FB: Region 4 fan poll

[smedindy]

I'm not the one who posted a long diatribe after they were tired.

[jknezek]

Getting into the mind of what a pollster believes to me is mocking them.

Yeesh. Standard debate practices is to get into your opponents mind so you can try and understand what they are talking about and try and refute it. You put forward that ExTP is much too rigid in his arguments. Aren't you in his mind? Doesn't seem mocking to me, or denigrating, just refutation, but it's the same thing. There is good empirical evidence for at least two of the three statements I made.

Poll Bias, or anchoring, is a well researched problem. That is point 1.

We all should agree that point 2 is a problem. Every talking head will tell you to lose early, not late, because you get punished extra for losing late in the season. Plenty of empirical data to support that assumption over the D1 poll years.

Point 3 is related to point 1. I probably can't find much evidence to support it specifically, but it is extremely logical based on what we know about poll bias.

Becoming aware of issues like this can sometimes help people recognize if/when they are doing it. Then again, other people just don't care or don't believe.

[wally_wabash]

All of this.  Maybe even less logical than talking oneself out of the most meaningful piece of data we can have (a head-to-head result) is how much results matter less and less over time.  And not 8 weeks.  We're talking about two weeks here.  You're absolutely right.  If Franklin and IWU have the exact same game results that they have gathered so far but you just shuffle the order in which those results happened, IWU would be ranked on every ballot and they would be ranked ahead of Franklin on every ballot.  It's not even a question.

You can't be so absolute. You can't have rigidity like that.

Is there any debate here at all that if IWU had beaten Franklin 4 days ago that Franklin would be ranked ahead of IWU? 

There's just a missing piece that you see that I don't.  I see that IWU beat Franklin head to head.  I see that Franklin got boatraced by the best team in the nation.  I see that IWU lost to Simpson.  There is nothing here that tells me that IWU beating Franklin was weird.  We might get there.  After a few more weeks, IWU might lose again to Augustana or something.  Franklin might beat MSJ and RHIT by 100.  But right now, there is no evidence to suggest that Franklin is better than IWU other than we thought Franklin was better than IWU to begin with and that's what we're going to stick with in spite of the result of the actual game that those two teams played. 

[smedindy]

I think the nuance is they are advocating a position of rigidity, which I disagree with. I can see your point, a bit, and maybe I have been a bit sensitive - but I CARE about my ballot and I do take nuance into account and it really cheeses ME off when people claim to think they know how, why I think, if I think by rote AND that there MUST be a sense of rigid order when there is NOT.

I happen to think that most of us are taking everything into account so there's not much poll bias in our smaller data set since most of the voters seem to research and do this for love, not because they are forced to and they pawn off the duties to the intern or the associate SID. I know I don't try to be biased. Maybe I'm speaking for myself in my rosy-jack utopian world.

And yakkers and mediots feed into the issue of #2, which I know I try to rise above.

[smedindy]

All of this.  Maybe even less logical than talking oneself out of the most meaningful piece of data we can have (a head-to-head result) is how much results matter less and less over time.  And not 8 weeks.  We're talking about two weeks here.  You're absolutely right.  If Franklin and IWU have the exact same game results that they have gathered so far but you just shuffle the order in which those results happened, IWU would be ranked on every ballot and they would be ranked ahead of Franklin on every ballot.  It's not even a question.

You can't be so absolute. You can't have rigidity like that.

Is there any debate here at all that if IWU had beaten Franklin 4 days ago that Franklin would be ranked ahead of IWU? 

There's just a missing piece that you see that I don't.  I see that IWU beat Franklin head to head.  I see that Franklin got boatraced by the best team in the nation.  I see that IWU lost to Simpson.  There is nothing here that tells me that IWU beating Franklin was weird.  We might get there.  After a few more weeks, IWU might lose again to Augustana or something.  Franklin might beat MSJ and RHIT by 100.  But right now, there is no evidence to suggest that Franklin is better than IWU other than we thought Franklin was better than IWU to begin with and that's what we're going to stick with in spite of the result of the actual game that those two teams played.

The date point of the FIRST game of the year, ON the road, and then the subsequent results. I had IWU ahead of Franklin early on, but it was kind of a thin margin. A close loss AT IWU didn't scream to me that Franklin is worse NOW than IWU. I wasn't tremendously impressed with IWU's result against Albion.

Location matters, too. Franklin loses to IWU at home, yeah that's different. But it was the first game on the road and it was a one-possession game and it wasn't a last minute score to get to one possession.

I think IWU, on a neutral field, wouldn't beat Franklin 51 times out of 100. They are 10 and 11 right now for me. And frankly anyone under Wheaton and Heidelberg for me is up in the air as to where they land.

[Li'l Giant]

Is there any debate here at all that if IWU had beaten Franklin 4 days ago that Franklin would be ranked ahead of IWU?   

Is that so surprising? I don't think it is. A result from the first game of the season is going to be considered differently than a result from the last game of the season, especially, if your poll philosophy has to do with ranking the teams as the sit *now* at the time of the ranking.

How often have we heard "it's better to lose early than lose late"?

[02 Warhawk]

It's all just fascinating.  Are we giving Franklin credit for taking a lead against Whitewater? 

Similar thing happened in the D3football.com poll. Whitewater lost a 1st place vote after the Franklin game, before getting it back the following week. I guess someone temporarily lost confidence in Whitewater since the game wasn't a blowout in the first quarter.

Some people forget Franklin is a very good football program, and it's an accomplishment in itself just to win against them on the road.

[ExTartanPlayer]

Now that we're all cheesed off (I hope it's a good smoked Gouda)...

I think digging into this stuff is informative because we learn how differently everyone's thought process works when ranking teams.  Seeing the range of opinions - not just the rankings, but the "why?" that comes out of discussions like this - is informative because playoff selection committees are also made up of people, with opinions and preconceived notions just like the people here, and those things will undeniably influence their decisions come playoff time, selection criteria be damned.

It's a plausible scenario that Simpson and IWU could end up on the board at the same time in Pool C playoff selection.  Let's pretend that Simpson goes 9-1 with a loss to Wartburg, and IWU goes 8-2 with losses to North Central and Simpson.  Suppose that we're down to the final Pool C slot and those two teams are on the board.  It's hard to imagine someone picking the 8-2 team with a h2h loss over a 9-1 team, but if someone here can formulate some mind-boggling explanation for why IWU should still be ranked ahead of Simpson, even if it's as dumb as "IWU beat Wheaton and we think Wheaton is better than Simpson, so we're going to assume IWU is better than Simpson even though Simpson actually played IWU and beat them" - undoubtedly someone on the committee could come to the same conclusion.

Or, suppose that NCC and UWSP/UWP are both on the board.  It's 8-2 NCC (with a loss to undefeated Wheaton and 8-2 UWSP) against 8-2 UWSP (with losses to undefeated UWW and 8-2 UWP).  Will they ignore the h2h result?  Will they acknowledge it, but say NCC beat UWP who beat UWSP, so the common-opponent outweighs the actual game between the two?  What if it's 8-2 NCC against 9-1 UWP (with a loss to UWW and a win over UWSP)?  Does NCC's h2h win over UWP outweigh the fact that they have two losses, one of which came against a common opponent that UWP beat?  That one is a lot more complicated, and there I can see a rational argument on both sides.

I'm not arguing for "rigidity" or against "nuance" - none of my posts said that h2h trumped all.  Really, go back and read them.  Here, I'll help:

"You don't have to be slaves to head-to-head results when there is data that trumps the head-to-head result, (example: if UWP beats UWSP in a few weeks, there's some sense in ranking NCC>UWSP because NCC>UWP and UWP>UWSP, contradicting last week's UWSP>NCC result) but right now there is none."

[FCGrizzliesGrad]

I'm usually the cause of oddball votes generating discussion... but for once it's not my fault. I had IWU at 7.
Now as to the other odd votes Wes mentioned... I may have been responsible for some of those.

[smedindy]

But ETP, there are results and other mitigating factors weighing in:

1. First game.
2. ROAD GAME! One possession road game.
3. IWU's less than impressive run against Albion and loss to Simpson.
4. Franklin's rebound after Whitewater.

Don't get cheesed off when others figure THAT into the equation and you loudly exclaim "H2H! H2H!" when those factors play into it!

[wally_wabash]

IWU played Albion and Simpson on the road. Why don't they get a pass?

[ExTartanPlayer]

1. First game.
2. ROAD GAME! One possession road game.
3. IWU's less than impressive run against Albion and loss to Simpson.
4. Franklin's rebound after Whitewater.

1. Some funky things happen in the first game.  The games still count in the standings.
2. This is your most valid point; I'm fine with calling a one-score road game a tossup.
3. Albion was in the playoffs last year and their "other" loss this year is to the UWSP team that just beat NCC.  Simpson was 7-3 last year and is currently 3-0.  Those are both pretty decent teams. I don't see how splitting close games with them is worse than...
4. Franklin's "rebound" has come against two teams that went 4-16 last year and are a combined 1-6 (with the lone win against Earlham).  They've blown the doors off both teams like they should, but still...those are two horrendous teams.  I don't see how two blowout wins against two teams they were really supposed to blow out outweighs a prior h2h loss.  IWU would have beaten those teams just as badly. 

(To be clear: I think Franklin has done an admirable job building their program and scheduling tough OOC, I know they're not scared of good competition)

If Wabash beat Wittenberg 42-35 at Hollett and Wittenberg blew out Hiram and Kenyon the next two weeks, would you use Witt's "rebound" as evidence that they deserved to move up in the rankings?

[smedindy]

It's the combo. Loss to Simpson is fine, if you will. But personally, I'm not sold on Albion.

[smedindy]

1. First game.
2. ROAD GAME! One possession road game.
3. IWU's less than impressive run against Albion and loss to Simpson.
4. Franklin's rebound after Whitewater.

1. Some funky things happen in the first game.  The games still count in the standings.
2. This is your most valid point; I'm fine with calling a one-score road game a tossup.
3. Albion was in the playoffs last year and their "other" loss this year is to the UWSP team that just beat NCC.  Simpson was 7-3 last year and is currently 3-0.  Those are both pretty decent teams. I don't see how splitting close games with them is worse than...
4. Franklin's "rebound" has come against two teams that went 4-16 last year and are a combined 1-6 (with the lone win against Earlham).  They've blown the doors off both teams like they should, but still...those are two horrendous teams.  I don't see how two blowout wins against two teams they were really supposed to blow out outweighs a prior h2h loss.  IWU would have beaten those teams just as badly. 

(To be clear: I think Franklin has done an admirable job building their program and scheduling tough OOC, I know they're not scared of good competition)

If Wabash beat Wittenberg 42-35 at Hollett and Wittenberg blew out Hiram and Kenyon the next two weeks, would you use Witt's "rebound" as evidence that they deserved to move up in the rankings?

Games still count but it's a feeling out period, I think. Your team in week one is not the team in week 5 or 7. Right now I think Franklin is a tick, a skosh, a smidge better than IWU.

Of course your hypothetical ignores the universe of other games. I guess that Albion game hangs with me - even though they were a playoff team last year. I don't trust the MIAA farther than I can throw them...

[smedindy]

On another note, what in the heck happened to the CCIW mid-pack teams? If we're seriously debating Illinois Wesleyan as ranked or unranked, then whither Carthage, Millikin, and Augie? Is the NCAC closing up with the CCIW?

*ducking*