.333) ("crapped the bed" vs. P5 teams)
Illinois 2-6 (.250) ("crapped the bed" vs. P5 teams)

SUMMARY: 9 (64.3%) of 14 Big Ten teams played poorly vs. P5 teams.

.
We are rated poorly by the computers as well.

Let's just be clear about this point - - the science/engineering of artificial intelligence hasn't advanced to the point at which computers can rate teams.

Computers just run algorithms in accordance with mathematical formulae that are created by their human programmers and spit out the results.

Many of the formulae that have been used (e.g., the RPI, etc.) have turned out to be inaccurate/unreliable (low to modest validity and reliability), and none of them have been found to be particularly satisfactory.

That's why so many of the NCAA selection committee's seedings often bore relatively little relation to the the teams' NET rankings. The committee knows better than to put too much faith in the flawed computer rankings, which are nothing more than "mechanical guesstimates."

Bear this in mind, as well: Some of the "computerized" rankings systems, such as the NET, have a massive "fudge factor." That's why the NCAA refuses to release the details of the NET ranking procedures.

There's a good reason why we have this expression: "Garbage in, garbage out."

.
(This post was last modified: 04-25-2021 09:18 AM by jedclampett.)
04-25-2021 09:11 AM
Find all posts by this user Quote this message in a reply
macgar32 Offline
Legend
*

Posts: 29,927
Joined: Dec 2007
Reputation: 649
I Root For: Memphis
Location: Bartlett
Post: #30
RE: Lunardi early 2022 bracketology
(04-25-2021 09:11 AM)jedclampett Wrote:  
(04-25-2021 08:20 AM)macgar32 Wrote:  
(04-25-2021 12:38 AM)jedclampett Wrote:  
(04-24-2021 10:48 PM)macgar32 Wrote:  
(04-24-2021 01:14 PM)jedclampett Wrote:  You've missed the point, which is that the pre-season rankings and pre-tournament seedings have systematically overestimated the quality of the ranked and seeded P5 teams, while systematically underestimating the quality of the ranked/unranked and seeded/unseeded non-P5 teams.

This has been a pervasive, persistent, and growing problem since the P5 came into existence.

For example:

........Number of non-P5 FB teams listed in the AP Top 25

.............Preseason AP Top 25.......Final AP Top 25..........Error Margin


2013..........1 non-P5 team............1 non-P5 teams........0 non-P5 teams

2014..........0 non-P5 teams...........3 non-P5 teams.......3 non-P5 teams

2015..........1 non-P5 teams...........3 non-P5 teams.......2 non-P5 teams

2016..........0 non-P5 teams..........3 non-P5 teams........3 non-P5 teams

2017..........1 non-P5 team............4 non-P5 teams........3 non-P5 teams

2018..........2 non-P5 teams...........6 non-P5 teams.......4 non-P5 teams

2019..........1 non-P5 teams...........7 non-P5 teams.......6 non-P5 teams

2020..........2 non-P5 teams...........8 non-P5 teams.......6 non-P5 teams


What the data in this table show is that, as a group, the sports journalists that vote in the pre-season football AP poll have been getting worse and worse at predicting how many non-P5 teams there would be in the final AP top 25 poll.

They have only increased the number of non-P5 teams in the pre-season polls very slightly since 2013, and their pre-season rankings suggest that they failed to take any notice of the fact that the number of non-P5 teams in the Final AP Top 25 increased seven-fold (i.e., by 600%) between 2013 and 2019.

.

Moreover, the claim that the AAC has had a "bad" OOC record is inaccurate. If it had, it wouldn't be the #6 FBS conference and the #7 D1 MBB conference.

No the point is the computer doesn't care about pre-season human polls.

So in order to get Q1 wins this conference cant crap the bed out of conference.

Don't crap the bed out of conference and teams would have many more opportunities to get Q1 wins...And the all important ability to get Q1 wins at home.

You haven't interpreted the data in the table yet. Instead, you've brushed it aside as if it's impertinent.

Based on the way you've chosen not to delve into the table, I assume that you haven't "grokked" it yet.

If you did "grok" it, you would know that the data show a tremendous and growing lack of correspondence between the pre-season and post-season polls over the past 8 years, with the lack of correspondence showing a consistent bias in favor of the P5 teams.

If you have - - please do share your thoughts on what the data show.

.

Now, on to the points you're trying to make.

Are you right in pointing out that winning more OOC games would help boost the AAC's reputation? Undeniably correct.

However, is that all there is to it? I don't think so.

Why not? Because, while the lower half of the AAC has struggled against P5 teams, the lower echelon teams in the P5 conferences have also struggled. The struggles of lower-tier conference teams are by no means not unique to the AAC or the non-P5 conferences.

Here are the Big Ten standings from the 2020 season (all P5 vs. P5 games):

East:

Ohio State 7-1 (.875)
Indiana 6-2 (.750)
Maryland 2-3 (.400) ("crapped the bed" vs. P5 teams)
Penn State 4-5 (.444) ("crapped the bed" vs. P5 teams)
Michigan 2-4 (.333) ("crapped the bed" vs. P5 teams)
Rutgers 3-6 (.333) ("crapped the bed" vs. P5 teams)
Michigan St. 2-5 (.286) ("crapped the bed" vs. P5 teams)

West:

Iowa 6-2 (.750)
Northwestern 6-2 (.750)
Wisconsin 3-3 (.500)
Minnesota 3-4 (.429) ("crapped the bed" vs. P5 teams)
Nebraska 3-5 (.375) ("crapped the bed" vs. P5 teams)
Purdue 2-4 (.333) ("crapped the bed" vs. P5 teams)
Illinois 2-6 (.250) ("crapped the bed" vs. P5 teams)

SUMMARY: 9 (64.3%) of 14 Big Ten teams played poorly vs. P5 teams.

.
We are rated poorly by the computers as well.

Let's just be clear about this point - - the science/engineering of artificial intelligence hasn't advanced to the point at which computers can rate teams.

Computers just run algorithms in accordance with mathematical formulae that are created by their human programmers and spit out the results.

Many of the formulae that have been used (e.g., the RPI, etc.) have turned out to be inaccurate/unreliable (low to modest validity and reliability), and none of them have been found to be particularly satisfactory.

That's why so many of the NCAA selection committee's seedings often bore relatively little relation to the the teams' NET rankings. The committee knows better than to put too much faith in the flawed computer rankings, which are nothing more than "mechanical guesstimates."

Bear this in mind, as well: Some of the "computerized" rankings systems, such as the NET, have a massive "fudge factor." That's why the NCAA refuses to release the details of the NET ranking procedures.

There's a good reason why we have this expression: "Garbage in, garbage out."

.

Regardless of how good\bad you feel the computers are...

If Memphis Beats VCU, Auburn and Western Kentucky the AAC may have had 4 teams into the tourney. The AAC controls its destiny...Not human polls.
(This post was last modified: 04-25-2021 10:49 PM by macgar32.)
04-25-2021 10:47 PM
Find all posts by this user Quote this message in a reply
Stickboy46 Online
Heisman
*

Posts: 7,271
Joined: Dec 2016
Reputation: 255
I Root For: Wichita State
Location:
Post: #31
RE: Lunardi early 2022 bracketology
(04-25-2021 07:43 AM)Atlanta Wrote:  
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......

Did you mean 1400? Because there are only 4500 scholarships available in D1 (353 x 13 in a normal year)
04-26-2021 01:29 PM
Find all posts by this user Quote this message in a reply
Atlanta Offline
Heisman
*

Posts: 9,360
Joined: Nov 2009
Reputation: 526
I Root For: Memphis Tigers
Location: Metro Atlanta
Post: #32
RE: Lunardi early 2022 bracketology
(04-26-2021 01:29 PM)Stickboy46 Wrote:  
(04-25-2021 07:43 AM)Atlanta Wrote:  
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......

Did you mean 1400? Because there are only 4500 scholarships available in D1 (353 x 13 in a normal year)

MBB + FB according to an ESPN radio discussion.
04-26-2021 01:34 PM
Find all posts by this user Quote this message in a reply
Stickboy46 Online
Heisman
*

Posts: 7,271
Joined: Dec 2016
Reputation: 255
I Root For: Wichita State
Location:
Post: #33
RE: Lunardi early 2022 bracketology
(04-26-2021 01:34 PM)Atlanta Wrote:  
(04-26-2021 01:29 PM)Stickboy46 Wrote:  
(04-25-2021 07:43 AM)Atlanta Wrote:  
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......

Did you mean 1400? Because there are only 4500 scholarships available in D1 (353 x 13 in a normal year)

MBB + FB according to an ESPN radio discussion.

Ahhh .... the +FB part ..
04-26-2021 01:38 PM
Find all posts by this user Quote this message in a reply
OldFan Offline
Bench Warmer
*

Posts: 195
Joined: Jun 2019
Reputation: 22
I Root For: Memphis
Location:
Post: #34
RE: Lunardi early 2022 bracketology
(04-26-2021 01:38 PM)Stickboy46 Wrote:  
(04-26-2021 01:34 PM)Atlanta Wrote:  
(04-26-2021 01:29 PM)Stickboy46 Wrote:  
(04-25-2021 07:43 AM)Atlanta Wrote:  
(04-25-2021 07:15 AM)Hurricane Drummer Wrote:  There's 1000+ players moving around in the portal and someone thought they could project next seasons brackets? Hmm.

Approaching 4000......

Did you mean 1400? Because there are only 4500 scholarships available in D1 (353 x 13 in a normal year)

MBB + FB according to an ESPN radio discussion.

Ahhh .... the +FB part ..

I was wondering too. I'm glad you asked and he clarified it.
04-26-2021 01:49 PM
Find all posts by this user Quote this message in a reply
Atlanta Offline
Heisman
*

Posts: 9,360
Joined: Nov 2009
Reputation: 526
I Root For: Memphis Tigers
Location: Metro Atlanta
Post: #35
RE: Lunardi early 2022 bracketology
Don't get distracted guys, there will be lots of transfer portal athletes left out in the cold with only a lost scholarship & no free education to show for it. And less than 2% of all college athletes will ever make any money at the pro level. Should make most want you stay & at least get a free education - very short sighted & poor counseling.
04-26-2021 11:22 PM
Find all posts by this user Quote this message in a reply
jedclampett Offline
All American
*

Posts: 3,312
Joined: Jul 2019
Reputation: 145
I Root For: Temple
Location:
Post: #36
RE: Lunardi early 2022 bracketology
(04-26-2021 11:22 PM)Atlanta Wrote:  Don't get distracted guys, there will be lots of transfer portal athletes left out in the cold with only a lost scholarship & no free education to show for it. And less than 2% of all college athletes will ever make any money at the pro level. Should make most want you stay & at least get a free education - very short sighted & poor counseling.

That's the side of the transfer portal story that hardly anyone has thought about very deeply.

There is a big tradeoff for the student athletes, because - as a group - what they're losing is just as significant as what they're gaining.

They have gained the freedom to transfer without having to sit out a year, but one of the effects of the change has been to cause a lot more roster instability.

Not only will some students end up regretting their decisions to enter the transfer portal; many others will end up being gently ushered out after having a talk with their coaches about how they could probably get a lot more playing time somewhere else.

It's too soon to know what kind of an impact the new transfer rules will have on graduation rates, but transferring and relocating is stressful, and the many demands placed on student-athletes can have an adverse affect on academic performance.

The long and short of it is that the benefits of the new & much more flexible transfer portal may end up being offset by the unanticipated disadvantages.

Many student athletes may find themselves looking back on the way things were when they could more or less count on finishing their bachelor's degrees at the universities that recruited them within 4-5 years, and wishing that things hadn't changed.

.
04-27-2021 07:56 AM
Find all posts by this user Quote this message in a reply
Shockit Offline
2nd String
*

Posts: 381
Joined: Jun 2017
Reputation: 20
I Root For: Wichita State
Location:
Post: #37
RE: Lunardi early 2022 bracketology
https://www.ncaa.com/news/basketball-men...all-season

Andy Katz’s Too early Rankings Not Lunardi but just as much Science which is zero.

#11 Houston
#21 Wichita
#22 Memphis
05-01-2021 09:05 PM
Find all posts by this user Quote this message in a reply
Joprior23 Online
1st String
*

Posts: 1,884
Joined: Mar 2016
Reputation: 27
I Root For: UH
Location:
Post: #38
RE: Lunardi early 2022 bracketology
We should have a really strong top 4 next season. After that is anyone’s guess. The only thing that looks guaranteed right now, outside of the top 4, is that ECU will be terrible.
05-01-2021 09:54 PM
Find all posts by this user Quote this message in a reply
jedclampett Offline
All American
*

Posts: 3,312
Joined: Jul 2019
Reputation: 145
I Root For: Temple
Location:
Post: #39
RE: Lunardi early 2022 bracketology
(05-01-2021 09:05 PM)Shockit Wrote:  https://www.ncaa.com/news/basketball-men...all-season

Andy Katz’s Too early Rankings Not Lunardi but just as much Science which is zero.

#11 Houston
#21 Wichita
#22 Memphis

Combined with Lundardi's estimates,

Houston: #4 seed (#12-16) (#14+#11=25/2=#12.5)

Memphis: #5 seed (#16-20) (#18+22=40/2=#20)

Wichita: #12 seed (#44-48) (#46+21=67/2=#33.5)

Lunardi & Katz, on average are predicting:

Houston ~#12.5 (lowest #3/highest #4 seed)

Memphis ~#20 (low #5 seed)

Wichita St. ~#33.5 (#9 seed)

.

If they're right, then the AAC would probably end up with 3 NCAA bids, which is the conference average since 2014.

.

Question: What does Katz know/think about Wichita State that Lunardi doesn't?

.
05-01-2021 10:40 PM
Find all posts by this user Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)


Copyright © 2002-2021 Collegiate Sports Nation Bulletin Board System (CSNbbs), All Rights Reserved.
CSNbbs is an independent fan site and is in no way affiliated to the NCAA or any of the schools and conferences it represents.
This site monetizes links. FTC Disclosure.
We allow third-party companies to serve ads and/or collect certain anonymous information when you visit our web site. These companies may use non-personally identifiable information (e.g., click stream information, browser type, time and date, subject of advertisements clicked or scrolled over) during your visits to this and other Web sites in order to provide advertisements about goods and services likely to be of greater interest to you. These companies typically use a cookie or third party web beacon to collect this information. To learn more about this behavioral advertising practice or to opt-out of this type of advertising, you can visit http://www.networkadvertising.org.
Powered By MyBB, © 2002-2021 MyBB Group.