2015 D3 Season: NATIONAL PERSPECTIVE

Started by D3soccerwatcher, February 08, 2015, 12:49:03 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

PaulNewman

Quote from: Mid-Atlantic Fan on October 21, 2015, 03:07:34 PM
Quote from: NCAC New England on October 21, 2015, 02:48:32 PM
Quote from: Flying Weasel on October 21, 2015, 02:42:54 PM
Whitworth: .468 SOS

Then the system is just inherently flawed.  Don't get how Midd can be .580+ with all of their cupcakes and Whitworth .468 and Kenyon .510 or whatever.  Two of the top five teams in the country with one barely ranked and other not ranked.  This isn't Randolph ranked #15 in the country not being ranked....it's two top 5 teams.

You can't flaw factual numbers and other teams records......

Are you kidding me?  Of course you can.  You can flaw how the numbers are reached and put together.

Mr.Right

Sitting at #7 in that region Kenyon has work left to do...I agree they have gotten screwed but to say they will be fine is a bit of an understatement. In that region you must be in the top 4-5 to have a chance usually

PaulNewman

Quote from: Mr.Right on October 21, 2015, 03:13:07 PM
Sitting at #7 in that region Kenyon has work left to do...I agree they have gotten screwed but to say they will be fine is a bit of an understatement. In that region you must be in the top 4-5 to have a chance usually

Of course, but the last week is far more important than the 1st.  Luther went from #2 to unranked and maybe without even losing.  Montclair was #1 or #2 and then #7 or #8 in week 3.  In other words, by next week this week's ranking will mean nothing.  If Kenyon had been #1 or #2 this week they would still have the same work to do. 

PaulNewman

Back to our regular programming, TMC easily handles Grove City 3-0.  Juniet with another goal and 2 more assists, putting him at 13 G and 19 A I believe. 

Mr.Right

Yes but looking at their schedule closely I do not see many solid wins except for Denison. The Case game was cancelled and they loss to DePauw...I am thinking they MUST beat OWU...I hear you though as OWU's schedule is almost worse...It is definitely a head scratcher

PaulNewman

Quote from: Mr.Right on October 21, 2015, 03:29:28 PM
Yes but looking at their schedule closely I do not see many solid wins except for Denison. The Case game was cancelled and they loss to DePauw...I am thinking they MUST beat OWU...I hear you though as OWU's schedule is almost worse...It is definitely a head scratcher

You don't count CMU as a solid win?

Mr.Right

I do but the committee does not as they are not even ranked

Flying Weasel

I find the NCAA's SOS calculations to be a seriously flawed way of gauging/measuring the stregnth of a schedule.  OWP and OOWP are factual mathematical data, yes.  But the strength of a schedule is not a purely quanitfiable thing.  Numerous attempts have be made to try to quantify it with formulas and equations, but they are all flawed.  For example, Bennett Rank has St. Lawrence at #3 in the nation right now and Messiah at No. 10?  Really?  Flawed!  Massey may do better, but still spits out some results that don't square with our own good judgment and discerment.  And the NCAA Rankings, by putting so much weight on a SOS calculation that can so easily be shown to be a flawed reflection/measurement of schedule difficulty, is consequently flawed.

PaulNewman

Mr.Right, remember that John Carroll was ranked #2 in the THIRD ranking and DID NOT get in.  I think placement this week, despite the obvious uproar (including my own), will end up being barely a blip.

Mid-Atlantic Fan

Quote from: NCAC New England on October 21, 2015, 03:09:58 PM
Quote from: Mid-Atlantic Fan on October 21, 2015, 03:07:34 PM
Quote from: NCAC New England on October 21, 2015, 02:48:32 PM
Quote from: Flying Weasel on October 21, 2015, 02:42:54 PM
Whitworth: .468 SOS

Then the system is just inherently flawed.  Don't get how Midd can be .580+ with all of their cupcakes and Whitworth .468 and Kenyon .510 or whatever.  Two of the top five teams in the country with one barely ranked and other not ranked.  This isn't Randolph ranked #15 in the country not being ranked....it's two top 5 teams.

You can't flaw factual numbers and other teams records......

Are you kidding me?  Of course you can.  You can flaw how the numbers are reached and put together.

Nope not kidding you actually. You know the criteria they use. You know the teams records and their opponents records. You know the formula. LastGuy has been doing calculations which have been spot on for the past two weeks. I don't know why this is a shock for you....

You keep eluding to wins vs ranked teams and vs solid teams which won't be used entirely until the next release of the rankings. In Kenyon's case you have other posters like Domino have showed you that Kenyon is 28/52 in region for SOS and the lowest of the 8 that are ranked. They have good wins vs CMU and Denison and a loss vs DePauw. Other than that they haven't played anybody. OWU game will be huge for them.

I think Kenyon is a good team and had them in my elite 8 predictions, but knowing that this first ranking is all SOS and win % I am not surprised where they are ranked. Just like Eastern in a region that I am more familiar with. They are #8/13 in the national polls, 2nd in NSCAA regional poll and yet they are #5 in the NCAA with an undefeated 12-0-2 record. I am not surprised by this because they play nobody, they are in a bad conference, and their SOS is right at the magical .500 mark. In Eastern's case the record vs ranked will help as they are 2-0. So I will be interested to see where they end up with that calculation added in.

My main point is that yes you may disagree but I am not sure why you are so shocked and outraged by something you probably should have expected if you follow the game at all.

PaulNewman

MAF, understand what you are saying, but I can still say the method of calculation is flawed which is different than suggesting that an error was made in their calculation.  As an aside, why do they even need a cmte if it's going to be solely numbers driven in their idiosyncratic way.  At any rate, I didn't see any warnings from anyone here to expect Kenyon to be #7 or unranked.  I understand this is irrelevant but just by the eye test there isn't any substantial difference in the schedules OWU vs Kenyon.  And again, putting Kenyon aside, cannot see how TMC and CWRU can be below some of those other teams.. What about CMU.  They played Messiah and Kenyon and play in the UAA.  They had to think when they made that schedule they were doing the right thing.  If Kenyon has ends up having 1-2 more blemishes and doesn't get in then I will express some outrage.  And it's outrageous to suggest that their season is riding on next week's game with OWU.  Even with the system as it is I do not believe it will play out that way.

TennesseeJed

Quote from: NCAC New England on October 21, 2015, 03:31:06 PM
Quote from: Mr.Right on October 21, 2015, 03:29:28 PM
Yes but looking at their schedule closely I do not see many solid wins except for Denison. The Case game was cancelled and they loss to DePauw...I am thinking they MUST beat OWU...I hear you though as OWU's schedule is almost worse...It is definitely a head scratcher

You don't count CMU as a solid win?

With respect to Kenyon in particular, they played most of their games at home, including all their more difficult games.  Because of the multipliers for home and away (so well articulated by FW in his article and other posts), it's clear to me that Kenyon (and others with a balance of their games at home) pay a heavy penalty for beating good teams at home, or even worse for losing to good teams at home.  Looked at in reverse, teams get very handsomely rewarded for winning away games and they get a bit of a break for losing away games...

In addition, no RvR impacts are yet factored into the calcs, so I can't recall who pointed this out, but the rankings that REALLY matter are next Wednesday's NCAA regional rankings, when all the "yeah, but we beat that team, and they lost to that team" stats are factored in...  I'm certainly not the local expert here, but I would expect for the rankings to be shaken up a bit when this week's games are factored in as well as the RvR's for all the teams.

Finally, the NCAA doesn't apparently provide the weights they use for SoS and W/L% in determining ranks, but as I posted on the Great Lakes thread, they can't be evenly weighted, or the multiplication of the two factors, and the corresponding sorted rank order would get us to the actual rankings put out by NCAA today.  I can only come up with 2-3 possible explanations, (though there may be many more...).  1) you have to know the weights of each factor to get to NCAA's rankings, or 2) there are subjective, qualitative factors at play that are not disclosed that are having material impacts to the rankings, and 3) there are other, potentially quantitative factors that are also undisclosed that are at play.

Mid-Atlantic Fan

#1287
Quote from: NCAC New England on October 21, 2015, 04:01:52 PM
MAF, understand what you are saying, but I can still say the method of calculation is flawed which is different than suggesting that an error was made in their calculation.  As an aside, why do they even need a cmte if it's going to be solely numbers driven in their idiosyncratic way.  At any rate, I didn't see any warnings from anyone here to expect Kenyon to be #7 or unranked.  I understand this is irrelevant but just by the eye test there isn't any substantial difference in the schedules OWU vs Kenyon.  And again, putting Kenyon aside, cannot see how TMC and CWRU can be below some of those other teams.. What about CMU.  They played Messiah and Kenyon and play in the UAA.  They had to think when they made that schedule they were doing the right thing.  If Kenyon has ends up having 1-2 more blemishes and doesn't get in then I will express some outrage.  And it's outrageous to suggest that their season is riding on next week's game with OWU.  Even with the system as it is I do not believe it will play out that way.

I cannot say I disagree with anything you have stated. I agree with your assessment. I just am not as surprised I guess because it's the NCAA we are talking about.  ;D

I think teams 5,6,7 are better than teams 2,3,4 in that region.

Flying Weasel

I don't know, but I highly doubt there is any mathematical weighting of the various primary criteria which are the following:
   • Win-loss percentage against Division III opponents
   • Division III head-to-head competition
   • Results versus common Division III opponents
   • Results versus ranked Division III teams at the time of selection
   • Division III strength of schedule

Once the win pct., SOS, record vs. ranked, etc. are compliled/computed, I highly doubt they are feeding those numbers into yet another equation/formula.  At that point it's a discussion. 

It seems that SOS and wins (more so than record) versus ranked teams are the best predictors of ranking and eventually selection.

Mr.Right

THE EAST IS  A MESS.......No Skidmore.....SUNYAC gets 3 of the top 4 seeds...Stevens as a #2....Liberty League gets shafted again....

My EAST Rankings:

1. Plattsburgh
2. Skidmore
3. Oneonta St
4. Hobart
5. Stevens
6. SLU
7. RPI


WHY in the hell is Cortland St even ranked?????? Can 1 sane person answer this, they are 9-4-2 with NOT 1 Good win


Looks like Skidmore needs to go clean and MUST beat RPI and SLU this week