D3 Top 25 Fan Poll

[ExTartanPlayer]

C-M isn't exactly a bad loss. They are 4-0, 2-0. Will they beat St. Thomas? I don't know, but it's not like St. Johns lost to a hopeless team. Why is this mitigating before the St. Thomas - CM Game? Wouldn't you take both games at face value right now and rank C-M over St. Johns over St. Thomas?  If any one of them loses again before the C-M / St. Thomas game you could re-evaluate then, but right now why mess up what seems a series of credible results on the field? Is there a reason to believe St. Thomas is significantly better than C-M?

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

[jknezek]

I think everyone is. Hence NCC dropping a ton, and UWSP getting added. It's just not being used to your liking.

Perhaps. But again, I just haven't seen anyone provide a reason why NCC should be above UWSP based on the data from this year. Either game data showing it was a fluky win or season data showing why the win shouldn't be considered the deciding factor on which team is better. I've seen people point out other instances, argue that polls can't move that much, or that good teams can't drop that far, or that hypothetically UWSP would not win again. But nothing data based on this season.

Without that data it is confusing that the result on the field is not the arbiter of "better team". From the pollsters point of view, why was the game played if it didn't determine who the better team was?

[02 Warhawk]

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

Now I assume this thinking goes out the door if NCC loses again, right? No way should you keep dropping UWP further and further down just so they're behind NCC, correct?

[ExTartanPlayer]

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

Now I assume this thinking goes out the door if NCC loses again, right? No way should you keep dropping UWP further and further down just so they're behind NCC, correct?

Yes!  I've said that several times already (*edit: heck, it was even in that post - "If another result contradicts this order..."). 

A direct quote from a few posts back:

"If there is other data within that same season to support it - i.e. Kenyon losing to several other teams that Wittenberg has beaten already - then, of course you consider those data points as well, and conclude that a larger body of evidence proves that Wittenberg is better than Kenyon.

In the specific cases we're arguing about, that data does not exist.  There is no result that supersedes UWSP 34, NCC 27.  If UWSP loses down the road to UWP, fine, the rankings will appropriately reflect how everyone weighs those results later."

[02 Warhawk]

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

Now I assume this thinking goes out the door if NCC loses again, right? No way should you keep dropping UWP further and further down just so they're behind NCC, correct?

Yes!  I've said that several times already.  A direct quote from a few posts back:

"If there is other data within that same season to support it - i.e. Kenyon losing to several other teams that Wittenberg has beaten already - then, of course you consider those data points as well, and conclude that a larger body of evidence proves that Wittenberg is better than Kenyon.

In the specific cases we're arguing about, that data does not exist.  There is no result that supersedes UWSP 34, NCC 27.  If UWSP loses down the road to UWP, fine, the rankings will appropriately reflect how everyone weighs those results later."

sorry...trying my best to get some work done and talk D3 football at the same time. Losing track of some posts. 

[ExTartanPlayer]

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

Now I assume this thinking goes out the door if NCC loses again, right? No way should you keep dropping UWP further and further down just so they're behind NCC, correct?

Yes!  I've said that several times already.  A direct quote from a few posts back:

"If there is other data within that same season to support it - i.e. Kenyon losing to several other teams that Wittenberg has beaten already - then, of course you consider those data points as well, and conclude that a larger body of evidence proves that Wittenberg is better than Kenyon.

In the specific cases we're arguing about, that data does not exist.  There is no result that supersedes UWSP 34, NCC 27.  If UWSP loses down the road to UWP, fine, the rankings will appropriately reflect how everyone weighs those results later."

sorry...trying my best to get some work done and talk D3 football at the same time. Losing track of some posts. 

No worries.  They're coming hot and heavy, and I'm probably neglecting my work a little too much for this discussion as well.

+K for the spirited discussion.

[BashDad]

Some argue or imply that that is what you do. Rank teams ahead of teams they beat, the hell with the other data points.

Who has argued that?  Please, quote a specific instance of someone in this thread saying that.

There's a strong implication in some of the comments above through the entire thread where those who take a more nuanced view of a H2H result may be called out. I went extreme with Kenyon / Wittenberg, but the St. Thomas / St. John's game is probably a better example where since it's a rivalry game, and St. John's has a loss as well, where the H2H between the two may NOT be as black and white as some would think. On that given day St. Thomas lost. Who is the better team, though? The loss to C-M by St. John's still lingers. IF C-M beats St. Thomas, then all in the MIAC is aligned.

Nuance is good! I'm all for nuance. More nuance, please.

But it truly is dumb to rank teams according to how you expect other people to rank them or because this is how polls my whole life have reflected similar results or because so-and-so plays so-and-so in week 8 and will surely lose. These are dumb ways to qualify teams in relation to each other. Dumb especially when you're arguing against a result on the field that has little additional data to mitigate the clearest of who-would-win-between-so-and-so answers: the scoreboard.

[02 Warhawk]

Some argue or imply that that is what you do. Rank teams ahead of teams they beat, the hell with the other data points.

Who has argued that?  Please, quote a specific instance of someone in this thread saying that.

There's a strong implication in some of the comments above through the entire thread where those who take a more nuanced view of a H2H result may be called out. I went extreme with Kenyon / Wittenberg, but the St. Thomas / St. John's game is probably a better example where since it's a rivalry game, and St. John's has a loss as well, where the H2H between the two may NOT be as black and white as some would think. On that given day St. Thomas lost. Who is the better team, though? The loss to C-M by St. John's still lingers. IF C-M beats St. Thomas, then all in the MIAC is aligned.

Nuance is good! I'm all for nuance. More nuance, please.

But it truly is dumb to rank teams according to how you expect other people to rank them or because this is how polls my whole life have reflected similar results or because so-and-so plays so-and-so in week 8 and will surely lose. These are dumb ways to qualify teams in relation to each other. Dumb especially when you're arguing against a result on the field that has little additional data to mitigate the clearest of who-would-win-between-so-and-so answers: the scoreboard.

We'll be looking forward to your poll next week.

[smedindy]

C-M isn't exactly a bad loss. They are 4-0, 2-0. Will they beat St. Thomas? I don't know, but it's not like St. Johns lost to a hopeless team. Why is this mitigating before the St. Thomas - CM Game? Wouldn't you take both games at face value right now and rank C-M over St. Johns over St. Thomas?  If any one of them loses again before the C-M / St. Thomas game you could re-evaluate then, but right now why mess up what seems a series of credible results on the field? Is there a reason to believe St. Thomas is significantly better than C-M?

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

The rivalry game factor is part of the equation. There's a lot of antipathy between the teams and that can affect play one way or another. If it was just a *normal* game, I would weigh this H2H a lot more, but you all missed the rivalry game part of my comment.

THAT is the nuance.

[Saxon73]

If this was the last game either team would play, which team would be ranked higher ??  UWSP or NCC

[jknezek]

The rivalry game factor is part of the equation. There's a lot of antipathy between the teams and that can affect play one way or another. If it was just a *normal* game, I would weigh this H2H a lot more, but you all missed the rivalry game part of my comment.

THAT is the nuance.

I wouldn't buy it, but I'll concede the valid point... +K

[BashDad]

Some argue or imply that that is what you do. Rank teams ahead of teams they beat, the hell with the other data points.

Who has argued that?  Please, quote a specific instance of someone in this thread saying that.

There's a strong implication in some of the comments above through the entire thread where those who take a more nuanced view of a H2H result may be called out. I went extreme with Kenyon / Wittenberg, but the St. Thomas / St. John's game is probably a better example where since it's a rivalry game, and St. John's has a loss as well, where the H2H between the two may NOT be as black and white as some would think. On that given day St. Thomas lost. Who is the better team, though? The loss to C-M by St. John's still lingers. IF C-M beats St. Thomas, then all in the MIAC is aligned.

Nuance is good! I'm all for nuance. More nuance, please.

But it truly is dumb to rank teams according to how you expect other people to rank them or because this is how polls my whole life have reflected similar results or because so-and-so plays so-and-so in week 8 and will surely lose. These are dumb ways to qualify teams in relation to each other. Dumb especially when you're arguing against a result on the field that has little additional data to mitigate the clearest of who-would-win-between-so-and-so answers: the scoreboard.

We'll be looking forward to your poll next week.

No, dude. There's no way I know enough about the 50 or so teams in consideration to order them in any kind of respectable way. And you don't either, along with everyone else. It's an impossible, frivolous task at all levels and everywhere. It's all silly. You're just--on top of all that--doing it poorly.

[BashDad]

C-M isn't exactly a bad loss. They are 4-0, 2-0. Will they beat St. Thomas? I don't know, but it's not like St. Johns lost to a hopeless team. Why is this mitigating before the St. Thomas - CM Game? Wouldn't you take both games at face value right now and rank C-M over St. Johns over St. Thomas?  If any one of them loses again before the C-M / St. Thomas game you could re-evaluate then, but right now why mess up what seems a series of credible results on the field? Is there a reason to believe St. Thomas is significantly better than C-M?

Exactly.  Same rationale I've been trying to use for UWSP > NCC > UWP.

If another result actually contradicts this order, adjust the rankings when that happens.  Not in anticipation that it will happen.

The rivalry game factor is part of the equation. There's a lot of antipathy between the teams and that can affect play one way or another. If it was just a *normal* game, I would weigh this H2H a lot more, but you all missed the rivalry game part of my comment.

THAT is the nuance.

The reason why this doesn't matter is that it never won't be a rivalry game. Maybe it changes how you qualify one of the teams against the other non-rivalry-opponents in the poll, but it doesn't make sense to dismiss the information about the two teams who played each other.

[BashDad]

I'm not talking specifically about SJU/UST by the way -- there's a loss involved, so knock yourself out with nuance, in that case. Go nuts.

[Bombers798891]

Ridiculous. Kenyon has mitigating data in terms of other losses. UWSP is undefeated. There isn't any mitigating data. You just want to ignore the result.

I don't think anyone's ignoring the result. There's a reason North Central went from 533 points in the polls to 289 while Stevens-Point went from 0 to 188. People clearly think more of Stevens-Point and less of North Central then they did last week. The result's not being ignored, it's just not being weighed as heavily as you'd like it to be by everyone

I know. The question a few of us are asking is why? Why aren't people using the result on the field?

They are. That's what prompted the shift. They're just not using it to the degree that you want.

People look at identical situations and interpret them differently.