PROVED: NU victory over EIU PROPELLED it into COACHES top 25!!!

[FeliSilvestris]

After NU beat (then and now) top-25 Stanford NU remained outside the top 25 (with 72 AP points, and 35 from the coaches). It was after NU beat EIU that the AP voters and the COACHES gave NU enough votes to make the top 25 (170 AP points, and 95 coaches' points). This suggests that the victory over 'lowly' EIU earned NU almost 100 additional AP pts, and 60 additional coaches' pts.

However, in another thread it has been claimed (without proof) that NU's EIU victory wasn't what propelled it into the top 25, but rather losses by other teams did.

WHO IS RIGHT? The numbers make that COMPLETELY CLEAR.

Let us look at the coaches poll, since presumably coaches are even more FB knowledgeable than AP voters (or should be).

After the 1st week results (including NU's victory over Stanford) the top 4 teams in the "others receiving votes" category were:
Miss St 97, Ariz St 66, Okla St 52, Northwestern 35.
If all these teams beat comparable opponents, NU would need 4 teams from the top 25 to drop out, in order to become #25.

The week-2 scores included:
#17 OkU 31-24 #23 Tenn
#15 LSU 21-19 Miss St
Tol 16-12 #18 Ark
BYU 35-24 #22 Bois St
OkSU 32-8 C Ark
AzSU 35-21 Cal Poly
NU 41-0 EIU

From these results, you'd expect that at least Okla St and Ariz St would stay ahead of NU (they won their games), and quite probably Miss St (which lost by only 2 to #15), as well as #23 Tenn (which lost by 7 to #17).

Yet, the new rankings included:
#24 NU
#25 Okla St
3 teams dropped out: #18 Ark, #22 Bois St, and #23 Tenn
Others receiving votes: ..., Ariz St 36, Miss St 35

Thus, not only did NU jump the ranked teams that dropped out (including Tenn which lost to #17), as well as Miss St (which lost to #15 by only 2).

NU also jumped Ariz St (which won its game but remained outside the top 25), as well as Okla St (which won its game and made the top 25 but behind NU).

Thus, NU's victory over 'lowly' EIU allowed it to jump multiple teams, including 2 which also won their games and 2 that narrowly lost to (higher) ranked teams.

Face it: NU victory over EIU was WELL APPRECIATED by the voting COACHES!!!

 

[nycat33]

Perhaps most of the coaches had the Northwestern vs Stanford game on their DVR and hadn't had a chance to watch it yet. I know I haven't watched the season premiere of Big Bang Theory yet and could be underrating this show. Do you have stats on the viewership habits of division 1 coaches, because that would be useful.

 

[villox]

Perhaps most of the coaches had the Northwestern vs Stanford game on their DVR and hadn't had a chance to watch it yet. I know I haven't watched the season premiere of Big Bang Theory yet and could be underrating this show. Do you have stats on the viewership habits of division 1 coaches, because that would be useful.

That's not how it works. Because the poll comes out by noon the next day, all of the coaches must rush home after their games on Saturday and are forced to watch high speed replays of each of the Top 30 games, clockwork orange style. That's how they make such informed decisions.

 

[gocatsgo2003]

 

[gocatsgo2003]

That's not how it works. Because the poll comes out by noon the next day, all of the coaches must rush home after their games on Saturday and are forced to watch high speed replays of each of the Top 30 games, clockwork orange style. That's how they make such informed decisions.

Or the coaches have absolutely nothing to do with it and the poll is typically filled out by some low-level staffer and/or someone from the Sports Information office. That's how it usually works.

(Pretty sure that's what you were implying, but didn't see the sarcasm font.)

 

[villox]

Or the coaches have absolutely nothing to do with it and the poll is typically filled out by some low-level staffer and/or someone from the Sports Information office. That's how it usually works.

(Pretty sure that's what you were implying, but didn't see the sarcasm font.)

Uh, no way. The accuracy of polls in the second week of the season are too important to allow low level staff to fill them out. And all 40+ or so coaches that make up the poll all must use the same criteria when judging teams which is why it's such a reliable ranking.

 

[Purple Pile Driver]

After NU beat (then and now) top-25 Stanford NU remained outside the top 25 (with 72 AP points, and 35 from the coaches). It was after NU beat EIU that the AP voters and the COACHES gave NU enough votes to make the top 25 material (170 AP points, and 95 coaches' points). This suggests that the victory over 'lowly' EIU earned NU almost 100 additional AP pts, and 60 additional coaches' pts.

However, in another thread it has been claimed (without proof) that NU's EIU victory wasn't what propelled it into the top 25, but rather losses by other teams did.

WHO IS RIGHT? The numbers made that COMPLETELY CLEAR.

Let us look at the coaches poll, since presumably coaches are even more FB knowledgeable than AP voters (or should be).

After the 1st week results (including NU's victory over Stanford) there were 3 teams ahead of NU in the "others receiving votes" category:
Mississippi State 97, Arizona State 66, Oklahoma State 52, Northwestern 35
If all these teams beat comparable opponents, NU would need 4 teams from the top 25 to drop out, become #25.
As it turned out, only 3 teams did lose and drop out: #18 Arkansas, #22 Boise Stat, #23 Tennessee
Also, OkSU beat C. Ark 32-8, ASU beat Cal Poly 35-21, and NU beat 'lowly' EIU 41-0.
But one of the 3 teams ahead of NU, MiSU lost but only by 2 pts to #15 LSU!!

From these results, you'd expect that at least OkSU and ASU would stay ahead of NU (they won their games), and probably MiSU also, since they only lost by 2 to #15...

Yet, here are the new rankings:
#24 NU
#25 OkSU
......
Others receiving votes: ..., ASU 36, MiSU 35

Not only did NU jumped MiSU (which lost to #15 by only 2), it also jumped ASU (which won its game but remained outside the top 25), and also jumped OkSU (which won its game and made the top 25 but behind NU).

NU victory over 'lowly' EIU allowed it to jump 3 teams (2 of which also won their games and one that lost by only 2 to #15 LSU).

Face it: NU victory over EIU was WELL APPRECIATED by the COACHES!!!

Are the capital letters shouting at us necessary to make another absurd assumption?

 

[rwhitney014]

Man, this must be what reading Pravda was like.

 

[thewildcat2011]

Man, this must be what reading Pravda was like.

Now that's funny.

 

[Hungry Jack]

Man, this must be what reading Pravda was like.

From my understanding, at least the Pravda was useful in the outhouse. You cannot wipe your arse with one of Feli's posts.

 

[Mr. Stupor]

feli: please get help.

 

[FeliSilvestris]

After NU beat (then and now) top-25 Stanford NU remained outside the top 25 (with 72 AP points, and 35 from the coaches). It was after NU beat EIU that the AP voters and the COACHES gave NU enough votes to make the top 25 material (170 AP points, and 95 coaches' points). This suggests that the victory over 'lowly' EIU earned NU almost 100 additional AP pts, and 60 additional coaches' pts....
Let us look at the coaches poll, since presumably coaches are even more FB knowledgeable than AP voters (or should be).

For completeness sake, I did a similar analysis of the AP rankings. The most amazing thing happened concerning Wiscky.

The AP ranking after week 1 included:
#25 Miss St 104,
Wisconsin 79,
BYU 74,
Northwestern 72

The week-2 game scores included:
#14 LSU 21-19 #25 Miss St
#19 Okla 31-24 #23 Tenn
Tol 16-12 #18 Ark
BYU 35-24 #20 Bois St
Wicky 58-0 Mia (OH)
OkSU 32-8 C. Ark

The new ranking included:
19 BYU 349
:::
23 Northwestern 170
24 Wis 158
25 Okla St 48

Four teams dropped out: #18 Ark, #20 Bois St , #23 Tenn, #25 Miss St

Not only did NU jumped all teams that dropped out, including Miss St (which lost by only 2 to #14 LSU), and Tenn (which lost by only 7 to #19 Okla).

Amazingly, NU also jumped Wis, which had beat Mia(OH) 58-0!!

Those AP voters were mighty impressed by NU's defeat of EIU....apparently even more impressed than they were by Wis's dismantling of Mia (OH), an FBS program (recall that Wis was ahead of NU in the previous poll!).

 

[Williesfan]

The Stanford W in week 2 coupled with blanking EIU was the catalyst.

 

[NUCat320]

Or the coaches have absolutely nothing to do with it and the poll is typically filled out by some low-level staffer and/or someone from the Sports Information office. That's how it usually works.

(Pretty sure that's what you were implying, but didn't see the sarcasm font.)

Hey, did you ever get to fill one out, GCG? (no sarcasm)

 

[FeliSilvestris]

The Stanford W in week 2 coupled with blanking EIU was the catalyst.

Yes, the Stanford win got us from not mentioned to "Others receiving votes".
But the amazing thing about the EIU win is that it enabled us, not only to jump teams that barely lost to ranked team, but even teams that themselves won their games (as Wisky in the AP poll, which destroyed FBS Mia(OH) 58-0!). Truly amazing!
Obviously, someone forgot to tell the AP voters AND the voting COACHES how horrible EIU is!

 

[EvanstonCat]

Still Going...

You are like the energizer rabbit. You just keep going and going and going.

Horse = dead. Actually smashed many times over into ground and unrecognizeable meat.

 

[TheC]

That's not how it works. Because the poll comes out by noon the next day, all of the coaches must rush home after their games on Saturday and are forced to watch high speed replays of each of the Top 30 games, clockwork orange style. That's how they make such informed decisions.

Actually, head coaches take this important role very seriously. Why do you think they all wear those headsets during the game? They are listening to the radio calls of games from around the country so they have all the info they need to make those picks.

 

[gocatsgo2003]

Hey, did you ever get to fill one out, GCG? (no sarcasm)

Nope.

 

[FeliSilvestris]

Actually, head coaches take this important role very seriously. Why do you think they all wear those headsets during the game? They are listening to the radio calls of games from around the country so they have all the info they need to make those picks.

Look, everyone is free to joke, make fun/light of the job coaches (and/or by extension AP voters) do while answering their ballot. Often the same people who discredit polls one day are the ones using those polls another day, when it suits their purpose.

I would hope that coaches take their membership in the voting panel (in representation of their school, their conference and the American Football Coaches Association) seriously, and view it as an honor. If most do, that may be enough for the poll to cast reasonably accurate results. A few non-serious ballots shouldn't affect the results drastically.

The poll organizers can also help their own cause by taking reasonable measures. For example, they may have a system to compare a coaches ballot against actual on-field results, whether weekly or by season's end. They may reward or recognize voters who prove most accurate by the chosen metric (especially those whose pre-season and early season ballots do better at season's end).

The organizers may also help voters by using today's technology to provide them with relevant information in a neat and easy to use way (using graphics, charts, video-clips when appropriate, etc., etc), "just a click away". For the preseason poll, the "information package" may emphasize previous season results, and key personnel changes (graduations, transfers, injuries, new-comers, etc). For the in-season polls, the package may provide scores sorted in various ways, but also performance information about the teams involved. Say, they see the NU/Duke score, clicking on Duke may open a window showing relevant information about Duke (conference, last-season results, current season results, previous opponents, notable facts such as key injuries, etc.). It may be a lot of work to set this up initially, but once the system is up and running, updating it on a weekly basis may be a few hours of work for a capable staffer, and some on the info may be computer-updated. Not sure that they do it this way, but if they don't it's only a matter of time before they do it.

Anyway, if poll results were "crazy" the list of "top 25" teams would look like randomly drawn, and "upsets" would be the norm rather than the exception. If the polls come REASONABLY close to on field results one has to assume that the voters are doing a reasonable (if imperfect) job.

 

[shakes3858]

Look, everyone is free to joke, make fun/light of the job coaches (and/or by extension AP voters) do while answering their ballot. Often the same people who discredit polls one day are the ones using those polls another day, when it suits their purpose.

I would hope that coaches take their membership in the voting panel (in representation of their school, their conference and the American Football Coaches Association) seriously, and view it as an honor. If most do, that may be enough for the poll to cast reasonably accurate results. A few non-serious ballots shouldn't affect the results drastically.

The poll organizers can also help their own cause by taking reasonable measures. For example, they may have a system to compare a coaches ballot against actual on-field results, whether weekly or by season's end. They may reward or recognize voters who prove most accurate by the chosen metric (especially those whose pre-season and early season ballots do better at season's end).

The organizers may also help voters by using today's technology to provide them with relevant information in a neat and easy to use way (using graphics, charts, video-clips when appropriate, etc., etc), "just a click away". For the preseason poll, the "information package" may emphasize previous season results, and key personnel changes (graduations, transfers, injuries, new-comers, etc). For the in-season polls, the package may provide scores sorted in various ways, but also performance information about the teams involved. Say, they see the NU/Duke score, clicking on Duke may open a window showing relevant information about Duke (conference, last-season results, current season results, previous opponents, notable facts such as key injuries, etc.). It may be a lot of work to set this up initially, but once the system is up and running, updating it on a weekly basis may be a few hours of work for a capable staffer, and some on the info may be computer-updated. Not sure that they do it this way, but if they don't it's only a matter of time before they do it.

Anyway, if poll results were "crazy" the list of "top 25" teams would look like randomly drawn, and "upsets" would be the norm rather than the exception. If the polls come REASONABLY close to on field results one has to assume that the voters are doing a reasonable (if imperfect) job.

OMG, this is your greatest work ever Feli!!! "I would hope that coaches take their membership in the voting panel (in representation of their school, their conference and the American Football Coaches Association) seriously, and view it as an honor." Where do you get this crap?

 

[TheC]

Look, everyone is free to joke, make fun/light of the job coaches (and/or by extension AP voters) do while answering their ballot. Often the same people who discredit polls one day are the ones using those polls another day, when it suits their purpose.

I would hope that coaches take their membership in the voting panel (in representation of their school, their conference and the American Football Coaches Association) seriously, and view it as an honor. If most do, that may be enough for the poll to cast reasonably accurate results. A few non-serious ballots shouldn't affect the results drastically.

The poll organizers can also help their own cause by taking reasonable measures. For example, they may have a system to compare a coaches ballot against actual on-field results, whether weekly or by season's end. They may reward or recognize voters who prove most accurate by the chosen metric (especially those whose pre-season and early season ballots do better at season's end).

The organizers may also help voters by using today's technology to provide them with relevant information in a neat and easy to use way (using graphics, charts, video-clips when appropriate, etc., etc), "just a click away". For the preseason poll, the "information package" may emphasize previous season results, and key personnel changes (graduations, transfers, injuries, new-comers, etc). For the in-season polls, the package may provide scores sorted in various ways, but also performance information about the teams involved. Say, they see the NU/Duke score, clicking on Duke may open a window showing relevant information about Duke (conference, last-season results, current season results, previous opponents, notable facts such as key injuries, etc.). It may be a lot of work to set this up initially, but once the system is up and running, updating it on a weekly basis may be a few hours of work for a capable staffer, and some on the info may be computer-updated. Not sure that they do it this way, but if they don't it's only a matter of time before they do it.

Anyway, if poll results were "crazy" the list of "top 25" teams would look like randomly drawn, and "upsets" would be the norm rather than the exception. If the polls come REASONABLY close to on field results one has to assume that the voters are doing a reasonable (if imperfect) job.

It is interesting to consider how much preparation is put into casting that coaches' vote every week. Not knowing for sure, I am in the camp that it is some Sports Info staffer or football office employee who is just pretty knowledgeable about college football.

 

[villox]

And if You showed me an actual head coach was spending more than 10 minutes filling out a poll instead of watching film and preparing for the next week's opponent I'd show you a coach who wouldn't have that job much longer.

 

[CoralSpringsCat]

Nope.

I know someone that has...

 

[FeliSilvestris]

It is interesting to consider how much preparation is put into casting that coaches' vote every week. Not knowing for sure, I am in the camp that it is some Sports Info staffer or football office employee who is just pretty knowledgeable about college football.

Perhaps it is a staffer in some cases, or the staffer may provide the coach a "suggested" ranking for him to make the final decision. It obviously depends on the coach. Anyway, what matters is whether the poll results are reasonably consistent with on field results. Obviously, if most weeks most victories are divided about equally between the higher and lower ranked team, the ranking would be useless. To the extend that isn't happening (higher ranked teams tend to win most games) whoever does the voting is doing OK.

Anyhow, what most voters probably do is to use their own previous week's ballot as basis for this week's, asking themselves if any of this weeks results warrants a change to their previous vote. Thinking (time) would be necessary mostly when a higher ranked team loses, or barely wins against much inferior competition. But as long as "upsets" are relatively few, a week's ballot may be nearly identical to the previous week's one. In some weeks it may be perfectly reasonable for a voter to submit exactly the same ranking as he did the previous week.

 

[iskaboo]

Felis, remember this?

http://bleacherreport.com/articles/...n-confirms-absolute-hypocrisy-of-coaches-poll

 

[FeliSilvestris]

Felis, remember this?

http://bleacherreport.com/articles/...n-confirms-absolute-hypocrisy-of-coaches-poll

What is the article's importance with respect to this thread?
It seems to be about a specific incident involving one coach that voted only once.
As I have said several times already, the key issue is whether the poll results are reasonably consistent with on field results. By all indications they reasonably are (the higher ranked team tends to win most times). Hence, whatever they are doing and whoever is actually doing it seems to be working reasonably well. Do you disagree?