Pool C

[fantastic50]

When it comes down to the last few picks, the Hoopsville guys pick three power conference teams with tough schedules but sub-.700 winning percentages (Amherst, North Central, UW-RF). On the other hand, I took three teams with weaker schedules but better records (LeTorneau, Loras, and Springfield).

I could see the committee going either way, or some combination. Last year's UW-O pick was along the lines of where Dave, etc. went, while in some other years, any team whose WP started with a 6 needed to be very strong in other criteria.

As always, Selection Monday will be fun!

Based on this approach, you would likely have taken St. Norbert last year. A non AQ team which, despite a lofty record, was never really considered because the overwhelming majority of their high win total was posted against extremely weak fellow conference teams. Teams like this would often finish in the middle of the pack in power conferences.

My final Pool C picks...

1) Babson (0.926, 0.585, 5-2, NE)
2) Susquehanna (0.800, 0.559, 5-5, MA)
3) Tufts (0.769, 0.570, 4-3, NE)
4) Rochester (0.840, 0.538, 4-2, EA)
5) Whitworth (0.852, 0.546, 1-3, WE)
6) Williams (0.704, 0.602, 7-5, NE)
7) Amherst (0.708, 0.602, 5-5, NE)
8) Wesleyan (CT) (0.760, 0.561, 4-3, NE)
9) New Jersey City (.750, 0.533, 6-3, AT)
10) UW-Whitewater (0.769, 0.568, 1-3, CE)
11) Emory (0.720, 0.551, 2-3, SO)
12) Hope (0.800, 0.525, 2-1, GL)
13) Salisbury (0.741, 0.548, 3-4, MA)
14) Cabrini (0.760, 0.532, 4-4, AT)
15) Mt St Joseph (0.760, 0.522, 2-3, GL)
16) Skidmore (0.731, 0.525, 5-1, EA)
17) St Lawrence (0.760, 0.524, 3-5, EA)
18) St Thomas (MN) (0.731, 0.530, 2-2, WE)
19) Augustana (.704, 0.543, 2-3, CE)
20) UW-Eau Claire (0.680, 0.572, 3-3, CE)
21) Keene State (0.679, 0.578, 1-3, NE)

I had to go back & look, but my missed picks last year were not in the high WP / low SOS category.  Both LeTourneau (.852 / .511 / 2-2 and proj. 3-4) and Loras (.731 / .543 / 2-3 and proj. 3-3) look a lot better than last year's St Norbert team (.792 / .506 / 1-1), but there are also a lot more competitive bubble teams this year, so I won't be surprised if both get left out.

[lmitzel]

So as someone who is trying to learn the selection process (I am well versed on how D-III hockey does it and I follow D-III hoops, but this is my first real time following the complete process), I am intrigued by Hobart.  People seem split on them getting in.  They have a very high WIN%, but their SOS would be the weakest of the Pool Cs (and weaker than many teams left out) and would be the only Pool C team with only one win against ranked teams (again less than many teams being left out).  Can WIN% be enough to get a team in?  On the hockey side I would predict that their SOS and RNK would sink them, but here people seem split 50/50.

Thanks, I'll hang up and listen!

Steve Wood
Plattsburgh State '99
Penn State '95

In an attempt to answer that question, Steve...

I've been slowly figuring out the process in the last couple years, but a lot of it centers on what the selection committee decides to prioritize. For reference, here's a couple charst D3Hoops published last year (source article here because you might be able to see them better):



Hobart looks to slot in on both charts in between the 50 percent and 90 percent, but closer to the 50 than the 90. Part of that may be a result of, at least according to the mock, them sitting at the table for the entire selection process. I'm sure that .524 mark and the only three RRO games were a big reason why they sat there. But again, you look at the chart, teams with worse win percentages and SOS's have made it. Last year the big deal was Oshkosh making it despite a .630 win percentage only because they had an absurd SOS. It's not out of the question that the committee opts to go to the other extreme, but I feel like SOS is more of a key.

Of course, people like Q, fantastic50, Dave, or Ryan could answer this far better than I could since they understand the process better and deeper than I do.

[fantastic50]

When it comes down to the last few picks, the Hoopsville guys pick three power conference teams with tough schedules but sub-.700 winning percentages (Amherst, North Central, UW-RF). On the other hand, I took three teams with weaker schedules but better records (LeTorneau, Loras, and Springfield).

I could see the committee going either way, or some combination. Last year's UW-O pick was along the lines of where Dave, etc. went, while in some other years, any team whose WP started with a 6 needed to be very strong in other criteria.

As always, Selection Monday will be fun!

Based on this approach, you would likely have taken St. Norbert last year. A non AQ team which, despite a lofty record, was never really considered because the overwhelming majority of their high win total was posted against extremely weak fellow conference teams. Teams like this would often finish in the middle of the pack in power conferences.

And that's exactly where River Falls (4th) and, if you think about it, Amherst, finished. Amherst finished in a 5-way tie for 1st and they could've been the 5th seed. NCC, as you know, finished 3rd.

Here's a thought experiment.  If I picked the best 21 teams from Pool C, as in those who be most likely to advance deep into the tournament based on the way they are playing right now (think of the Alabama vs Ohio State argument in December), here is what my predictive model would pick:
1) UW-P
2) Whitman
3) St John's
4) Marietta
5) Bethel
6) North Central
7) UW-O
8) Wesleyan
9) UW-RF
10) Ohio Northern
11) Ripon
12) LeTourneau
13) IWU
14) Hamilton
15) Emory
16) CNU
17) Ohio Wesleyan
18) St Olaf
19) Swarthmore
20) Amherst
21) Loras
---
22) Wooster
23) Middlebury
30) Albright
31) Hobart
33) F&M
43) Springfield
63) NJCU
64) Keene St

[fantastic50]

Comparing Pool C picks, the Hoopsville selections, TitanQ, and I all agree on 17 teams (listed here in consensus order)...
1) Hamilton
2) Wesleyan
3) UW-Platteville
4) Whitman
5) Swarthmore
6) St John's
7) Middlebury
8) Wooster
9) Marietta
10)New Jersey City
11) Emory
12) Christopher Newport
13) St Olaf
14) UW-Oshkosh
15) Albright
16) Franklin & Marshall
17) Illinois Wesleyan

We're all over the place for the last four...
North Central (Hoopsville & TitanQ)
Amherst (Hoopsville & TitanQ)
Hobart (Hoopsville & me)
Loras (TitanQ & me)
Springfield (TitanQ & me)
LeTorneau (me)
UW-River Falls (Hoopsville)

On North Central/IWU, remember, it's almost certain IWU can't get in unless NCC does.  So if you don't have NCC in, you realistically don't have IWU in either.

I was projecting IWU ahead of North Central in the final regional rankings.  Are you thinking that NCC gets to the table first because of the 2-1 head-to-head (which is not in my model), or did Dave have inside info from the committee that this is, in fact, the case?

[Greek Tragedy]

In the CCIW board a post shows NCC ahead in 3 of 4 categories: SOS, vRRO and H2H.  I would put NCC ahead of IWU too, but I'm not a math professor or a Hall of Famer.

[nescac1]

Again, very interesting stuff fantastic50!  Once brackets are out I'd be really curious to see how your predictive model sees things shaking out ...

[fantastic50]

Again, very interesting stuff fantastic50!  Once brackets are out I'd be really curious to see how your predictive model sees things shaking out ...

I had seen the numbers, didn't know how accurate they were. I'm new to these boards. In the past, how accurate have these projections been to actual Pool C bids?

If I remember correctly, fantastic50 was only off by one bid last season.

I missed two last year, and one the year before that.  There are several folks here who are just as good or better with calling those berths on this last weekend.  The tools that I have available are most beneficial a week or two ahead of the selections, because I can predict final SOS, etc. with decent accuracy.

Looking at the consensus, I expect to be outpicked this year by Dave, Bob, etc.  I think that the value of my simulations comes in earlier, as I have predicted since late January that the Centennial would likely get three bids (Swarthmore, Hopkins, F&M), even though the Diplomats had a .493 SOS at the time.

[nescac1]

I should have been more clear, I mean I'd be curious to see how your model sees teams advancing through the brackets once match-ups are posted ... if it does that sort of thing. 

[fantastic50]

In the CCIW board a post shows NCC ahead in 3 of 4 categories: SOS, vRRO and H2H.  I would put NCC ahead of IWU too, but I'm not a math professor or a Hall of Famer.

LOL, well said; I'm converted!  The Titans' sole advantage is one extra win, so they stay ahead only if the committee cringes at the "6" in NCC's .692 WP, which has happened in some years, but seems unlikely after last year.

[fantastic50]

For what's worth, the consensus of my mathematical modeling class (building predictive models as a graded mini-project) was as follows.  This was purely driven by past & current data on WP, SOS, vRRO, and RR, as very few of them follow the D3 game beyond our campus, if at all.

1) Hamilton
2) Wesleyan
3) UW-Platteville
4) Middlebury
5) Wooster
6) Marietta
7) Whitman
8) Emory
9) St. John's
10) Swarthmore
11) NJCU
12) Christopher Newport
13) LeTourneau
14) UW-O
15) Hobart
16) Albright
17) Salem St.
18) F&M
19) IWU
20) St. Olaf
21) UW-RF

[WUPHF]

Based on this approach, you would likely have taken St. Norbert last year. A non AQ team which, despite a lofty record, was never really considered because the overwhelming majority of their high win total was posted against extremely weak fellow conference teams. Teams like this would often finish in the middle of the pack in power conferences.

That is incredible interesting.

Thanks for sharing.

[Greek Tragedy]

Does someone have Oshkosh's final numbers from last year? I would like to see a comparative with River Falls this year.

[fantastic50]

Does someone have Oshkosh's final numbers from last year? I would like to see a comparative with River Falls this year.

UW-O was .630 / .602 / 5-7

[Greek Tragedy]

Does someone have Oshkosh's final numbers from last year? I would like to see a comparative with River Falls this year.

UW-O was .630 / .602 / 5-7

Thanks. River Falls is behind in two of those three categories. I think they are something like .586 and 3-7.

[fantastic50]

So as someone who is trying to learn the selection process (I am well versed on how D-III hockey does it and I follow D-III hoops, but this is my first real time following the complete process), I am intrigued by Hobart.  People seem split on them getting in.  They have a very high WIN%, but their SOS would be the weakest of the Pool Cs (and weaker than many teams left out) and would be the only Pool C team with only one win against ranked teams (again less than many teams being left out).  Can WIN% be enough to get a team in?  On the hockey side I would predict that their SOS and RNK would sink them, but here people seem split 50/50.

Thanks, I'll hang up and listen!

Steve Wood
Plattsburgh State '99
Penn State '95

In an attempt to answer that question, Steve...

I've been slowly figuring out the process in the last couple years, but a lot of it centers on what the selection committee decides to prioritize. For reference, here's a couple charst D3Hoops published last year (source article here because you might be able to see them better):



Hobart looks to slot in on both charts in between the 50 percent and 90 percent, but closer to the 50 than the 90. Part of that may be a result of, at least according to the mock, them sitting at the table for the entire selection process. I'm sure that .524 mark and the only three RRO games were a big reason why they sat there. But again, you look at the chart, teams with worse win percentages and SOS's have made it. Last year the big deal was Oshkosh making it despite a .630 win percentage only because they had an absurd SOS. It's not out of the question that the committee opts to go to the other extreme, but I feel like SOS is more of a key.

Of course, people like Q, fantastic50, Dave, or Ryan could answer this far better than I could since they understand the process better and deeper than I do.

Here's a chart (admittedly a bit "busier") with 2013-17 teams (small o's if in, small x's if out).  The stars are for the 2018 candidates (with Whitman off the top of the chart).  The lines are the 50/50 bubble for teams with the given number of wins vs RRO.  Hobart is the red "x" at .808 WP and .525 SOS.  They're above the 50/50 line, but not far enough above to be anything close to a "lock."