2014 D3 Season: National Perspective

PaulNewman · October 29, 2014, 08:33:19 PM

Mules level with a PK. Under 20. I think I would watch that one Mr.Right.

Mr.Right · October 29, 2014, 08:49:08 PM

Actually I went with my heart and got Bates v Colby on, 8th / 7th place on the line. Colby scored early and lead 1-0. Bates better pick up the pace in this one in the 2nd half as they need two goals to keep their season alive. Colby has hit the crossbar and for some reason benched their keeper midway thru the 1st half. I did not see an injury but who knows

Mr.Right · October 30, 2014, 12:05:44 AM

After looking at the Great Lakes regional rankings I wanted to see why everyone was going crazy over Geneva being 7th. After looking, that might be one of the most egregious mistake in the rankings I have ever see. A 8-4-2 record with a .522 SOS with ONE good win against Carnegie Mellon and a bunch of bad losses including a 6-0 thrashing by John Carroll. It seems the region is down but I would think some of those NCAC teams like DePauw or even Hiram who have a better SOS. This region will not get more than 2 Pool C's. Loser of the Kenyon / OWU NCAC final, I assume John Carroll will win their league but if they do not they would get the other. Maybe one of the two UAA teams also gets in but it is a stretch.

varsityjog · October 30, 2014, 12:48:45 AM

Hey all new to the scene but I was hoping some one could shed some light on why Lycoming College dropped out of the Mid-Atlantic region completely... With a win in their last game against an underrated Hood College, who was tied for second in conference play at the time, taking them to a 11-2-3 record and a few teams ahead of them in the region losing it just doesn't seem make sense as too their disappearance.

Flying Weasel · October 30, 2014, 01:19:25 AM

Lycoming may have been bitten by the .500 SOS threshold that seems to be being used. Their SOS was right at .500 the previous week, but dropped to .497 this week. Of course, they were always a borderline team given their low SOS, so maybe the committee honestly felt upon looking it over again this week that there were eight better teams from the region.

Domino1195 · October 30, 2014, 07:37:37 AM

Denison at DePauw for the last tourney spot - wouldn't you love to be a player on either of those teams this week! First half of OWU-Denison was spent primarily between the penalty areas. When OWU was able to get inside the area they scored - I think they may have only had 3-4 shots on goal in the first half. Left when the score was 3-1; Denison's lone goal was kind of fluky as a long ball was cleared out for a throw, ball tossed I the area, OWU doesn't clear and Denison curls one in off the back post.

OWU possessed very well - Denison couldn't/wouldn't keep the ball on the ground. Hoping for OWU Kenyon rematch in the tourney final. OWU still has to answer questions re their back line. They seem to be good for one or two lapses each game, and as they saw in the Rose Hulman game last year - it only takes one . . .

varsityjog · October 30, 2014, 10:21:38 AM

Quote from: Flying Weasel on October 30, 2014, 01:19:25 AM
Lycoming may have been bitten by the .500 SOS threshold that seems to be being used. Their SOS was right at .500 the previous week, but dropped to .497 this week. Of course, they were always a borderline team given their low SOS, so maybe the committee honestly felt upon looking it over again this week that there were eight better teams from the region.

Thanks for the quick response,

Will the SOS rise after teams they've beaten like Rochester moved up to 4th in their region?

Seems odd that teams like Catholic, Hopkins, Haverford, and Swathmore all have 5+ losses and are ahead of Lycoming, regardless of SOS thats 3 more losses. Although I agree SOS is a big factor and teams like Lycoming do lack in that department, it feels like the committee puts a bit too much emphasis on it.

PaulNewman · October 30, 2014, 10:39:32 AM

How does Geneva get ahead of Wabash???

jknezek · October 30, 2014, 10:54:11 AM

Since schedules are not equal, SOS is always going to be very important. If you play and lose to top competition, that has to be weighted as well as playing and beating weak competition. SOS isn't perfect, but it is necessary. While teams can't always control the schedules, and even sometimes when they do set up a competitive match a few years ahead it may not pan out, if you are a "second chance" school, meaning non-AQ, you are going to need some meat on your schedule. Without it, there is a lot of skepticism about whether you deserve that Pool C second chance, or how highly you deserve to be ranked as a Pool A. Beating the weakest teams doesn't mean you are good, playing and losing competitively to good teams is a better indicator of where you belong when the playoff field is composed of those good teams.

lastguyoffthebench · October 30, 2014, 11:20:53 AM

Quote from: NCAC New England on October 30, 2014, 10:39:32 AM
How does Geneva get ahead of Wabash???

Wabash seems to be victim of the SOS with .489

PaulNewman · October 30, 2014, 11:40:51 AM

I guess.

Wabash has beaten Rose-Hulman, Hiram, Oberlin, Allegheny, and then #2 Kenyon. And tied DePauw. Lost to OWU (not a bad loss). Only bad loss is Denison. And only 2 losses overall and plays in a tougher conference, going 1-1-1 with the big three.

Geneva did beat an unranked CMU and next best win was Allegheny. Other good results were ties with Lycoming and Thomas More. Losses to LaRoche, John Carroll (not bad loss), Grove City, Fredonia St. Had 4 losses at time of ranking. And, btw, lost last night to Waynesburg.

Flying Weasel · October 30, 2014, 01:24:54 PM

I'm pretty convinced that a .500 SOS threshold is being applied as there's really no other explanation for Luther going from second in the North Region to unranked in one-week. And it would then explain the absence of Wabash and possibly others.

In the Great Lakes region there's a bunch of sub .500 SOS teams in addition to Wabash with winning pct.'s .750 or better that might otherwise be in the mix for the last spot or two (Penn State-Behrend, Thomas More, Medaille, Grove City) so the committee didn't have to go reaching for Geneva and Capital. Ohio Northern came from unranked to No. 6 because with the addition of the "vs. ranked" criteria in the second week, they all of a sudden have 3 wins vs. ranked that they didn't have the previous week. Nevermind those wins were against Case, DePauw, and Capital--not the most impressive list of ranked teams.

I also have observed over the years that the committee does not seem to get hung up on losses vs. ranked teams. It's the wins vs. ranked more so than the full record vs. ranked that they seem to key in on. At least that's my impression. And along with that, I'm not sure they really look at how "bad" your losses are and weigh some losses more heavily than others. Overall, I'm OK with both these things. You play a tough schedule and come out .500 against ranked teams, that says you belong in that group. 2-2-1 vs. ranked? I'm not going to penalize the 2 losses if being compared to a 1-0-1 record vs. ranked. And upsets happen and you can lose 1-0 to a team you outshoot 30-1. These things happen in soccer over the course of a season. Unless the committee is going to be looking at dozens and dozens of individual box scores, they can't reliably determine which losses were really "bad" losses.

The committee seems to really love SOS and wins vs. ranked and weight those two criteria very heavily. Unfortunately, both those measures are very flawed. The Ohio Northern example of wins vs. ranked teams illustrates that point very well. Depending on the strength/depth of a region, there's can be a big difference between a couple win against lowly ranked teams and the higher ranked teams, and yet you get the feeling the committee doesn't bother to look that deep on a case-by-case basis. And the SOS formula is very flawed. A quick current example is that Coast Guard's SOS is the same as Williams (.585) despite Williams adding Babson, RPI, and Skidmore to their NESCAC slate. The SOS formula is very simplistic and can't reliably capture who truly is a tough opponent. I don't think any formula ever can, but you could certainly do better then the one they are using.

The funny thing is that I fully agree with and support taking strength-of-schedule into account and I totally agree and support giving wieght to quality wins and being competitive (even if tying/losing) against top competition. But the specifics of how the committee tries to take those things into account is very unreliable and seriouly flawed in my opinion. A lot more discerment and judgment is required apart from number-crunching.

PaulNewman · October 30, 2014, 01:43:47 PM

Good job, FW.

So was Luther above .500 the week before?

I don't know the details of the how they determine the SOS, but, bottom line, is Luther going to be left out of the tournament? I just looked at their schedule and doesn't look that weak to me. They have their conference games and then they played against two of the UWs and also St Olaf, Carleton and Macalester. They beat all three of the latter, which includes highly rated St Olaf. They have a sterling overall record of 15-2-1 (only losses to Loras and Wartburg). Just by the "eye test" you can tell they belong in the tournament.

Flying Weasel · October 30, 2014, 01:56:44 PM

Yes, Luther's SOS the previous week was .519 and then fell to .493.

jknezek · October 30, 2014, 02:11:02 PM

One thing we all have to remember is that championship criteria is universal across D3. So the guidelines of using SOS and RROs to rank teams are used in all team sports. As is the A/B/C system. It's a fairly good system, especially when you consider that the Regional Committees meet only a couple times. The people on those committees all have other full time jobs. So this criteria, and this system, have to be both universally applicable and streamlined for time.

Would it be nice if the committee members spent as much time watching games as some of us do? Of course, but primarily they are coaching their own teams or managing their own departments at the same time. It's just not feasible. So simple metrics help distill data across a large database. There is still flexibility within the guidelines, so SOS and RROs aren't straitjackets, but they are very good starting points for both the teams and time available.

Do we argue about the last couple teams in and the first couple left out? Of course. No system is going to prevent that since everyone has an opinion. Do we quibble about the seedings? Of course, but those are also driven by money. In the end, the game is played out on the field, with the top teams included, and only the fringes being questionable. Given all the variables and constraints, we are fortunate to have things this good.