NE Region General Questions

[Ryan Scott (Hoops Fan)]

Win/Loss percentage is an important but can easily be out-weighed by a weak SOS... resulting in those voting to think... sure they have one loss... their SOS is in the bottom half of the region.

Just a thought.

Yeah, but has it really been such a disparity like this year?  They'd move some teams aside for a really strong SOS and more losses, but I just don't recall the kind of disparity that we see with MIT and Albertus happening before.  I could be wrong.

[Dave 'd-mac' McHugh]

It's the first regional ranking... we can't make that kind of deduction from that! Plus, we had similar moves last year and the year before... it was actually kind of welcoming instead of based more on W/L that had teams with really easy records ranked a bit too high.

Let's see what happens this week... and then next week before we try and figure out what is getting more weight.

[Hugenerd]

Blindly using an OWP value is just as egregious as blindly using a WP.  Is it really enough to play teams with good records and lose (and I say 'play' instead of schedule because the LEC and NESCAC each have multiple ranked teams in their conference, which really boosts their values). For example, one may think that a 17-5 record with a 0.54 OWP is really great, until one realizes that the weighted OWP of the 17 games that team has won is only 0.44.  To illustrate this point, in the table below, I have calculated OWP values of each of the NE ranked teams (from last week) in games they have won, as well as their record vs. 0.500 or better teams (where the head-to-head results between teams are not counted. For example, if I were doing a calculation for Amherst, and the team I was considering was Brandeis, the record I would use is 11-9, instead of 12-9, because the head-to-head result would not count).

Team	OWP	OWP in wins	(OWP in wins)-OWP	Record vs. 0.500+
Amherst	0.603	0.577	-0.026	12-2
Middlebury	0.571	0.522	-0.049	12-2
RIC	0.601	0.501	-0.100	11-5
West Conn	0.568	0.517	-0.051	10-2
WPI	0.616	0.608	-0.008	12-3
East Conn	0.536	0.442	-0.094	7-5
MIT	0.486	0.482	-0.004	12-1
Keene St.	0.553	0.491	-0.062	9-4
Wesleyan	0.501	0.447	-0.054	9-3
Tufts	0.548	0.470	-0.078	7-5
Becker	0.484	0.454	-0.030	8-4
Albertus Magnus	0.433	0.434	+0.001	6-1

After you take out losses from the weighted OWP calculation, the only 2 numbers that jump out at you, on the high side, are WPI and Amherst.  On the low end, you have East Conn, Albertus Magnus, Becker, and Wesleyan. Everyone else, however, is essentially 0.500+-0.200, which, in my opinion, is just noise, and at that point, I think you have to take a strong look at record and other primary criteria.  If I were ranking them, which obviously I am not in any official sense, I would put Amherst as a clear number 1.  After that, things get a bit less clear, but I would put them in the following groups.  In the 3 slots behind Amherst, I would put 2. Middlebury, 3. MIT, and 4. West Conn.  MIT and West Conn could be flipped, but I had West Conn below MIT because of their 3 losses to average or sub-par teams. The next group of three I would have is 5. WPI 6. Keene St. and 7. RIC (Keene has the head-to-head win vs. RIC).  In the final group, I would put 8. Albertus Magnus, 9. Wesleyan, 10. East Conn, and 11. Becker.  I think Tufts will drop out and did not calculate figures for everyone else, so I dont know who would replace them.  Again, this is just how I would use the OWP values, and I think it makes sense.  Otherwise, as you can see from the table, it is possible to boost your overall OWP ~0.100 (which from what I hear is extremely significant) by losing 4-6 games, without really showing you can beat the types of teams you are losing to.

[augie_superfan]

HN,

I would agree that the first regional rankings weren't perfect.  Hopefully they will continue to evolve and the committee members will take a closer look at results vs. common teams and those things to get a better feel for the teams...not that they shouldn't already have that feeling 20 some games into the season.  My problem with OWP stems from relying on this "average" number.  By taking the average, you really lose valuable information about the difficulty of the games played.  Simple example below of 2 different schedules, both with an OWP of 0.500

Sched A:  games vs. teams with win % of: 0.8, 0.8, 0.3, 0.3, 0.3
Sched B:  games vs. teams with win % of: 0.5, 0.5, 0.5 ,0.5, 0.5

So, if you have 2 teams that went 4-1 versus these different schedule, their "numbers" would look the same but they actually played a schedule of different difficulty (in my opinion).  Hopefully that makes some sense and is just one drawback to the OWP/OOWP idea.

Also, to go off something you have alluded to earlier, we seem to make a big deal between SOS values of like 0.52 and 0.48 or something like that.  How different are these values really?  It would be nice to look at the distribution of these numbers, probably peaked around 0.5 but wonder how tightly they are packed around that number (i.e. standard deviation).  Knightslappy?

[walzy31]

Win/Loss percentage is an important but can easily be out-weighed by a weak SOS... resulting in those voting to think... sure they have one loss... their SOS is in the bottom half of the region.

Just a thought.

Yeah, but has it really been such a disparity like this year?  They'd move some teams aside for a really strong SOS and more losses, but I just don't recall the kind of disparity that we see with MIT and Albertus happening before.  I could be wrong.

Similar thing happened last year when Amherst was 21-0 but had played something like 16 games at home and had a low SOS (Tufts and Wesleyan were down and the RIC game got snowed out) which resulted in them being 5th in the region. The only stand out win was at home over Williams who came in at #2 in the regional rankings with one loss. Amherst's SOS would have been higher if they had player more away games, and this year Amherst is benefiting from more away games and a better Opponents SOS.

[amh63]

Guys who play with numbers.....remember the weighting factors for home and away games were adjusted this year in a more "even" manner.   Thus the SOS impact for away and home games will be different between last year and this year.....but your point is well taken Walzy.

[Dave 'd-mac' McHugh]

The adjustment was slight... certainly closer to one another, but still gives a weight-advantage to away games.

[Hugenerd]

The adjustment was slight compared to last year, but still pretty significant.  Away games are weighted 1.25 and home games weighted 0.75.  Just as an example, according to that methodology, a win at home vs. a 23-2 team counts LESS, than a win on the road against a 14-11 team.  I would not consider that slight.  Thats like saying beating a team like Johnson and Wales or Wheelock on the road is about as tough as beating Amherst or Middlebury at home.

[Hugenerd]

Blindly using an OWP value is just as egregious as blindly using a WP.  Is it really enough to play teams with good records and lose (and I say 'play' instead of schedule because the LEC and NESCAC each have multiple ranked teams in their conference, which really boosts their values). For example, one may think that a 17-5 record with a 0.54 OWP is really great, until one realizes that the weighted OWP of the 17 games that team has won is only 0.44.  To illustrate this point, in the table below, I have calculated OWP values of each of the NE ranked teams (from last week) in games they have won, as well as their record vs. 0.500 or better teams (where the head-to-head results between teams are not counted. For example, if I were doing a calculation for Amherst, and the team I was considering was Brandeis, the record I would use is 11-9, instead of 12-9, because the head-to-head result would not count).

Team	OWP	OWP in wins	(OWP in wins)-OWP	Record vs. 0.500+
Amherst	0.603	0.577	-0.026	12-2
Middlebury	0.571	0.522	-0.049	12-2
RIC	0.601	0.501	-0.100	11-5
West Conn	0.568	0.517	-0.051	10-2
WPI	0.616	0.608	-0.008	12-3
East Conn	0.536	0.442	-0.094	7-5
MIT	0.486	0.482	-0.004	12-1
Keene St.	0.553	0.491	-0.062	9-4
Wesleyan	0.501	0.447	-0.054	9-3
Tufts	0.548	0.470	-0.078	7-5
Becker	0.484	0.454	-0.030	8-4
Albertus Magnus	0.433	0.434	+0.001	6-1

After you take out losses from the weighted OWP calculation, the only 2 numbers that jump out at you, on the high side, are WPI and Amherst.  On the low end, you have East Conn, Albertus Magnus, Becker, and Wesleyan. Everyone else, however, is essentially 0.500+-0.200, which, in my opinion, is just noise, and at that point, I think you have to take a strong look at record and other primary criteria.  If I were ranking them, which obviously I am not in any official sense, I would put Amherst as a clear number 1.  After that, things get a bit less clear, but I would put them in the following groups.  In the 3 slots behind Amherst, I would put 2. Middlebury, 3. MIT, and 4. West Conn.  MIT and West Conn could be flipped, but I had West Conn below MIT because of their 3 losses to average or sub-par teams. The next group of three I would have is 5. WPI 6. Keene St. and 7. RIC (Keene has the head-to-head win vs. RIC).  In the final group, I would put 8. Albertus Magnus, 9. Wesleyan, 10. East Conn, and 11. Becker.  I think Tufts will drop out and did not calculate figures for everyone else, so I dont know who would replace them.  Again, this is just how I would use the OWP values, and I think it makes sense.  Otherwise, as you can see from the table, it is possible to boost your overall OWP ~0.100 (which from what I hear is extremely significant) by losing 4-6 games, without really showing you can beat the types of teams you are losing to.

So the new rankings are out and they make a lot more sense.  Although I didnt get the exact rankings correct, I got the 4 groups I defined correctly in my post above (1. Amherst, 2-4. Middlebury, West Conn, MIT, 5-7. Keene, RIC, WPI, and 8-11. Wesleyan, East Conn, Albertus, and Becker).   

[pjunito]

Nerd, you were right...

How far down do you think keene and Western will drop because of their loses yesterday?  Albertus will improve their SOS slightly with a win on Saturday. They beat JWU last night, who was 12-12 on the year and host St. Joe's on Saturday who is 17-7. But, it looks like Albertus should take no chances; they need to win the GNAC tournament to ensure a spot in the national tournament.

[Hugenerd]

Nerd, you were right...

How far down do you think keene and Western will drop because of their loses yesterday?  Albertus will improve their SOS slightly with a win on Saturday. They beat JWU last night, who was 12-12 on the year and host St. Joe's on Saturday who is 17-7. But, it looks like Albertus should take no chances; they need to win the GNAC tournament to ensure a spot in the national tournament.

I don't think those losses affect them much at all, as losses don't seem to count for much this year.  Those games will only help their OWP value also. I think if MIT wins tonight, it is possible they pass West Conn next week, but I can't see them dropping more than a couple of spots. Maybe RIC jumps up ahead of them also. Albertus may be stuck in that 10 spot, with East Conn's big win, although they are the only team in the region undefeated against regionally ranked opponents. Their low OWP may just be too big a hurdle to overcome under the current criteria.

[pjunito]

Thanks for the insight Nerd..

Can someone answer this question for me?  I thought I read earlier that the national committee (if there is one) when selecting the B and C teams, don't take the regional rankings into considerations as much as we may think. If a team like Albertus (who is ranked in top 25 but 10th in region) receives the automatic bid, where would they be seeded in the tournament? Would they be a 4 seed because of their poor SOS and low OWP? or could they be at a 3 or 2 seed?

[Hugenerd]

Thanks for the insight Nerd..

Can someone answer this question for me?  I thought I read earlier that the national committee (if there is one) when selecting the B and C teams, don't take the regional rankings into considerations as much as we may think. If a team like Albertus (who is ranked in top 25 but 10th in region) receives the automatic bid, where would they be seeded in the tournament? Would they be a 4 seed because of their poor SOS and low OWP? or could they be at a 3 or 2 seed?

They dont really seed in D3, but it is highly unlikely they would host.  Geography would be a higher consideration, but they do try to break up top teams, if possible without causing much additional travel.

[pjunito]

OK, so there is no seeding like the D1 or D2 (regionally) tournaments. I know the schools are broken up into 4 team pools. In 2010, Albertus played William Patterson and MIT played DeSales. WP was the 1 and Albertus the 4. I would not expect Albertus to host; not sure if their gym is equipped even if they were in the top of the region. I am looking around the Northeast trying to figure out where Albertus may end up...

I guess I wanted to know how likely would it be that a team like Albertus would face a RIC or East Conn in the first round of the NCAA. 

And nerd, you are my go to guy!

[Hugenerd]

They may seed within each pod to determine home/away jerseys, but there is no region seedings like D1 (and I am sure they try to split up top teams from each region as much as possible).  And honestly, your guess is as good as mine in terms of Albertus' first round opponent.  They dont have to play someone in this region either.  They could easily be moved to the middle atlantic region also.  They are in a location where it wouldnt be too difficult to move them around to a few different places.