Pool C in 2016

[Ralph Turner]

So wally, when you do these projected brackets mid-season, do you usually do it without projecting team's future records? I'm asking because you had SJFU as your second Pool C team, but they're going to probably have two losses if they're not Pool A, which would obviously affect their odds.

I'm currently trying to assign probabilities for teams to get in based on regressions of prior performances (specific to Pool C, Pool A is pretty straight-forward), so hopefully I'll have something meaningful to add by the end of the week.

If you can mathematically model the selection committee's choices you should get a Nobel.   

I think of these mid-season projections as more of a "if the season ended today" kind of snapshot.  When I started doing this a few years ago, mostly I was just curious to see how the at-large picture evolves over the last month or so of the season.  I don't try to project future results too much, except in cases where two teams (generally from the same conference) are kind of hitched together but will play one another at a later date.  I think I've had this happen with the MIAC a couple of times, particularly when Bethel was a serious player there along with STT, STJ, and C-M.  Even trying to get the current "snapshot", it doesn't make sense to have four teams from the same league in play because that's just not a thing that's going to happen.  I could/should have done this with HSU/ETBU this week, which would have bumped Muhlenberg up onto the table (the Mules would not have been selected in any case). 

You're right about SJF.  9-1 SJF would be a conference champion.  8-2 SJF...well, 8-2 SJF teams have actually had good luck in these situations, so maybe that's not a terrible spot for the Cardinals after all.

I always have trouble denying the East Region (for football especially) an at large bid when there are enough at large bids for 1 per region.  That being said, the East is a mess and I think that 10-0 CWRU knocks out the 6th Pool C team on the table.

When the committee sits down, I think that a 9-1 HSU/ETBU and 10-0 CWRU come out from the South.  The West will get 2 bids (Johnnies and UW-someone). The North and East will get 1 Pool C bid each.

[wally_wabash]

CWRU still has to get to 10-0.  The hay is very, very far from being in the barn where that is concerned. 

I think CWRU are a pick 'em (if not a slight dog) in St. Louis this weekend and I think they're a definite underdog against Westminster.  Favored against CMU, but when you play CMU you're always in danger of getting Benger'd. 

[jknezek]

CWRU still has to get to 10-0.  The hay is very, very far from being in the barn where that is concerned. 

I think CWRU are a pick 'em (if not a slight dog) in St. Louis this weekend and I think they're a definite underdog against Westminster.  Favored against CMU, but when you play CMU you're always in danger of getting Benger'd.

The important thing to remember is CWRU's opponents so far this season are 14-35, with St. Vincent being the only team above .500 at 4-3.  CWRU's last three opponents are 16-5, with the worst being CMU at 4-3. At this point, all we really know about CWRU is they are above average and have played a very below average schedule. Case has the 223rd ranked schedule in DIII with a .409 SOS. Since we don't count the NESCAC schools, that's 223 out of 237. Almost the 95th percentile. Now of course it will improve with their last 3 games, and winning games is important, but at 7-0 against their slate, I don't feel real confident that they are heading for the playoffs.

[HansenRatings]

CWRU only has about a 1-in-4 chance of finishing undefeated according to my projections, so it's not likely, but really not that unlikely.

If you're interested, I've projected how the SOS rankings will look by season's end. I have italicized every team my model believes is most likely to earn their conference's Pool A bid, had underlined UMHB as the most likely Pool B team, and I have bolded every potential Pool C team that is likely to finish at or below two losses.

Wabash's SOS is probably going to be the worst SOS of any team with two or fewer losses, with only CWRU and Eureka even close.

UW-Oshkosh is going to get in. They'll probably finish second in the country in SOS, and there's only a 10% chance they lose any game down the stretch.

Fisher's 8-2 magic might hold too, depending on their ranking by the RAC--I'm projecting them to have the third-highest SOS among likely Pool C teams (behind UW-O and UW-P). If any other East team at 8-2 gets in, I would bet on Hobart, with a projected SOS at #34

Wheaton has the next highest SOS, and I think IWU will be regionally-ranked, giving them another boost.

I'm not so sure about Dubuque's Pool C chances. Teams likely to be ranked above them in SOS also with one loss from Pool C: Franklin, Hardin-Simmons, SAA runner-up, Wheaton, UW-O. They'll probably need to be ranked above SJU in the West to have a shot.

[ExTartanPlayer]

CWRU only has about a 1-in-4 chance of finishing undefeated according to my projections, so it's not likely, but really not that unlikely.

Sounds about right to me.

[wally_wabash]

If you're interested, I've projected how the SOS rankings will look by season's end. I have italicized every team my model believes is most likely to earn their conference's Pool A bid, had underlined UMHB as the most likely Pool B team, and I have bolded every potential Pool C team that is likely to finish at or below two losses.

Am I understanding the linked sheet is what you're projecting to be final NCAA SOS figures?  I think Wabash's OWP can be, at worst, 0.444.  That difference, with the 2/3 weight on that portion of the math, would be about a 0.06 boost in NCAA SOS, which is a pretty big jump.  I'm not saying that gets Wabash's SOS totally out of jail, but it does certainly get them out of the 220s.

[HansenRatings]

If you're interested, I've projected how the SOS rankings will look by season's end. I have italicized every team my model believes is most likely to earn their conference's Pool A bid, had underlined UMHB as the most likely Pool B team, and I have bolded every potential Pool C team that is likely to finish at or below two losses.

Am I understanding the linked sheet is what you're projecting to be final NCAA SOS figures?  I think Wabash's OWP can be, at worst, 0.444.  That difference, with the 2/3 weight on that portion of the math, would be about a 0.06 boost in NCAA SOS, which is a pretty big jump.  I'm not saying that gets Wabash's SOS totally out of jail, but it does certainly get them out of the 220s.

Thanks for pointing that out. I had an error in my lookup functions.

[HansenRatings]

Should be fixed now. Wabash is at #177

[wesleydad]

Wally, thanks for doing this.  As things now stand in the NJAC, Wesley has no chance of being in the pool C discussion.  They win out, they win the NJAC.  Another loss and they have no shot at anything.  Well, unless all kinds of crazy goes on in the NJAC and they end up in some tie for the AQ and somehow win the tiebreaker.  From what I can see the only teams left in the pool C discussion from the NJAC are Salisbury and Frostburg, but since they play each other someone gets a second loss.  If either of them wins out they win the AQ, so in the end the NJAC has no viable pool C candidates.

[wally_wabash]

Thanks, wesleydad.  One benefit of doing these early projections is getting alerted to specific situations like this one in the NJAC. 

I wouldn't throw dirt on any 2-loss NJAC runner up just yet.  There's a lot of things that can happen in the East region that would allow for a 2-loss NJAC runner up to be first in the at-large line.  Stranger things have happened. 

[Neverwas]

Come for the content stay for the one liners. 

Thanks for all the time and effort on this week's projection (and the weeks ahead), Wally.  This is always fun to read.  Plus, I truly love the one-liners/catchphrases. 

[D3MAFAN]

Thanks, wesleydad.  One benefit of doing these early projections is getting alerted to specific situations like this one in the NJAC. 

I wouldn't throw dirt on any 2-loss NJAC runner up just yet.  There's a lot of things that can happen in the East region that would allow for a 2-loss NJAC runner up to be first in the at-large line.  Stranger things have happened.

Personally, I can't see a three loss Wesley team getting to the board and getting taken as a Pool C, they are in control of their own destiny and would need to win out. Regarding the remaining NJAC, the only two loss team would be Salisbury or CNU and that would be a reach, Salisbury would need Albright to finish strong and become RR. Salisbury and CNU would also need Frostburg, CNU/Salisbury or Wesley to be RR at the end of the year to give them 2 wins over RR opponents.  Maybe if Rowan wins out and sneaks back into the discussion.

Edit: I revised this paragraph so many times, now created my own conundrum 

[HansenRatings]

I'm guessing Wally is waiting on the Regional Rankings to make a post, so I'll start things off this week. I recently tried to model which teams the selection committee is most likely to select for Pool C contention, after accounting for Pools A & B. The model isn't perfect, but over the last five seasons, it would have predicted all but one or two of the selections correct each season. I'll probably have a blog post later in the week to go into specific details, but essentially it tries to predict regional rankings to predict who will be selected. I used this in conjunction with my regular season-end projection model to give each team a probability of being selected to the tournament. Here are my results:



Most interesting thing to me here is that my model thinks that Muhlenberg's resume will be closer to that of a typical playoff participant than the HSU/ETBU winner.

[wally_wabash]

Travel day yesterday, so I'm a day behind.  I'll have freshly updated eliminator tables sometime tonight and then I'll crank out a projection tomorrow after the RRs are released.  This week's at-large projection is going to look a little different than last week.  I think I'm going to try and integrate more of my eliminator stuff into the projection this week- so for instance, (spoiler) Wesley is eliminated from Pool C this week as I can't find a scenario where they can be 8-2 and not win the NJAC AQ (h/t to wesleydad for pointing that out last week).  So even though I'm going to project Salisbury as the NJAC champion and Wesley may well be the first or second highest ranked at-large team in the East region in Wednesday's official rankings, I'm going to exclude Wesley from the at-large analysis. 

I'm guessing Wally is waiting on the Regional Rankings to make a post, so I'll start things off this week. I recently tried to model which teams the selection committee is most likely to select for Pool C contention, after accounting for Pools A & B. The model isn't perfect, but over the last five seasons, it would have predicted all but one or two of the selections correct each season. I'll probably have a blog post later in the week to go into specific details, but essentially it tries to predict regional rankings to predict who will be selected. I used this in conjunction with my regular season-end projection model to give each team a probability of being selected to the tournament. Here are my results:

Most interesting thing to me here is that my model thinks that Muhlenberg's resume will be closer to that of a typical playoff participant than the HSU/ETBU winner.

This is cool.  Don't sleep on the CC runner up, gang.  Every year we kind of forget about the CC past Hopkins, and every year that league's 9-1 runner up is right in the mix at the last couple of rounds of voting. 

[ITH radio]

Should be interesting to see what comes out on Weds. My guess for ER would be something like this:

1. Alfred (highest SOS of 8-0 teams, arguably best conf, etc)
2. Stevenson 8-0 (better SOS than SLU)
3. SLU 8-0 (could see them lower bc of bad SOS but usually comm rewards 8-0 league leaders from the "top 4"
4. SJFC 7-1 (best SOS of 1 loss ER teams)
5. Hobart 7-1 (bc of high SOS)
6. Salisbury 7-1
7. Frostburg 7-1 (w. solid SOS of 115)
8. WNE 8-0 (maybe crazy bc of low SOS but have seen undefeated teams from NE get in Wk 10)
9. Salve 8-0 (see above)
10. Husson 6-1 (low SOS but only 7 point loss to #1 ALF gets them on board IMO)

Ironically only really two teams on this list has a Pool C shot (SLU, if the lose to HOB and beat WPI in Wk 11 to go 9-1 and ALF if they lose to SJF and end up 9-1 - even in that scenario, I could see SLU being left out bc of SOS).