Pool C -- 2015

[wally_wabash]

Agree with you guys about once ranked, always ranked.  In football, we have a hard time separating and selecting teams because there is a lack of data and almost non-existent intersecting data.  Ignoring the results of those first two sets of regional rankings is throwing away valuable data that can help distinguish teams from one another. 

AO drew comparisons with the basketball committee- and maybe ORAR isn't better for basketball, but basketball also has 2.5x the data points with which to define teams.  Division III likes to have the same set of selection rules across all of its championships, which is a nice idea on the surface, but can fall flat in practice because basketball is different than football is different than field hockey is different than baseball.  I don't see the harm in first recognizing that these sports are different and then adopting selection criteria that are manicured to that sport and the data available to give the committee the best opportunity to select and seed teams. 

AO also mentioned that the basketball national committee isn't shy about rearranging the regional rankings.  My recollection is that the football committee chairperson has been asked about this previously and they have stated that they are reluctant to undo those regional rankings.  I'd need confirmation from the In The HuddLLe guys on that, but that seems to be what I remember. 

[AO]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region. 

[wabndy]

Division III likes to have the same set of selection rules across all of its championships, which is a nice idea on the surface, but can fall flat in practice because basketball is different than football is different than field hockey is different than baseball.

Basketball has held onto the primary criteria preference of in-region competition that football used to hold.  I presume someone in the competition committee finally figured out that with some teams only playing 7 games against any D3 opponents, making further distinctions for in-region was needlessly tossing out additional data points.  The football prechampionships manual continues to play lip service to the importance of in-region competition, but they have effectively gutted that by not making it a primary or secondary criteria.

[wally_wabash]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

I don't think that's really a problem at all.  Win percentage is a criteria and losing damages win percentages (piping hot take right there).  Remember- these rankings are very different from the subjective balloting that goes on at D3football.com and the AFCA. 

[AO]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

I don't think that's really a problem at all.  Win percentage is a criteria and losing damages win percentages (piping hot take right there).  Remember- these rankings are very different from the subjective balloting that goes on at D3football.com and the AFCA.

I think some committees do use the "results against regionally ranked opponents" to keep a DePauw in the rankings.  Let's compare two teams:
Team A: 8-2 lost to unranked Ohio Wesleyan and an undefeated highly ranked team
Team B: 9-1 lost to unranked Ohio Wesleyan.  Did not play any ranked teams.
In this scenario I think you have to throw win pct. out the door and look at additional criteria.

[wally_wabash]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

I don't think that's really a problem at all.  Win percentage is a criteria and losing damages win percentages (piping hot take right there).  Remember- these rankings are very different from the subjective balloting that goes on at D3football.com and the AFCA.

I think some committees do use the "results against regionally ranked opponents" to keep a DePauw in the rankings.  Let's compare two teams:
Team A: 8-2 lost to unranked Ohio Wesleyan and an undefeated highly ranked team
Team B: 9-1 lost to unranked Ohio Wesleyan.  Did not play any ranked teams.
In this scenario I think you have to throw win pct. out the door and look at additional criteria.

You don't throw win percentage out.  You look at all of the criteria.  It's not like an order of tiebreak kind of thing.  All of the pieces matter. 

[AO]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

I don't think that's really a problem at all.  Win percentage is a criteria and losing damages win percentages (piping hot take right there).  Remember- these rankings are very different from the subjective balloting that goes on at D3football.com and the AFCA.

I think some committees do use the "results against regionally ranked opponents" to keep a DePauw in the rankings.  Let's compare two teams:
Team A: 8-2 lost to unranked Ohio Wesleyan and an undefeated highly ranked team
Team B: 9-1 lost to unranked Ohio Wesleyan.  Did not play any ranked teams.
In this scenario I think you have to throw win pct. out the door and look at additional criteria.

You don't throw win percentage out.  You look at all of the criteria.  It's not like an order of tiebreak kind of thing.  All of the pieces matter.

Same difference?  Throw it out or consider it "tied" and move on to additional criteria.

[USee]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

Your problem doesn't go away with your solution. If you don't have ORAR you have the inverse of the same problem. Depauw drops out after losing in the final week and North Central sneaks in by winning their last 3 games (but losing all the ones that matter). That benefits Wheaton and Platteville and hoses Wabash. ORAR gives you more data in a data starved process.

[AO]

Once ranked always ranked should come back. They don't rank the teams until after week 7. There is much less of a chance of pretenders in week 7 than in week 2 or 3. The way it is now penalizes teams like Wabash who will play Depauw in the last game. If Depauw drops that game they may not be ranked in the final rankings and, while Depauw is not a top 5 team in the region, they are certainly a top 8-12 type team. Contrast that with ONU who lost 2 games near the front of their schedule and by virtue of winning out may slide into the top 10 at the end (a phenomena that appeared to have helped St Thomas last year and penalized UWP). At the margin, once ranked always ranked is a positive criteria that helps more than it hurts in my opinion.

So if ONU only gets to #11 and DePauw dropped to #13 we should give more credit for beating DePauw?  The final rankings are the only one's that matter.  That way there is a zero percent chance of pretenders in the top ten.   The real problem with this scenario is ending the rankings at an arbitrary point at 10 and committees dropping teams like DePauw out of the rankings for losing to the #1 or #2 team in the region.

Your problem doesn't go away with your solution. If you don't have ORAR you have the inverse of the same problem. Depauw drops out after losing in the final week and North Central sneaks in by winning their last 3 games (but losing all the ones that matter). That benefits Wheaton and Platteville and hoses Wabash. ORAR gives you more data in a data starved process.

I agree that someone always gets hosed when wins against #10 counts for a lot more than wins against #11.   But if you're going to draw the line, you can't count teams that dropped from the ten as being better than teams who just missed the ten.   That's more data, but it's inconsistent, incomplete data.

[USee]

You are focused on the smallest part of the problem. If you use ORAR in our example, Depauw drops to #11, 12, or 13 and North Central moves up to #10. That solves two of the three problems in that both these teams could be #10 so giving them both credit treats them as co-#10s. It doesn't solve the issue of ONU going from #13 and finishing at #11, thereby giving some teams credit for beating a team ranked lower, but that problem exists with or without ORAR. Use the example of ONU and JCU. ONU could conceivably sneak in to #10 and JCU falls to #11 simply because JCU plays Mt Union later. That's not fair. It doesn't have playoff implications this year but a similar situation could.

[AO]

Use the example of ONU and JCU. ONU could conceivably sneak in to #10 and JCU falls to #11 simply because JCU plays Mt Union later. That's not fair. It doesn't have playoff implications this year but a similar situation could.

As I argued with my team A, team B scenario, this loss to Mount should not move JCU to #11.   However, the order you play teams should have no impact on the final regional rankings.  To give every team a fair shot, committees need to re-evaluate every team in the rankings based upon the final results.  Nobody "sneaks" above another team.  It is fully earned according to their full schedule, regardless of when the games were played.

[Andy Jamison - Walla Walla Wildcat]

Maybe this has already been discussed but my buddy brought up the idea of reseeding the field for the Quarterfinals as the NCAA travel restrictions are lifted by then (I believe). This way there would be a guarantee that all 4 games have the best chance of being competitive. The playoff committee would be able to seed the teams 1-8 and go from there.

How on earth do you intend to guarantee competitive games in the regional final round by reshuffling?  How can you possibly know that those games, no matter how you match them up, would be competitive?   

The travel restrictions are not lifted, really ever.   By the time you get 3 rounds deep, it's difficult (but not impossible) to ensure teams will be within the 500 mile radius.  It's definitely not a free for all in the quarterfinals though.  If the committee were to shuffle the deck intermittently throughout the tournament, they'd do so in order to save dollars and not find competitive balance (or what you think is competitive balance). 

Here was last year's final 8.  How would you have "re-seeded" them to make for a better tournament? 
Wartburg vs. UWW
Linfield vs. Widener
John Carroll vs. Mount Union
Hobart vs. Wesley

As a rule - No two East teams play each other in the Quarters - They have to play outside the East

To have seeded last year I'll use both after the fact and how I would have done it:

After first two rounds in 2014 - #1 UWW #2 MUC #3 Wesley #4 Linfield #5 Wartburg #6 John Carroll #7 Widener #8 Hobart

UWW vs Hobart
MU vs Widener
Wesley vs JC
Linfield vs Wartburg

After the fact

#1 UWW #2 MUC #3 Linfield #4 Wartburg #5 JC #6 Wesley #7 Widener #8 Hobart

UWW vs Hobart
MUC vs Widener
Linfield vs Wesley
Wartburg vs JC

Neither really accomplishes what I want which is to get the East down to 1 team by the 3rd round - aka Wesley or whatever teams beats them.

So this idea doesn't seem to work (or at least for 2014)

[wally_wabash]

Maybe this has already been discussed but my buddy brought up the idea of reseeding the field for the Quarterfinals as the NCAA travel restrictions are lifted by then (I believe). This way there would be a guarantee that all 4 games have the best chance of being competitive. The playoff committee would be able to seed the teams 1-8 and go from there.

How on earth do you intend to guarantee competitive games in the regional final round by reshuffling?  How can you possibly know that those games, no matter how you match them up, would be competitive?   

The travel restrictions are not lifted, really ever.   By the time you get 3 rounds deep, it's difficult (but not impossible) to ensure teams will be within the 500 mile radius.  It's definitely not a free for all in the quarterfinals though.  If the committee were to shuffle the deck intermittently throughout the tournament, they'd do so in order to save dollars and not find competitive balance (or what you think is competitive balance). 

Here was last year's final 8.  How would you have "re-seeded" them to make for a better tournament? 
Wartburg vs. UWW
Linfield vs. Widener
John Carroll vs. Mount Union
Hobart vs. Wesley

As a rule - No two East teams play each other in the Quarters - They have to play outside the East

To have seeded last year I'll use both after the fact and how I would have done it:

After first two rounds in 2014 - #1 UWW #2 MUC #3 Wesley #4 Linfield #5 Wartburg #6 John Carroll #7 Widener #8 Hobart

UWW vs Hobart
MU vs Widener
Wesley vs JC
Linfield vs Wartburg

After the fact

#1 UWW #2 MUC #3 Linfield #4 Wartburg #5 JC #6 Wesley #7 Widener #8 Hobart

UWW vs Hobart
MUC vs Widener
Linfield vs Wesley
Wartburg vs JC

Neither really accomplishes what I want which is to get the East down to 1 team by the 3rd round - aka Wesley or whatever teams beats them.

So this idea doesn't seem to work (or at least for 2014)

At least we know with zero uncertainty what your motivations are here.  For a guy who is concerned about east coast bias, this is as blatantly east coast biased as it gets. 

[jknezek]

Neither really accomplishes what I want which is to get the East down to 1 team by the 3rd round - aka Wesley or whatever teams beats them.

So this idea doesn't seem to work (or at least for 2014)

And that should slam this conversation closed...

[smedindy]

Neither really accomplishes what I want which is to get the East down to 1 team by the 3rd round - aka Wesley or whatever teams beats them.

Ah, geez. Someone with no earthly idea of the spirit and ethos of D3.

"I want a hamburger...no a cheeseburger. I want a hot dog. I want a milkshake. I want potato chips..."