Pool C -- 2014

[smedindy]

And the next thing is but Whitewater!  But Mount Union!  They're always looking for games!  There's no upside to playing those teams and (probably) losing in September.  Or maybe you Buff State your way into a win in September.  You still have to beat them again in November/December.  If I'm catching that lightning in a bottle, I want it to be in the tournament, not in a September game.

That leaves you with little margin for error. 8-2 teams in "C" do happen, of course, but you're usually last on the bubble and sweating it out.

[wally_wabash]

And the next thing is but Whitewater!  But Mount Union!  They're always looking for games!  There's no upside to playing those teams and (probably) losing in September.  Or maybe you Buff State your way into a win in September.  You still have to beat them again in November/December.  If I'm catching that lightning in a bottle, I want it to be in the tournament, not in a September game.

That leaves you with little margin for error. 8-2 teams in "C" do happen, of course, but you're usually last on the bubble and sweating it out.

8-2 at-larges happen for teams in leagues that have been granted favored nations status.  Franklin, had they beaten IWU and lost to say the hypothetical conference champion MSJ and also to UWW, isn't getting in.  Less hypothetical is the case of Wittenberg- there's a really, really strong possibility that Wittenberg could lose to a really good 10-0 Wabash team and to FCS Butler and not get in.  Perhaps most applicable is the case of Thomas More.  Thomas More isn't getting in and they "stepped up" and played one of the power teams.  9-1 Thomas More had a shot.  8-2 Thomas More most likely doesn't. 

[AO]

also, I hope to never hear from the committee chair again "but we discounted their SOS advantage because they got that from being in a strong conference"!!!!!!!!!!!!

Less of a problem in football with few non-conference games but we should be adding data, not throwing it out.

[wally_wabash]

also, I hope to never hear from the committee chair again "but we discounted their SOS advantage because they got that from being in a strong conference"!!!!!!!!!!!!

Less of a problem in football with OOWP meaning less but we should be adding data, not throwing it out.

Amen to that, which is why the decision to get rid of the once ranked, always ranked clause irks me.  We're ignoring useful and really meaningful data in a scenario where we already don't have a ton of data to work with. 

[AO]

also, I hope to never hear from the committee chair again "but we discounted their SOS advantage because they got that from being in a strong conference"!!!!!!!!!!!!

Less of a problem in football with few non-conference games but we should be adding data, not throwing it out.

Amen to that, which is why the decision to get rid of the once ranked, always ranked clause irks me.  We're ignoring useful and really meaningful data in a scenario where we already don't have a ton of data to work with.

is it really too much of a burden for the committee to rank all the teams?  I agree with the decision that a team that was top ten but is no longer should be valued less than a end of season top ten team, but that win should still help those teams as it's still a good win, just not a top-ten win.

[SaintsFAN]

Thomas More isn't getting in and they "stepped up" and played one of the power teams.  9-1 Thomas More had a shot.  8-2 Thomas More most likely doesn't.

I really, really WISH you were wrong here.  Sigh.  Oh well. 

[SaintsFAN]

*One more reason why SOS is complete garbage.  You can't ever know that the team you scheduled a home and home with in 2011 is going to be any good in 2014.  What makes it worse is the dialogue that happens afterward about how Team A (who got selected) went out and scheduled and played good teams while Team B (who did not get selected) played a weak schedule.  Team A didn't schedule good teams, the teams on Team A's schedule turned out to be good- there's a huge difference.

The only way to really guarantee an great SOS opponent is to schedule Mt. Union, UMHB, Wesley, Linfield, Wheaton, North Central, Wabash or Whitewater. Franklin may be one of those (they were 5-5 in 2005...)They haven't had a meh SOS record in a decade. Linfield was 6-3 a few years ago. Wesley was .500 a decade ago. So was Whitewater (wow...times change!). Wabash was 6-4 in 2004. I don't know if I'd trust another MIAC school, an OAC school or a WIAC school to guarantee a 7 win or more opponent.

I may have missed one or two...


(My thinking is 6-4 doesn't help SOS much at all).

Thomas More has played an undefeated Wesley, an undefeated W&J and a 7 win Waynesburg team.  Their SOS advantage from those games will be completely nullified by playing in a weaker than usual PAC. 

[hazzben]

In general, if a school schedules a team that recently (say, last 5 years) has been above average in a strong, typically deep conference (MIAC, WIAC, CCIW, E8, etc.) or very good in a middling conference (IIAC, NWC, etc.) I give them props for trying to go out and schedule good teams.

Right, but a above average team in those conferences still could be 6-4 or 5-5. That doesn't help much. Maybe in OOWP, but then there's the OAC where everyone plays nine conference games, so the OOWP and OWP tend to normalize to .500 anyway.

5-5 would be about the definition of average. Possibly even meaning they went 3-5 in conference. For me, 'above average' means you're usually above .500 in conference play over the course of those 5 years. And typically, for those teams, from those conferences, they're probably doing pretty well out of conference then as well.

And notice I didn't list the OAC. Basically, unless you're scheduling Mount, I'm just not sure you can count on the team you've scheduled being good. That's not to say the OAC #2 isn't good. It's just to say we usually have no idea who that #2 is going to be 3-4 years from now.

[ExTartanPlayer]

*One more reason why SOS is complete garbage.  You can't ever know that the team you scheduled a home and home with in 2011 is going to be any good in 2014.  What makes it worse is the dialogue that happens afterward about how Team A (who got selected) went out and scheduled and played good teams while Team B (who did not get selected) played a weak schedule.  Team A didn't schedule good teams, the teams on Team A's schedule turned out to be good- there's a huge difference.

The only way to really guarantee an great SOS opponent is to schedule Mt. Union, UMHB, Wesley, Linfield, Wheaton, North Central, Wabash or Whitewater. Franklin may be one of those (they were 5-5 in 2005...)They haven't had a meh SOS record in a decade. Linfield was 6-3 a few years ago. Wesley was .500 a decade ago. So was Whitewater (wow...times change!). Wabash was 6-4 in 2004. I don't know if I'd trust another MIAC school, an OAC school or a WIAC school to guarantee a 7 win or more opponent.

I may have missed one or two...


(My thinking is 6-4 doesn't help SOS much at all).

Thomas More has played an undefeated Wesley, an undefeated W&J and a 7 win Waynesburg team.  Their SOS advantage from those games will be completely nullified by playing in a weaker than usual PAC.

Thomas More's SOS also took a dent from Hanover losing all of their close games (48-42 to RHIT, 31-28 to Defiance, 20-14 to Bluffton, 37-35 to Anderson) and currently sitting at 1-7 instead of their usual .500-level play. 

Besides the PAC's down season, another thing that hurt TMC is that the newly-unbalanced PAC schedule ended up with them playing all of the bottom-feeders and missing 6-3 Bethany.  Give them a game with Bethany at 5-4 or 6-3 instead of winless Grove City and they get another little boost.

Which, as we're saying all over the place in this conversation, shows how silly it is to use the SOS metric as such a big component here.  So much of this is influenced by things that are out of a team's control and rather unpredictable.  Thomas More's SOS could be .05 points higher with a couple of those random bounces on the schedule.

I'm not feeling all that bad for TMC, or any two-loss team, as I'm a believer that any wild-card bids are a crapshoot and that everybody's (or most everybody) has got a path to the dance by winning their conference. 

[AO]

I think what we're all getting here is that we want to see Margin of Victory added to the data.

[smedindy]

In general, if a school schedules a team that recently (say, last 5 years) has been above average in a strong, typically deep conference (MIAC, WIAC, CCIW, E8, etc.) or very good in a middling conference (IIAC, NWC, etc.) I give them props for trying to go out and schedule good teams.

Right, but a above average team in those conferences still could be 6-4 or 5-5. That doesn't help much. Maybe in OOWP, but then there's the OAC where everyone plays nine conference games, so the OOWP and OWP tend to normalize to .500 anyway.

5-5 would be about the definition of average. Possibly even meaning they went 3-5 in conference. For me, 'above average' means you're usually above .500 in conference play over the course of those 5 years. And typically, for those teams, from those conferences, they're probably doing pretty well out of conference then as well.

And notice I didn't list the OAC. Basically, unless you're scheduling Mount, I'm just not sure you can count on the team you've scheduled being good. That's not to say the OAC #2 isn't good. It's just to say we usually have no idea who that #2 is going to be 3-4 years from now.

What I meant was in the OAC and NCAC, with nine-game schedules, you're playing everyone once. The conference records for the teams under the runner up (if they have just one conference loss) will be slightly under .500. They have one extra game, and it's usually happened that even if they sweep the non-conference you're just going to get to a game over .500 in OWP at best.

Example - JCU this year, if they lose to mount will have played nine OAC opponents. Their nine opponents will have an OAC record of 37-44. The OAC went 6-3 in non-conference outside of JCU so JCU's OWP is 43-47 before taking their game into account. They played St. Vincent who at 3-5 makes them now 46-52 with two games to go. Conceivably, a runner up OAC team could get to 55-45 but that's with an OAC sweep and the runner up being the only team to beat their opponent. A huge leap. Normally, they're around .500 or just shy in OWP.

It's the OOWP that's the variable, but you can't expect everyone to schedule killers. I mean, you could tell Marietta, Muskingum and Wilmington to go play Wabash, W&J and Wittenberg, but I doubt if they would and I doubt if those three schools would schedule the dregs anyway. JCU is helped by Muskingum playing Waynesburg and Otterbein playing SJF, but Marietta played Kenyon and Capital played Earlham. Urp. 

But you are right that the 2nd or 3rd place OAC teams tend to rotate - now it's 'Berg and JCU. In the past it was ONU and B-W. Sometimes Capital and Otterbein. You just don't know.

[hazzben]

And I'd agree 100%, SoS is tough for 10 team leagues.

It's one of the reasons I wasn't torn up (but definitely annoyed) that Mac jumped over to the MWC rather than rejoin the MIAC. I'd much rather be able to schedule a quality non-con than have another team on the schedule that would likely finish somewhere south of 3-7.

[wally_wabash]

I think what we're all getting here is that we want to see Margin of Victory added to the data.

No, I don't think that's a good idea really at all.   

[Ralph Turner]

Updating your conference-by-conference list of all potential one-loss teams that could end up in Pool C...the list of possibilities reallllllly thinned out this week thanks to a few upsets.

(*Note: I know that playoff selection for Pool C does not go by conference, it's just the easiest way for me to tick off all possible one-loss teams in Pool C, thus giving an idea how many 1-loss teams may end up on the board)

...

Narrowing the above list down, here are conferences that realistically could produce a 1-loss Pool C candidate:

CC - Muhlenberg
MAC - DelVal/Widener loser
MWC - Carroll
NCAC - Witt/Wabash loser
OAC - JCU/Mount loser
ODAC - Emory & Henry (needs to beat Guilford in finale)
UMAC - St. Scholastica (would have to lose to Greenville)
WIAC - UW-Oskhosh*

I'll go a step further and say that the MAC runnerup, NCAC runnerup, and OAC runnerup are the "locks" to be in Pool C with one loss.  Also a strong possibility for Muhlenberg, Carroll, Emory & Henry (although two of those teams must beat a reasonably challenging opponent to secure their 9-1 records).  UW-Oshkosh could be 6-4, but would have a 6-1 Division III record.  In summary, there will probably be no more than seven teams in Pool C with one loss, and possibly as few as four or five.  The door should be open for a few two-loss candidates.

Plus a few Pool B leftovers. Wesley is in for the first Pool B bid, but in the absence of the first RR, here are some Pool B's that may be on the Table.

Framingham State
Centre
Texas Lutheran (which must rank above Thomas More to have a chance).


+1 xTP!

[AO]

In general, if a school schedules a team that recently (say, last 5 years) has been above average in a strong, typically deep conference (MIAC, WIAC, CCIW, E8, etc.) or very good in a middling conference (IIAC, NWC, etc.) I give them props for trying to go out and schedule good teams.

Right, but a above average team in those conferences still could be 6-4 or 5-5. That doesn't help much. Maybe in OOWP, but then there's the OAC where everyone plays nine conference games, so the OOWP and OWP tend to normalize to .500 anyway.

5-5 would be about the definition of average. Possibly even meaning they went 3-5 in conference. For me, 'above average' means you're usually above .500 in conference play over the course of those 5 years. And typically, for those teams, from those conferences, they're probably doing pretty well out of conference then as well.

And notice I didn't list the OAC. Basically, unless you're scheduling Mount, I'm just not sure you can count on the team you've scheduled being good. That's not to say the OAC #2 isn't good. It's just to say we usually have no idea who that #2 is going to be 3-4 years from now.

What I meant was in the OAC and NCAC, with nine-game schedules, you're playing everyone once. The conference records for the teams under the runner up (if they have just one conference loss) will be slightly under .500. They have one extra game, and it's usually happened that even if they sweep the non-conference you're just going to get to a game over .500 in OWP at best.

Example - JCU this year, if they lose to mount will have played nine OAC opponents. Their nine opponents will have an OAC record of 37-44. The OAC went 6-3 in non-conference outside of JCU so JCU's OWP is 43-47 before taking their game into account. They played St. Vincent who at 3-5 makes them now 46-52 with two games to go. Conceivably, a runner up OAC team could get to 55-45 but that's with an OAC sweep and the runner up being the only team to beat their opponent. A huge leap. Normally, they're around .500 or just shy in OWP.

It's the OOWP that's the variable, but you can't expect everyone to schedule killers. I mean, you could tell Marietta, Muskingum and Wilmington to go play Wabash, W&J and Wittenberg, but I doubt if they would and I doubt if those three schools would schedule the dregs anyway. JCU is helped by Muskingum playing Waynesburg and Otterbein playing SJF, but Marietta played Kenyon and Capital played Earlham. Urp. 

But you are right that the 2nd or 3rd place OAC teams tend to rotate - now it's 'Berg and JCU. In the past it was ONU and B-W. Sometimes Capital and Otterbein. You just don't know.

Smed.  You have to exclude the John Carroll games in your calculations.   The OWP will always be exactly .500 counting only conference games since the OAC plays a round robin.  36-36.

I think what we're all getting here is that we want to see Margin of Victory added to the data.

No, I don't think that's a good idea really at all.   

of course it is.  You don't have to give extra credit to Mount for winning 100-0 versus 70-0.  People love to point out exceptions to the margin of victory rule rather than point out all the cases where the margin gave us meaningful predictive data.