(03-08-2019 08:55 AM)gusrob Wrote:  Memphis and UConn have the same conference record. 5 and 6 seed battle.

I Googled "UConn Basketball Schedule" and it pulled up women's basketball schedule

.....UConn still has ECU to play. That's a win. If we lose, I'm thinking we're the 6 seed.

If UConn and MEMPHIS end up tied in Conference Record, do we get the 5 seed?

Both they Daily Memphian and CA reported. There is absolutely no way Memphis doesn’t get the 5. It’s done.

(03-08-2019 09:01 AM)UMTiger02 Wrote:  

(03-08-2019 08:55 AM)gusrob Wrote:  Memphis and UConn have the same conference record. 5 and 6 seed battle.

I Googled "UConn Basketball Schedule" and it pulled up women's basketball schedule

.....UConn still has ECU to play. That's a win. If we lose, I'm thinking we're the 6 seed.

If UConn and MEMPHIS end up tied in Conference Record, do we get the 5 seed?

Both they Daily Memphian and CA reported. There is absolutely no way Memphis doesn’t get the 5. It’s done.

You are kind

(03-08-2019 08:55 AM)gusrob Wrote:  Memphis and UConn have the same conference record. 5 and 6 seed battle.

I Googled "UConn Basketball Schedule" and it pulled up women's basketball schedule

.....UConn still has ECU to play. That's a win. If we lose, I'm thinking we're the 6 seed.

If UConn and MEMPHIS end up tied in Conference Record, do we get the 5 seed?

Google is not your friend.

UConn is 5-12 in conference play. We are 10-7. We could tie Wichita State if we lose and they win but we beat them twice this year. I don't think we lose to Tulsa at home and I don't think it's even close.

Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

(03-08-2019 09:09 AM)mairving Wrote:  

(03-08-2019 08:55 AM)gusrob Wrote:  Memphis and UConn have the same conference record. 5 and 6 seed battle.

I Googled "UConn Basketball Schedule" and it pulled up women's basketball schedule

.....UConn still has ECU to play. That's a win. If we lose, I'm thinking we're the 6 seed.

If UConn and MEMPHIS end up tied in Conference Record, do we get the 5 seed?

Google is not your friend.

UConn is 5-12 in conference play. We are 10-7. We could tie Wichita State if we lose and they win but we beat them twice this year. I don't think we lose to Tulsa at home and I don't think it's even close.

You are not kind

(03-11-2019 03:30 PM)cmt Wrote:  Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

(03-12-2019 12:46 PM)Tiger87 Wrote:  

(03-11-2019 03:30 PM)cmt Wrote:  Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Quote:The pre-season ratings will be degraded as real data accumulates. Starting next Monday, the influence of the initial rating will be dropped gradually each day until it reaches zero on the morning of January 23. This seems like a long ways away, but this date was chosen for a couple of reasons.

(03-12-2019 03:34 PM)macgar32 Wrote:  

(03-12-2019 12:46 PM)Tiger87 Wrote:  

(03-11-2019 03:30 PM)cmt Wrote:  Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Quote:The pre-season ratings will be degraded as real data accumulates. Starting next Monday, the influence of the initial rating will be dropped gradually each day until it reaches zero on the morning of January 23. This seems like a long ways away, but this date was chosen for a couple of reasons.

Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

(03-12-2019 05:18 PM)Tiger87 Wrote:  

(03-12-2019 03:34 PM)macgar32 Wrote:  

(03-12-2019 12:46 PM)Tiger87 Wrote:  

(03-11-2019 03:30 PM)cmt Wrote:  Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Quote:The pre-season ratings will be degraded as real data accumulates. Starting next Monday, the influence of the initial rating will be dropped gradually each day until it reaches zero on the morning of January 23. This seems like a long ways away, but this date was chosen for a couple of reasons.

Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

It is technically not a false statement. But the preseason rankings are what form the first day rankings which form the second day rankings which form the third day rankings, etc, etc, so they are no longer in the formula by the end of January, but they, in part, created all the numbers that are still in the formula, so they are still there.

(03-08-2019 09:01 AM)UMTiger02 Wrote:  

(03-08-2019 08:55 AM)gusrob Wrote:  Memphis and UConn have the same conference record. 5 and 6 seed battle.

I Googled "UConn Basketball Schedule" and it pulled up women's basketball schedule

.....UConn still has ECU to play. That's a win. If we lose, I'm thinking we're the 6 seed.

If UConn and MEMPHIS end up tied in Conference Record, do we get the 5 seed?

Both they Daily Memphian and CA reported. There is absolutely no way Memphis doesn’t get the 5. It’s done.

yeah.....my sharing of google's goofiness hasn't been taken as intended.

Thanks for the serious reply. I appreciate the attempted help.

(03-13-2019 07:58 AM)cmt Wrote:  

(03-12-2019 05:18 PM)Tiger87 Wrote:  

(03-12-2019 03:34 PM)macgar32 Wrote:  

(03-12-2019 12:46 PM)Tiger87 Wrote:  

(03-11-2019 03:30 PM)cmt Wrote:  Looking back, since December, the biggest change in AAC Kenpom standings are USF, up 116 from December and Tulane, down 103. Memphis came in 3rd with a plus 51 and a fifth place finish. What a couple more wins could have done. Obviously NET likes the AAC better than Kenpom does, but I think I agree with Kenpom more, I guess we'll see what happens in the next 2 weeks.

How can Dunleavey go winless in the AAC? Embarrassing. Of course they have a shot at redemption Thursday? No.

I am going to enjoy the next 2 weeks, the more we win, the more I'll enjoy, but everyday gets us a little closer to next season and we'll be back where we deserve to be, not duking it out for 5th place.

This was where everybody stood on 12/3/2018, ranking is Kenpom

1. Cincinnati 7-1, 36 loss to OSU, best win @UNLV and Mississippi @home
2. Houston 6-0, 41 beat BYU and Oregon, lost to nobody
3. UCF 6-2, 43 lost to FAU and Missouri, beat Alabama, ST Joe's
4. Temple 7-1, 75 beat Missouri, Georgia and St Joe's, lost to VCU
5. Wichita State, 82 4-3 lost to LaTech, Davidson and Alabama, beat Baylor
6. SMU 6-3, 87 lost to Lipscomb, beat Lamar
7. UCONN 6-2, 88 beat Syracuse, lost to Iowa and Arizona
8. Memphis 3-4, 123 beat (86) Yale, lost to Charleston (116)
9. Tulsa 5-3, 128 beat California Baptist, lost to SIU, Nevada and Utah
10. Tulane, 190 2-5 lost to UT Martin
11. USF 6-2, 228 lost to Georgetown and the Citadel, beat FAMU
12. ECU 6-4, 256 includes loss to High Point, best win UT Rio Grande Valley?

Here's how it all turned out, NET, then Kenpom, then change in Kenpom

1. Houston 16-2 (29-2) 7, 13 <+28>
2. Cincinnati 14-4 (25-6) 23, 34 <+2>
3. UCF 13-5 (23-7) 28, 37 <+6>
4. Temple 13-5 (23-8) 49, 69 <+6>
5. Memphis 11-7 (19-12) 51, 72 <+51>
6. Wichita State 10-8 (17-13) 91, 100 <-18>
7. USF 8-10 (19-12) 86, 112 <+116>
8. Tulsa 8-10 (18-13) 90, 120 <+8>
9. UCONN 6-12 (15-16) 98, 101 <-13>
10. SMU 6-12 (14-16) 110, 117 <-30>
11. East Carolina 3-15 (10-20) 254, 269 <-13>
12. Tulane 0-18 (4-26) 297, 293 <-103>

KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Quote:The pre-season ratings will be degraded as real data accumulates. Starting next Monday, the influence of the initial rating will be dropped gradually each day until it reaches zero on the morning of January 23. This seems like a long ways away, but this date was chosen for a couple of reasons.

Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

It is technically not a false statement. But the preseason rankings are what form the first day rankings which form the second day rankings which form the third day rankings, etc, etc, so they are no longer in the formula by the end of January, but they, in part, created all the numbers that are still in the formula, so they are still there.

Not true

Eventually all of the rating is 100% based on efficiency

First game SOS is determined by the 30 games that came after it.

(03-13-2019 09:47 PM)macgar32 Wrote:  

(03-13-2019 07:58 AM)cmt Wrote:  

(03-12-2019 05:18 PM)Tiger87 Wrote:  

(03-12-2019 03:34 PM)macgar32 Wrote:  

(03-12-2019 12:46 PM)Tiger87 Wrote:  KenPom had a poor read on the AAC this year from the jump. Hard to put faith in his numbers for us after the preseason bias he had built in.

Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Quote:The pre-season ratings will be degraded as real data accumulates. Starting next Monday, the influence of the initial rating will be dropped gradually each day until it reaches zero on the morning of January 23. This seems like a long ways away, but this date was chosen for a couple of reasons.

Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

It is technically not a false statement. But the preseason rankings are what form the first day rankings which form the second day rankings which form the third day rankings, etc, etc, so they are no longer in the formula by the end of January, but they, in part, created all the numbers that are still in the formula, so they are still there.

Not true

Eventually all of the rating is 100% based on efficiency

First game SOS is determined by the 30 games that came after it.

Which was based on the preseason rankings. We'll just have to disagree.

(03-14-2019 06:41 AM)cmt Wrote:  

(03-13-2019 09:47 PM)macgar32 Wrote:  

(03-13-2019 07:58 AM)cmt Wrote:  

(03-12-2019 05:18 PM)Tiger87 Wrote:  

(03-12-2019 03:34 PM)macgar32 Wrote:  Any preseason bias he had built in is 100% removed...

He metric is 100% data driven at this point. I think he says the preseason ratings expire in late January.

Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

It is technically not a false statement. But the preseason rankings are what form the first day rankings which form the second day rankings which form the third day rankings, etc, etc, so they are no longer in the formula by the end of January, but they, in part, created all the numbers that are still in the formula, so they are still there.

Not true

Eventually all of the rating is 100% based on efficiency

First game SOS is determined by the 30 games that came after it.

Which was based on the preseason rankings. We'll just have to disagree.

It is math... There is nothing to disagree about.

(03-16-2019 07:24 AM)macgar32 Wrote:  

(03-14-2019 06:41 AM)cmt Wrote:  

(03-13-2019 09:47 PM)macgar32 Wrote:  

(03-13-2019 07:58 AM)cmt Wrote:  

(03-12-2019 05:18 PM)Tiger87 Wrote:  Yeah, they all say that. But unless he has full transparency on his algorithm, how do you know what he places the most and least value on? He also shows us a "luck" metric. Luck is not included in his rankings, but it is based upon results varying from his predicted outcome. So this alone tells you that he discounts actual results to a certain extent. So for teams with high luck, he is undervaluing their results, and teams with low luck he is overvaluing.

The fact that he had SMU preseason #1 in the AAC and it took him SO LONG to move away from that tells me all I need to know.

I think he is great at measuring individual analytics, and decent at predicting games. But his is certainly NOT the best ranking tool.

It is technically not a false statement. But the preseason rankings are what form the first day rankings which form the second day rankings which form the third day rankings, etc, etc, so they are no longer in the formula by the end of January, but they, in part, created all the numbers that are still in the formula, so they are still there.

Not true

Eventually all of the rating is 100% based on efficiency

First game SOS is determined by the 30 games that came after it.

Which was based on the preseason rankings. We'll just have to disagree.

It is math... There is nothing to disagree about.

Then stop it.