Playoffs odds and ends

[Professor Chaos]

Some news over the last few days about this year's FCS playoffs. The new TV slots are out: https://www.inforum.com/fcs-semifinals-going-to-national-television-for-first-time-televised-by-abc

https://twitter.com/SamHerderFCS/status/1851092193215164580

Some notable things here - the Friday night quarterfinal is promoted from ESPN2 to ESPN. The Friday night semifinal is gone but both semifinals on Saturday will air on ABC (although they're going up against CFP first round games on TNT and an NFL game on NBC). There will be a Saturday quarterfinal on ABC again (last year was the first time in my memory that happened). The championship game is now on a Monday night (January 6th at 7PM ET) but that gets it away from any NFL competition since the NFL regular season ends on Sunday January 5th (no Monday night game(s) in the final week of the NFL season).


Also announced this week that that playoff selection committee is officially scrapping the STATS and Coaches polls from their toolbox and using a combination of computer metrics instead to judge things like SOS:

https://twitter.com/SamHerderFCS/status/1850964365588984100


Finally, the FCS playoff selection committee is unveiling their midseason top 10 (the only ranking they'll release until Selection Sunday) tomorrow 10/30 on the 2PM ET broadcast of College Football Live on ESPN2.

[FUBeAR]

The 1st news is all good.

The 2nd part is INSANE. FUBeAR is no fan of the polls, but this is FAR WORSE. It seems to FUBeAR that FCS peeps would have already been in TotallyTriggered FreakOut mode over it. Maybe this will raise awareness.

As of today KPI has Rhode Island as the #2 Seed and Drake as the #3 Seed. Sound good to y’all?

[ElCid]

The 1st news is all good.

The 2nd part is INSANE. FUBeAR is no fan of the polls, but this is FAR WORSE. It seems to FUBeAR that FCS peeps would have already been in TotallyTriggered FreakOut mode over it. Maybe this will raise awareness.

As of today KPI has Rhode Island as the #2 Seed and Drake as the #3 Seed. Sound good to y’all?

You need to look at it in its entirety. It ain't JUST KPI. So relax.

[FUBeAR]

You need to look at it in its entirety. It ain't JUST KPI. So relax.
OK … Of the “voters,” 25% of them have URI as the #2 Seed and Drake as the #3 Seed.

And, that gives you great confidence in the system?

[Mocs123]

Maybe they should just use the AGS Poll instead? It seems like the most accurate.

[ElCid]

OK … Of the “voters,” 25% of them have URI as the #2 Seed and Drake as the #3 Seed.

And, that gives you great confidence in the system?

You still aren't looking at it in its entirety. You are stuck on them being #2. The KPI "ALONE" will "never" be used to get to the final outcome so you harping on it makes zero sense. You are making a false equivalency believing that it equates to 25% of voters, when a closer comparison would be 25% of each voter themselves. And even then it wouldn't necessarily be true. It's simply one of four factors weighted as appropriate in each voter's mind.

If they are still #2 after ALL 4 factors are used, then come back and we can talk.

However, I'm a little leery about using a rating system (Massey, or any other private/commercial one) where the algorithm isn't 100% transparent and cypherable by others. Wouldn't take much to tweak it on the sly. Not saying he would, but if it can happen, and can't be verified, it could introduce doubt. It isn't going to be 100% objective, so the committee will still wield an idiot check function, but it's the close calls that we need to be concerned with.

And this doesn't even speak to the inherent flaws in Massey. But it does seem to be consistent in its objectivity. But if a team is somewhat isolated in its schedule exposure, it seems a bit unreliable. It also weighs the past season a bit heavily early on regardless of subjective factors (lost players, injuries, coaches, etc). But by seasons end, this prior weighting is corrected out.

[FUBeAR]

the committee will still wield an idiot check function
As Coach Bill S. might say … “ah, there’s the rub”

FUBeAR believes the Committee Members have been, as Papa FUBeAR used to say to, and about, FUBeAR, “…as worthless as mammary glands on a boar hog” …or something like that.

FUBeAR believes, from his assessment of their “work,” they must spend 99% of their time eating NCAA-sponsored hotel cuisine, lying to each other about how wonderful their own Athletics Programs are, networking for their next job, and/or chasing, well, y’know … FUBeAR has no idea where they convene, but he always envisions it as a swanky hotel in the French Quarter in New Orleans. Then, the other 1% of the time, they look at the STATS Poll, the poopbutt “Coaches” Poll, and jiggle those around to build-a-bracket that best optimizes their 2 goals - pissing off the fewest number of people they know personally AND preserving the lucre the NCAA so deeply covets.

And, that’s FUBeAR’s non-cynical voice speaking. You should hear what cynical FUBeAR thinks about the Committee.

Adding something as goofy as KPI appears to be can only gum up their ineptitude even further - no matter how much it does / doesn’t count.

[ElCid]

As Coach Bill S. might say … “ah, there’s the rub”

FUBeAR believes the Committee Members have been, as Papa FUBeAR used to say to, and about, FUBeAR, “…as worthless as mammary glands on a boar hog” …or something like that.

FUBeAR believes, from his assessment of their “work,” they must spend 99% of their time eating NCAA-sponsored hotel cuisine, lying to each other about how wonderful their own Athletics Programs are, networking for their next job, and/or chasing, well, y’know … FUBeAR has no idea where they convene, but he always envisions it as a swanky hotel in the French Quarter in New Orleans. Then, the other 1% of the time, they look at the STATS Poll, the poopbutt “Coaches” Poll, and jiggle those around to build-a-bracket that best optimizes their 2 goals - pissing off the fewest number of people they know personally AND preserving the lucre the NCAA so deeply covets.

And, that’s FUBeAR’s non-cynical voice speaking. You should hear what cynical FUBeAR thinks about the Committee.

Adding something as goofy as KPI appears to be can only gum up their ineptitude even further - no matter how much it does / doesn’t count.

Now that all seems to be a very accurate estimation.

[Professor Chaos]

As Coach Bill S. might say … “ah, there’s the rub”

FUBeAR believes the Committee Members have been, as Papa FUBeAR used to say to, and about, FUBeAR, “…as worthless as mammary glands on a boar hog” …or something like that.

FUBeAR believes, from his assessment of their “work,” they must spend 99% of their time eating NCAA-sponsored hotel cuisine, lying to each other about how wonderful their own Athletics Programs are, networking for their next job, and/or chasing, well, y’know … FUBeAR has no idea where they convene, but he always envisions it as a swanky hotel in the French Quarter in New Orleans. Then, the other 1% of the time, they look at the STATS Poll, the poopbutt “Coaches” Poll, and jiggle those around to build-a-bracket that best optimizes their 2 goals - pissing off the fewest number of people they know personally AND preserving the lucre the NCAA so deeply covets.

And, that’s FUBeAR’s non-cynical voice speaking. You should hear what cynical FUBeAR thinks about the Committee.

Adding something as goofy as KPI appears to be can only gum up their ineptitude even further - no matter how much it does / doesn’t count.
FUBeAR is far too cynical is his takes on the selection committee methinks. As El Cid pointed out FPI is being used as a tool (I'd assume primarily to quantify SOS) not as a guideline for how they should seed - although I do agree that KPI is low grade garbage. They don't just look at the STATS or Coaches polls either because their seedings and selections have more closely mirrored the AGS Poll than either of those (link (https://www.anygivensaturday.com/showthread.php?216083-Which-poll-is-the-most-accurate-playoff-predictor)). They're going to screw up from time to time but the pitchfork mafia against them led by FUBeAR is not very fair in their criticisms IMO.

[Preferred Walk-On]

Also announced this week that that playoff selection committee is officially scrapping the STATS and Coaches polls from their toolbox and using a combination of computer metrics instead to judge things like SOS:

https://twitter.com/SamHerderFCS/status/1850964365588984100

Finally, the FCS playoff selection committee is unveiling their midseason top 10 (the only ranking they'll release until Selection Sunday) tomorrow 10/30 on the 2PM ET broadcast of College Football Live on ESPN2.
Can anybody point me to the ESPN FPI for FCS teams? I'm sure I am just missing something here.

[Professor Chaos]

Can anybody point me to the ESPN FPI for FCS teams? I'm sure I am just missing something here.
I'm assuming the FCS one isn't published online anywhere - the FBS one is here: https://www.espn.com/college-football/fpi

Might be nice, if the focus here is "transparency", to have all of these ratings the committee is using easily accessible.

[KnightoftheRedFlash]

Games on ABC, even in sacrificial lamb timeslots against the CFP and the NFL, is a good sign of growth.

[ElCid]

FUBeAR is far too cynical is his takes on the selection committee methinks. As El Cid pointed out FPI is being used as a tool (I'd assume primarily to quantify SOS) not as a guideline for how they should seed - although I do agree that FPI is low grade garbage. They don't just look at the STATS or Coaches polls either because their seedings and selections have more closely mirrored the AGS Poll than either of those (link (https://www.anygivensaturday.com/showthread.php?216083-Which-poll-is-the-most-accurate-playoff-predictor)). They're going to screw up from time to time but the pitchfork mafia against them led by FUBeAR is not very fair in their criticisms IMO.

You don't think they are in a back smoke filled room at a 5 Star hotel trading lies? Lol. It's fun to picture though.

But I do agree with you that all these tools they use are just that, simply data points to organize their thoughts, not an objective means of choosing or ranking the playoff participants. The danger is that it could morph into that if they aren't careful. People love to abdicate to purely objective measures if allowed to or it supports their position.

[Preferred Walk-On]

I'm assuming the FCS one isn't published online anywhere - the FBS one is here: https://www.espn.com/college-football/fpi

Might be nice, if the focus here is "transparency", to have all of these ratings the committee is using easily accessible.
I can't even see it with my ESPN+ subscription? Must need to bundle that and pick the "no ads" option to view it. :D

You are correct though, transparency is important. Kind of like seeding the "top" 16 teams. Color me skeptical, but I also believe that the seeds will magically be as close as possible to being as "regional" matchups. We'll see.

Cynical FUBeAR must be soiling his drawers now that there are only two metrics available (Massey and KPI).

[ElCid]

I can't even see it with my ESPN+ subscription? Must need to bundle that and pick the "no ads" option to view it. :D

You are correct though, transparency is important. Kind of like seeding the "top" 16 teams. Color me skeptical, but I also believe that the seeds will magically be as close as possible to being as "regional" matchups. We'll see.

Cynical FUBeAR must be soiling his drawers now that there are only two metrics available (Massey and KPI).

Yup. I thought that about regional results when they first announced the change to 16. But I'll wait to see. It could happen by accident and circumstances, so to be fair, it might be best to wait a couple, three years to see if it's a real outcome or just a fear.

[Professor Chaos]

I can't even see it with my ESPN+ subscription? Must need to bundle that and pick the "no ads" option to view it. :D

You are correct though, transparency is important. Kind of like seeding the "top" 16 teams. Color me skeptical, but I also believe that the seeds will magically be as close as possible to being as "regional" matchups. We'll see.

Cynical FUBeAR must be soiling his drawers now that there are only two metrics available (Massey and KPI).
If it was out there it would be here I think: https://www.espn.com/college-football/fpi/_/group/81

But that's a dead link.

[FUBeAR]

FUBeAR is far too cynical is his takes on the selection committee methinks. As El Cid pointed out FPI is being used as a tool (I'd assume primarily to quantify SOS) not as a guideline for how they should seed - although I do agree that KPI is low grade garbage. They don't just look at the STATS or Coaches polls either because their seedings and selections have more closely mirrored the AGS Poll than either of those (link (https://www.anygivensaturday.com/showthread.php?216083-Which-poll-is-the-most-accurate-playoff-predictor)). They're going to screw up from time to time but the pitchfork mafia against them led by FUBeAR is not very fair in their criticisms IMO.
Fair?

As Papa FUBeAR taught FUBeAR and FUBeAR, in turn, taught FUBeAR Jr & Little Miss FUBeAR - the only “fair” we know comes to town once a year and has rides and cotton candy…and hot tubs for sale, these days.


BTW - KPI rates EVERY Non-FCS Team, on an FCS Team’s schedule - from Alabama to Lincoln (CA) as the 34th best FCS Team to determine SoS - now, THAT is some damn, high-quality algorithmic modeling. They had to use some serious AI to come up with that piece of work!

[Bisonator]

I'd say anything that has Rhode Island and Drake in the top 5 is complete garbage. WTF is this KPI? There is something wrong with their algorithm if that's what they are getting.

I'd rather the committee keep the polls in play at least use everything at your disposal to come up with your selections and seedings. What's the harm in that?

[Preferred Walk-On]

Yup. I thought that about regional results when they first announced the change to 16. But I'll wait to see. It could happen by accident and circumstances, so to be fair, it might be best to wait a couple, three years to see if it's a real outcome or just a fear.
It has always been my opinion that you play who they put in front of you. Period. From my biased perspective, this has been true when NDSU has been seeded lower (or not seeded) than they should be (IMO), when they placed all 5 MVFC teams on the same side of the bracket, and even when they sent SDSU to NDSU in early rounds of the playoffs due to regionality. You have to win 4 (or 5 games), and you have to beat those teams eventually, so it really doesn't matter.

From a transparency point of view each committee member should have their rankings published to see if the seeds match the consensus of the committee. It is quite simple. Pick the teams that qualify, then rank them from 1-24 (I guess you don't have to show 17-24). There are 10 members, so do it like a poll (24 pt for 1st, 1 pt for 24th) and show the results. Now the final eight can be slotted in across from 9-16, and nobody will really care at that point (and little regionality bias, presumably only with the 1st round matchups). By doing this, it becomes difficult to manipulate the rankings of the 10 members to fit alternative seeding patterns (unless they want to sit and do the math). Ten is a large enough number to do this, so why not? Again, the teams would already have been selected, and it would just be a matter of filling in the bracket based on numbers (not region).

[Bisonator]

Maybe they should just use the AGS Poll instead? It seems like the most accurate.
This. They should just let AGS do the selections and seeding. xnodx

[dbackjon]

Fair?

As Papa FUBeAR taught FUBeAR and FUBeAR, in turn, taught FUBeAR Jr & Little Miss FUBeAR - the only “fair” we know comes to town once a year and has rides and cotton candy…and hot tubs for sale, these days.


BTW - KPI rates EVERY Non-FCS Team, on an FBS Team’s schedule - from Alabama to Lincoln (CA) as the 34th best FCS Team to determine SoS - now, THAT is some damn, high-quality algorithmic modeling. They had to use some serious AI to come up with that piece of work!

Dback agrees with FuBear that the KPI is a complete joke

[CHIP72]

Games on ABC, even in sacrificial lamb timeslots against the CFP and the NFL, is a good sign of growth.

Perhaps, but TV time slots that have less competition are even better.

I was just looking up the TV ratings for various DI-AA/FCS postseason games in 2023, in response to discussion on Reddit, specifically the FCS subreddit, about the DI-AA/FCS semifinals moving to ABC (I said good luck competing with the two NFL games and DI-A/FBS college football playoff games at the same time). Here are the TV ratings/audiences for various games; the following information was taken from the Sports Media Watch site:

DI-AA/FCS quarterfinals:

*12/08 (Fri): Furman/Montana, 8:00 PM ET (ESPN2) - 679K viewers, 0.39 rating

*12/09 (Sat): Villanova/South Dakota State, 12:00 PM ET (ESPN) - 1.85M viewers, 1.1 rating

*12/09 (Sat): North Dakota State/South Dakota, 2:30 PM ET (ABC) - 1.09M viewers, 0.7 rating

Notice how VU/SDSU had a much larger audience than NDSU/USD, despite the fact the latter was on ABC while the former was on ESPN. The NDSU/USD game mostly overlapped with the Army/Navy game (on CBS) however while the VU/SDSU had no linear TV college football competition.

DI-AA/FCS semifinals & Celebration Bowl:

*12/15 (Fri): Albany/South Dakota State, 7:00 PM ET (ESPN2) - 476K viewers, 0.29 rating

*12/16 (Sat): Howard/Florida A&M, 12:00 PM ET (ABC) - 1.51M viewers, 0.9 rating

*12/16 (Sat): North Dakota State/Montana, 4:30 PM ET (ESPN2) - 790K viewers, 0.48 rating

The Celebration Bowl was helped immensely by being on ABC while the two playoff games were televised on ESPN2. However, the Celebration Bowl also had the lowest ratings of any of the three college games on ABC that day, and both semifinal games had smaller TV audiences than any of the six DI-A/FBS bowl games that weekend (all Saturday games). In turn, the DI-A/FBS bowl game ratings were dwarfed by the ratings for the three NFL games that day, despite the fact the NFL games were on the NFL Network and the college bowl games were on ABC or ESPN.

DI-AA/FCS national championship game:

*01/07 (Sun): Montana/South Dakota State, 2:00 PM ET (ABC) - 1.04M viewers, 0.6 rating

The championship game had fewer viewers than two of the quarterfinal playoff games, including one televised on ESPN, and the Celebration Bowl. The low viewership was entirely due to direct competition with NFL games taking place at the same time.

The biggest factor in getting more eyeballs watching DI-AA/FCS games is limited competition in its time slot. The Villanova/South Dakota State quarterfinal, televised on ESPN, had the largest TV viewing audience of any DI-AA/FCS postseason game in 2023. That game was largely unopposed in its time slot (there were no DI-A/FBS games or NFL games at the same time, and the only other college games were lower classification playoff games on streaming only ESPN+). Getting on over-the-air television also helped, but not as much as not having competition from DI-A/FBS games or especially NFL games. Both the Celebration Bowl and National Championship game were on ABC, but the former had less NFL (and DI-A/FBS) competition and largely because of that attracted more TV viewers.

The two games that will benefit from the revised DI-AA/FCS postseason schedule and network assignments IMO will be 1) the Celebration Bowl (now a week earlier and held prior to the Army-Navy Game while still being televised on ABC) and especially 2) the national championship game, now played on a Monday night and not receiving any NFL competition.


Sent from my iPhone using Tapatalk

[ElCid]

It has always been my opinion that you play who they put in front of you. Period. From my biased perspective, this has been true when NDSU has been seeded lower (or not seeded) than they should be (IMO), when they placed all 5 MVFC teams on the same side of the bracket, and even when they sent SDSU to NDSU in early rounds of the playoffs due to regionality. You have to win 4 (or 5 games), and you have to beat those teams eventually, so it really doesn't matter.

From a transparency point of view each committee member should have their rankings published to see if the seeds match the consensus of the committee. It is quite simple. Pick the teams that qualify, then rank them from 1-24 (I guess you don't have to show 17-24). There are 10 members, so do it like a poll (24 pt for 1st, 1 pt for 24th) and show the results. Now the final eight can be slotted in across from 9-16, and nobody will really care at that point (and little regionality bias, presumably only with the 1st round matchups). By doing this, it becomes difficult to manipulate the rankings of the 10 members to fit alternative seeding patterns (unless they want to sit and do the math). Ten is a large enough number to do this, so why not? Again, the teams would already have been selected, and it would just be a matter of filling in the bracket based on numbers (not region).

I actually agree. I basically don't care about regionalization in the end. As an example, does it actually matter if you play a team down the road from you instead of a team on the opposite coast. You have to play who is on the schedule. You may have to play any team eventually.

Sure, how far a team gets in the playoffs can be just as much a matter of schedule as how good you are. But it only matters for people who are wrapped up in a dick measuring contest of claiming who made it to the quarterfinal or semi final, or final. Some teams have an easier path than others. Some teams have to play a season opponent again in the playoffs. So what. You still have to beat them now or later. If you don't beat them, but beat a patsy or relatively weaker team instead, it doesn't make you any better just because you made it to the next round easily. It always amazes me how some fans of certain teams hang their hat on the hard or easy schedule their team has to play in the playoffs. Or how many conference teams are on one side of the bracket or the other. Again, so what. You still got to win the same number of games.

[caribbeanhen]

This. They should just let AGS do the selections and seeding. xnodx

AGS should submit a bracket to these hucksters and publicize it

by the way, thanks for asking WTF is KPI….

regionalization blows