I didn't say that KP doesn't play a factor in selection. It's just a hilarious system. On this recent stretch where temple has knocked off many teams ranked highly in the KP system they've hardly moved up his system. This is because of his "luck" factor. Also known as the tough team that knows how to win close games factor.

(02-15-2016 11:53 AM)TempleU22 Wrote:  I didn't say that KP doesn't play a factor in selection. It's just a hilarious system. On this recent stretch where temple has knocked off many teams ranked highly in the KP system they've hardly moved up his system. This is because of his "luck" factor. Also known as the tough team that knows how to win close games factor.

This 'luck' factor in KP is a joke. You want a team that knows how to win the close ones, not be dismissive of them. Look at Villanova the last two years. Beat a lot of teams by double digits, but when it comes time to be in the crucible, they melted. It is why I think the 2014 UCONN team made a strong run at the end. Knew how to be gritty when it counted, and was prepared for it by actual game experience.

(02-15-2016 11:07 AM)stever20 Wrote:  The committee though used the advanced metrics more last year than ever before. Really don't think that's a 1 year thing quite frankly. It's why UConn and Cincy are in a lot better shape than folks think- and why Temple really has a ton of work to do to get in the hunt for an at large.....

SMU missed in 2014 with a top 30 ranking in KenPom and Sagarin.

(02-15-2016 01:11 PM)MarloStanfield Wrote:  

(02-15-2016 11:07 AM)stever20 Wrote:  The committee though used the advanced metrics more last year than ever before. Really don't think that's a 1 year thing quite frankly. It's why UConn and Cincy are in a lot better shape than folks think- and why Temple really has a ton of work to do to get in the hunt for an at large.....

SMU missed in 2014 with a top 30 ranking in KenPom and Sagarin.

Non-Conference SOS of 303. The committee was strictly punishing that. I didn't agree, but that's the reason they got left out. It was too bad to ignore.

Ball don't lie.

(02-15-2016 12:36 PM)fishpro1098 Wrote:  

(02-15-2016 11:53 AM)TempleU22 Wrote:  I didn't say that KP doesn't play a factor in selection. It's just a hilarious system. On this recent stretch where temple has knocked off many teams ranked highly in the KP system they've hardly moved up his system. This is because of his "luck" factor. Also known as the tough team that knows how to win close games factor.

This 'luck' factor in KP is a joke. You want a team that knows how to win the close ones, not be dismissive of them. Look at Villanova the last two years. Beat a lot of teams by double digits, but when it comes time to be in the crucible, they melted. It is why I think the 2014 UCONN team made a strong run at the end. Knew how to be gritty when it counted, and was prepared for it by actual game experience.

"Luck factor" isn't calculated by how you do in close games. There is a lot more to it than this. It's also not a part of kenpom's equation, it's a figure he uses to illustrate which teams probably aren't as good as their record indicates.

Lets use the Houston game as an example. The Houston game likely is a killer result on Temple's kenpom rating. Houston is 81 according to Kenpom and Temple is 86. The 86th best team would be expected to beat the 81st best team on its home court (probably a 3-5 point favorite). I haven't done the statistical analysis but my guess is the type of team that would lose to the 81st team at home by 27 is expected to be somewhere below 200 in kenpom's ratings (and possibly pretty far below that). In that one game Temple's performance is essentially that of a sub 200 team. It only counts as a loss and to a fairly decent team, but it's a very bad performance that suggests Temple isn't very good.

Now lets take Temple's home win over UConn. The 86th best team would probably be a slight underdog (5-7 points maybe) to beat the 23rd ranked kenpom team. Temple's 5 point win is say 10-12 points better than their expected result in that game. (again i'm fudging the numbers slightly, but you get my point). It would probably be a team ranked about 30-35 that would be expected to win by 5 over the 23rd ranked team at home. So for that game... Temple has a top 30-35 performance. It's only one win but that win suggests Temple is pretty good. If Temple had that type of performance all year they'd be significantly higher in the rankings.

Early in the year Temple beat Minnesota at home by 5. Minnesota is 179. I suspect a team ranked 170-180ish would be expected to beat minnesota at home by 5. So even though that performance is a win, it is a win that kenpom views as "lucky" because that kind of performance wouldn't merit a 5 point win against a team better than 170 in the kenpom ranking.

The result isn't lucky in and of itself (according to kenpom's methodology). Basically kenpom is aggregating all of you performances and determining what your expected record would be performing at that level all year. The "luck" factor is about whether your actual results outperform that number. It works this way because kenpom adjusts for level of competition. Temple's close wins over UConn aren't considered "lucky" in kenpom because Temple actually performed at a higher level than their aggregate ranking. Luck comes in when a team performs below that aggregate ranking.

It works the other way too. Had Temple lost a close game at UConn or lost a close game at Cincinnati (say 2 points in each). Those performances would have been "unlucky" because Temple performed at a higher level than their aggregate ranking.

(02-15-2016 11:40 AM)TempleU22 Wrote:  I love how fans most cinci and UCONN fans won't simply acknowledge the flaws behind Kenpom and Sagarin. It's just as flawed of a system as RPI. I understand that you love seeing your team that high and will fight for the ratings until the bitter end. I would think that the committee is smart enough to see the results that temple has accumulated in actual game action, not some silly metric the values how much you win by.

Who said the other ratings systems don't have flaws? They all do. That's why we have a human committee to weigh ALL the data. Temple fans are just horses with blinders on only seeing what they want to. UC is on the bubble because they have good ratings in some metrics, and bad ones in others. That bolded statement is laughable. The RPI is by far the most flawed metric, because it oversimplifies records.

By the way, this RPI system Temple is clinging to so desperately, still has them behind SMU, UConn, Tulsa, and Cincinnati.

You fail to see that a team's whole season counts. Sorry you messed the bed in the non-conference. If you had capitalized and got a couple even decent wins, you wouldn't be in this position right now. When you only beat teams with 200+ RPIs/KPs/Sagarins etc in the non-conference, you need to be almost perfect in conference. And losing by 27 to UH, to ECU, and Memphis is far from perfect, even when paired with wins over SMU, UConn 2x, UC 2x, and Tulsa. Those wins just got you in the conversation. Now you need to go over the top.

You beat Nova on Wednesday, and a lot of your problems are fixed.

Nobody believes that the committee should seed based on a just following straight down kenpom or sagarin, but both kenpom and sagarin are ratings system where 100% of your rating is based on your on court performance. When you get into the nitty gritty of methodology maybe you think one thing or another should be different and teams can be playing better or worse than their rating (the rating is merely an assessment of how they have played over the course of the entire season). Nobody would argue the systems are perfect. But they tell you far more about a basketball team than RPI.

It's certainly possible that both these systems are missing on how good Temple is right now. That Temple has figured it out and some of the bad performances that both systems are factoring in are not a good reflection of Temple. The committee can have that argument, but as a rule they are really good, predictive rating systems.

RPI only 25% of your rating is based on on court performance. Once a season begins, regardless of how a team plays, it can only effect 25% of it's RPI rating. That's probably the biggest reason the RPI ratings are flawed. (though weighing all home wins equally, all home losses equally, all road wins equally and all road losses equally is another)

(02-15-2016 01:43 PM)bearcatmark Wrote:  Nobody believes that the committee should seed based on a just following straight down kenpom or sagarin, but both kenpom and sagarin are ratings system where 100% of your rating is based on your on court performance. When you get into the nitty gritty of methodology maybe you think one thing or another should be different and teams can be playing better or worse than their rating (the rating is merely an assessment of how they have played over the course of the entire season). Nobody would argue the systems are perfect. But they tell you far more about a basketball team than RPI.

It's certainly possible that both these systems are missing on how good Temple is right now. That Temple has figured it out and some of the bad performances that both systems are factoring in are not a good reflection of Temple. The committee can have that argument, but as a rule they are really good, predictive rating systems.

RPI only 25% of your rating is based on on court performance. Once a season begins, regardless of how a team plays, it can only effect 25% of it's RPI rating. That's probably the biggest reason the RPI ratings are flawed. (though weighing all home wins equally, all home losses equally, all road wins equally and all road losses equally is another)

Quiet you. No, I refuse to listen. I will play with my abacus with my head in the sand, and there is nothing you can do to stop me. /sarcasm

Interesting fact:

If you averaged Sagarin, Pomeroy, KPI, BPI, RPI and LRMC last year as of Selection Sunday, you got all the teams right.

Well, UCLA, Miami and Florida tied for the last at large bid, but Florida had a losing record and UCLA had a much better RPI than both Miami and Florida, so you would have gotten all the teams right if you used common sense on the last pick.

We'll see how the 6-rank average does this year.

Way too long of a post to say Temple is inconsistent. We know.

(02-15-2016 11:30 AM)bearcatmark Wrote:  

(02-15-2016 11:04 AM)HP-TBDPITL Wrote:  The committee doesn't use RPI as their guide for whether teams get in or not.

The committee uses RPI for seeding purposes after the teams are selected...and also have to factor in splitting up conference teams, look at sites, etc...

It really is as simple as that.

This just isn't true. Committee has always looked to RPI when it tries to figure out who you beat and who you lost to. Record vs. top 25 RPI, top 50RPI, and top 100 RPI are consistently talked about by committee in terms of selection. Your actual RPI number is less important though I think you'll find that teams get dropped a seed line or two if their RPI is significantly lower than the rest of their resume implies they should be seeded.

That said they aren't limited by that and can make quality judgments independent of these things (and often when they do it ends up being among their most baffling decisions). I only take cursory looks at any of the other measures for selection purposes because I almost never hear the committee talk about them.

You actually agreed with my point in bold. They look at your RPI number for seeding more than look at it in regards to selection.

(02-15-2016 01:57 PM)CougarRed Wrote:  Interesting fact:

If you averaged Sagarin, Pomeroy, KPI, BPI, RPI and LRMC last year as of Selection Sunday, you got all the teams right.

Well, UCLA, Miami and Florida tied for the last at large bid, but Florida had a losing record and UCLA had a much better RPI than both Miami and Florida, so you would have gotten all the teams right if you used common sense on the last pick.

We'll see how the 6-rank average does this year.

Here are the averages currently:

SMU 15
UConn 33
Cincy 38
Tulsa 52
Temple 71
Houston 81

Under that scenario, UConn and Cincy are in. Tulsa is among the first 4 out. Temple is in only because it leads the conference - like Houston, it must win the conference tourney.

I had to recheck myself, but how close your games are is pretty much exactly what contributes to the luck factor.

Teams that win a lot of close games and have a few decisive losses always end up with a high luck factor. Look, nobody here is arguing that temple is decisively ahead of the other AAC bubble teams but to dismiss the on court action and look at a silly mathematic formula is absurd.

The NCAA tournament has 32 automatic qualifiers, teams that win their conferences. The other 36 are at-large selections, so picking 1-68 RPI, or Kenpom, or an average, doesn't work.
Until very recently, the selection was based on merit - what did your team do during the season to make them tournament worthy? I am an admittedly biased, still-bitter-about-last-year, Temple fan, who saw or heard for the 1st time that UCLA got in on the eyeball test - for a good loss to AZ. The Owls were 0-5 against SMU and Tulsa, I think one win in that group would have got us in.
Bottom line, the P5 are going after the other conferences. The American may only get 1 team this year. UC and UConn are not locks, though they should be. The PAC may get 8, and they are terrible.

First let me say I’m not disputing any of this, I just don’t understand how they come up with these numbers and I’m hoping someone could explain it to me.

Let’s say it plays out where Temple loses a close game to Nova, beats Houston but loses to Tulsa, and wins the remaining games against Memphis, UCF and Tulane. They would finish the regular season 20-10 and most likely in first place alone.

Based on the above numbers Temple would be behind UConn and UC who they have beaten twice, behind Tulsa who they split with and even though they beat a top 10 ranked SMU and their only 2 “bad losses” being at Memphis and at ECU by 2 and 3 they would be on the outside looking in?

I understand Temple played poorly in its OOC but their 6 OOC losses would be to teams with RPI’s of 8, 18, 44, 29, 64 and 2, and all but one of those games were played prior to December 15th.

It was said that last year Temple didn’t get in because all their quality wins came early in the season and they didn’t beat the top of the conference teams. This year they beat everyone at the top at least once but may not get in because of what happened in November and early December?

Again I’m not saying anyone is wrong, it just makes no sense to me and I just don’t get it.

(02-15-2016 04:09 PM)TempleU22 Wrote:  I had to recheck myself, but how close your games are is pretty much exactly what contributes to the luck factor.

Teams that win a lot of close games and have a few decisive losses always end up with a high luck factor. Look, nobody here is arguing that temple is decisively ahead of the other AAC bubble teams but to dismiss the on court action and look at a silly mathematic formula is absurd.

Here is the key point on luck from kenpom's glossary:
Luck - A measure of the deviation between a team’s actual winning percentage and what one would expect from its game-by-game efficiencies.

If every team was exactly even that would mean winning a lot of close games would be "lucky" and losing a lot of close games would be "unlucky", but again kenpom adjusts for Strength of Opponent. So when Temple wins a close game at UConn that isn't really considered "lucky" as Temple played at an efficiency level where it would beat a ton of teams. When Temple wins a close home game over Minnesota that is considered "lucky" as Temple is playing at an efficiency level that is closer to a team in the low 100s than a team in the 80s.

(02-15-2016 04:09 PM)TempleU22 Wrote:  I had to recheck myself, but how close your games are is pretty much exactly what contributes to the luck factor.

Teams that win a lot of close games and have a few decisive losses always end up with a high luck factor. Look, nobody here is arguing that temple is decisively ahead of the other AAC bubble teams but to dismiss the on court action and look at a silly mathematic formula is absurd.

Agree, have watched Temple several times this year, not just against UC. Good, tough team that has risen to the occasion in several big games. Love that Decosey kid.

(02-15-2016 01:43 PM)bearcatmark Wrote:  (though weighing all home wins equally, all home losses equally, all road wins equally and all road losses equally is another)

There's a significant amount of research out there showing that home court advantage is consistent across all venues in all sports (with small exceptions for altitude).

RPI is a flawed metric, but I think it does a better job of explaining a team's resume than Pomeroy or Sagarin. And I'm a huge Pomeroy fanboy.

(02-15-2016 01:31 PM)stxrunner Wrote:  

(02-15-2016 11:40 AM)TempleU22 Wrote:  I love how fans most cinci and UCONN fans won't simply acknowledge the flaws behind Kenpom and Sagarin. It's just as flawed of a system as RPI. I understand that you love seeing your team that high and will fight for the ratings until the bitter end. I would think that the committee is smart enough to see the results that temple has accumulated in actual game action, not some silly metric the values how much you win by.

Who said the other ratings systems don't have flaws? They all do. That's why we have a human committee to weigh ALL the data. Temple fans are just horses with blinders on only seeing what they want to. UC is on the bubble because they have good ratings in some metrics, and bad ones in others. That bolded statement is laughable. The RPI is by far the most flawed metric, because it oversimplifies records.

By the way, this RPI system Temple is clinging to so desperately, still has them behind SMU, UConn, Tulsa, and Cincinnati.

You fail to see that a team's whole season counts. Sorry you messed the bed in the non-conference. If you had capitalized and got a couple even decent wins, you wouldn't be in this position right now. When you only beat teams with 200+ RPIs/KPs/Sagarins etc in the non-conference, you need to be almost perfect in conference. And losing by 27 to UH, to ECU, and Memphis is far from perfect, even when paired with wins over SMU, UConn 2x, UC 2x, and Tulsa. Those wins just got you in the conversation. Now you need to go over the top.

You beat Nova on Wednesday, and a lot of your problems are fixed.

I'll agree with a lot of this. Not sure I ever had blinders about Temple. It is a team that found itself in mid season, still has flaws, even got lucky, and is gritty as hell. I like watching them play.