Top 25 talk

[Hugenerd]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

[Hugenerd]

Knightslappy, do you have the ability to compile conference OWP/OOWP?  I realize that, because some teams have started conference play and other have not, they won't be perfect (in a perfect world, we'd be able to just look at the non-con schedules)... but that might shed more light on the results.

I averaged each conference members OWP and OOWP and then calculated a SOS number for each conference. I know the averaging method isn't exact, but I'm not set up to calculate an exact conference OWP and OOWP. I hope this gets us close enough for discussion purposes. Sorted by Strength of Schedule using the handbook's defined 2/3 and 1/3 rule:

Thanks, +1, great work!  

Even if it isn't 100% exact, it is likely pretty close... and it should help to shed some light on how good the conferences really are.

Would you think that SOS*Winning percentage would be a good way to come up with a composite number?

Oh, and it doesn't matter either way (just interested to know), but are these OWP and OOWP for all opponents, just D-III or just in-region D-III?

RPI is usually calculated 0.25*winning percentage + 0.75*SOS

[KnightSlappy]

Knightslappy, do you have the ability to compile conference OWP/OOWP?  I realize that, because some teams have started conference play and other have not, they won't be perfect (in a perfect world, we'd be able to just look at the non-con schedules)... but that might shed more light on the results.

I averaged each conference members OWP and OOWP and then calculated a SOS number for each conference. I know the averaging method isn't exact, but I'm not set up to calculate an exact conference OWP and OOWP. I hope this gets us close enough for discussion purposes. Sorted by Strength of Schedule using the handbook's defined 2/3 and 1/3 rule:

Thanks, +1, great work!  

Even if it isn't 100% exact, it is likely pretty close... and it should help to shed some light on how good the conferences really are.

Would you think that SOS*Winning percentage would be a good way to come up with a composite number?

Oh, and it doesn't matter either way (just interested to know), but are these OWP and OOWP for all opponents, just D-III or just in-region D-III?

These numbers are in-region only, same data I use for regional rankings.

The RPI calculation is 0.25 x WP + 0.75 x SOS. I could do that for each conference as well.

[Pat Coleman]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

Thanks -- I needed the laugh.

[Hugenerd]

Knightslappy, do you have the ability to compile conference OWP/OOWP?  I realize that, because some teams have started conference play and other have not, they won't be perfect (in a perfect world, we'd be able to just look at the non-con schedules)... but that might shed more light on the results.

I averaged each conference members OWP and OOWP and then calculated a SOS number for each conference. I know the averaging method isn't exact, but I'm not set up to calculate an exact conference OWP and OOWP. I hope this gets us close enough for discussion purposes. Sorted by Strength of Schedule using the handbook's defined 2/3 and 1/3 rule:

CONF     OWP      OOWP     SOS
WIAC     0.5987   0.5432   0.5802
MACC     0.5875   0.5312   0.5688
UAA      0.5773   0.5248   0.5598
CCIW     0.5515   0.5317   0.5449
SUNYAC   0.5557   0.5151   0.5422
USAC     0.5717   0.4827   0.5420
MASCAC   0.5735   0.4687   0.5386
MIAC     0.5446   0.5222   0.5372
NWC      0.5438   0.5227   0.5367
E8       0.5376   0.5282   0.5344
NJAC     0.5319   0.5279   0.5306
OAC      0.5348   0.5178   0.5292
MIAA     0.5302   0.5117   0.5240
ODAC     0.5173   0.5266   0.5204
MWC      0.5223   0.5149   0.5198
CC       0.5184   0.5189   0.5185
NESCAC   0.5161   0.5207   0.5176
LAND     0.5208   0.5105   0.5173
LL       0.5152   0.5075   0.5126
MACF     0.5057   0.5249   0.5121
LEC      0.5080   0.5165   0.5109
CAC      0.5069   0.5172   0.5103
CUNYAC   0.5046   0.5091   0.5061
GNAC     0.5193   0.4791   0.5059
CCC      0.5108   0.4857   0.5024
NAC      0.4939   0.5087   0.4988
NATHC    0.4899   0.4993   0.4930
IIAC     0.4801   0.5040   0.4880
HCAC     0.4845   0.4909   0.4866
GSAC     0.4569   0.5402   0.4847
SCAC     0.4835   0.4821   0.4831
PrAC     0.4763   0.4953   0.4827
NCAC     0.4750   0.4929   0.4810
ASC      0.4654   0.4942   0.4750
NEWMAC   0.4428   0.5386   0.4747
NECC     0.4611   0.4966   0.4729
SKY      0.4614   0.4894   0.4707
AMCC     0.4531   0.5009   0.4690
CSAC     0.4551   0.4843   0.4648
SCIAC    0.4061   0.4994   0.4372
UMAC     0.4018   0.4638   0.4224
IND      0.3921   0.4285   0.4042
NEAC     0.3820   0.4467   0.4036
SLIAC    0.3635   0.4344   0.3871

This does shed a lot of light on the Massey ratings.  The only reason it seems there is a WIAC "bias" in the Massey ratings is because the WIAC not only has the highest winning percentage in d3, but also the highest SOS. Therefore, there really isnt any bias at all, the WIAC has just done the best against the best competition in d3.  Therefore, any team that plays a WIAC opponent has their OWP and OOWP numbers bumped up.  I am assuming you would see the same thing with any conference so long as they played they toughest schedule and won 80%+ of their games.

[sac]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

Thanks -- I needed the laugh.

You can laugh it off all you want but it doesn't change the fact that its an inaccurate rating because its full of mistakes.

You've posted D3 OWP and OOWP on your site for several years now and I would think you would understand how one error can string through a lot of data to create multiple inaccuracies.   Massey's counting an exhibition in its rating would be similar to you calculating someone's OWP or OOWP using a game thats out of region or using an NAIA opponent.  Regardless of the error, its an error and corrupts the data set your looking at and trying to compare with other data that is accurate.

[Pat Coleman]

I do know that one extra result that fits within the data sample doesn't make most of the D-III ratings completely useless.

[sac]

I do know that one extra result that fits within the data sample doesn't make most of the D-III ratings completely useless.

Um yes it would, how is that hard to understand........and its not just one error, its several.  I'm sure Ferris State/Calvin isn't the only exhibition being counted and based on past history many NAIA results go completely unreported.

[Ralph Turner]

I do know that one extra result that fits within the data sample doesn't make most of the D-III ratings completely useless.

Um yes it would, how is that hard to understand........and its not just one error, its several.  I'm sure Ferris State/Calvin isn't the only exhibition being counted and based on past history many NAIA results go completely unreported.

I would like to see the calculations done with both results to give me an idea of the scope of the error and its imapct on Calvin's numbers.

The fact that 25 games will dilute the impact is also to be considered.

[Hugenerd]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

Thanks -- I needed the laugh.

You can laugh it off all you want but it doesn't change the fact that its an inaccurate rating because its full of mistakes.

You've posted D3 OWP and OOWP on your site for several years now and I would think you would understand how one error can string through a lot of data to create multiple inaccuracies.   Massey's counting an exhibition in its rating would be similar to you calculating someone's OWP or OOWP using a game thats out of region or using an NAIA opponent.  Regardless of the error, its an error and corrupts the data set your looking at and trying to compare with other data that is accurate.

I would argue that the ratings are perfectly accurate, the sources that provide Massey the ratings are where the inaccuracies are.  Massey explicitly states all his data sources and calculates his ratings based on the assumption that those are accurate.  I think your real beef should be with Presto or the NCAA or wherever the source of these reporting errors are.  If I buy an HP laptop in a factory sealed box from Walmart and when I open the box there are 3 keys missing from the keyboard, I could go back and yell at the Walmart employees until I am blue in the face but the real fault is HP's, not Walmart's for trusting HP to provide a product with no errors.

[sac]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

Thanks -- I needed the laugh.

You can laugh it off all you want but it doesn't change the fact that its an inaccurate rating because its full of mistakes.

You've posted D3 OWP and OOWP on your site for several years now and I would think you would understand how one error can string through a lot of data to create multiple inaccuracies.   Massey's counting an exhibition in its rating would be similar to you calculating someone's OWP or OOWP using a game thats out of region or using an NAIA opponent.  Regardless of the error, its an error and corrupts the data set your looking at and trying to compare with other data that is accurate.

I would argue that the ratings are perfectly accurate, the sources that provide Massey the ratings are where the inaccuracies are.  Massey explicitly states all his data sources and calculates his ratings based on the assumption that those are accurate.  I think your real beef should be with Presto or the NCAA or wherever the source of these reporting errors are.  If I buy an HP laptop in a factory sealed box from Walmart and when I open the box there are 3 keys missing from the keyboard, I could go back and yell at the Walmart employees until I am blue in the face but the real fault is HP's, not Walmart's for trusting HP to provide a product with no errors.

Oh my god.......whoever.

[Hugenerd]

F&M's site has it incorrect also, as Pat mentioned it would, but the boxscore linked on the site has the score 58-44.  If this error were to be picked up by Massey, it could compromise the integrity of his entire rating system.  I just hope, for the sake of all of us, that this error is picked up before it can do some real damage to someone's Massey rating.

Thanks -- I needed the laugh.

You can laugh it off all you want but it doesn't change the fact that its an inaccurate rating because its full of mistakes.

You've posted D3 OWP and OOWP on your site for several years now and I would think you would understand how one error can string through a lot of data to create multiple inaccuracies.   Massey's counting an exhibition in its rating would be similar to you calculating someone's OWP or OOWP using a game thats out of region or using an NAIA opponent.  Regardless of the error, its an error and corrupts the data set your looking at and trying to compare with other data that is accurate.

I would argue that the ratings are perfectly accurate, the sources that provide Massey the ratings are where the inaccuracies are.  Massey explicitly states all his data sources and calculates his ratings based on the assumption that those are accurate.  I think your real beef should be with Presto or the NCAA or wherever the source of these reporting errors are.  If I buy an HP laptop in a factory sealed box from Walmart and when I open the box there are 3 keys missing from the keyboard, I could go back and yell at the Walmart employees until I am blue in the face but the real fault is HP's, not Walmart's for trusting HP to provide a product with no errors.

Oh my god.......whoever.

My point is that if these errors are just random due to random reporting errors there is nothing we can do about it.  It is impractical to think that Massey or any other source can independently verify each game score.  You do the calculations based on the data you have, it doesnt invalidate the whole set of rankings.

[Hugenerd]

Middlebury down by 18 with 9 minutes to go against Colby Sawyer.

68-58 Colby Sawyer on top, 5 minutes to play.

72-63 CSC, under 4.

77-68 CSC, 1:45 to play.

80-70 CSC, 1:06 left.

84-75 CSC, :37 left.

FINAL: 86-76 CSC

Middlebury picks up their first loss.  Big win for Colby Sawyer.

[Hugenerd]

Another top 25 team in trouble:

Concordia Texas leads Mississippi College 49-38 at the half.

[Ralph Turner]

Another top 25 team in trouble:

Concordia Texas leads Mississippi College 49-38 at the half.

CTX has probably the best set of athletes in the ASC to run with Mississippi College.

CTX head coach Stan Bonewitz loves the run-and-gun.