RPI - Pomeroy - Sagarin Rankings 12-09-09

[Niall of the Nine Hostage]

Interesting how our lack of quality opponents have affected the current rankings. The Pomeroy is really confusing. Last week UNC was higher than us, but after we defeated them, they jumped to 40. We moved up about 9 spots. Its early for all these systems but it seems like the Sagarin is probably the closest in terms of predicting tonights spread.

Real Time RPI http://www.realtimerpi.com/
7 Connecticut 6-1 0.6935 14 0.6313 Bige 0-0
38 Kentucky 8-0 0.6214 170 0.4952 Sec 0-0


Pomeroy http://kenpom.com/rate.php
35 Connecticut
65 Kentucky


Sagarin Rankings http://www.kiva.net/~jsagarin/sports/cbsend.htm
16 Kentucky
17 Connecticut

IMO UConn doesn't have the inside game to keep our bigs in check. Cousins will have a big game. He just doesn't have the experience in playing against bigs his own size. I'm sure he didn't get much of it in HS, and it was glaring in the UNC game. Tonight he'll be back to form.

Wall has everyone's attention. Dyson has something to prove tonight and I think that will be his mistake. He has a tendency to make bad decisions and I think Wall will get into his head. Kind of like a Cousins self destruct against UNC, although not quite as bad as Cuz's. UConn will go as Dyson goes.... At least that's what I'm hoping.

If we can just manage the manageable stuff we will be ok. Like hydrating well John, suck it up and quit whining Cuz, reduce the turnovers to under 15, and hit those dog gone free throws, especially the front ends of the 1 and 1. I think we'll go to the line a lot tonight.

A healthy Bledsoe would go a long way. They have 2 fine guards.

[JFCats22]

Our pathetic strength of schedule is hurting us in these rankings. After playing UCONN and UL that will increase some going into SEC play.

[Disgrantled]

Our pathetic strength of schedule is hurting us in these rankings.

Not sure why you have such an issue with this.

[Will Lavender]

Re: the Pomeroy numbers, schedule has nothing to do with it. The numbers are adjusted for competition.

What's hurting us there is the fact that we turn it over all the time. We don't play very fast, we turn it over a lot = low offensive efficiency. The difference between 14 turnovers in a game and 20+ is a world; for example, Carolina jumped about 15 spots after the loss on Saturday, while we dropped. That's because they were eight or so more possessions more efficient than we were, and that's a noticeable difference.

If this team finds a way to put a string of games together where we're turning it over 12-15 times a game, we'll make a jump of about 50 spots because we're in the green (and that's good) in a ton of numbers inside our individual Pomeroy profile.

[blewis18]

Re: the Pomeroy numbers, schedule has nothing to do with it. The numbers are adjusted for competition.

What's hurting us there is the fact that we turn it over all the time. We don't play very fast, we turn it over a lot = low offensive efficiency. The difference between 14 turnovers in a game and 20+ is a world; for example, Carolina jumped about 15 spots after the loss on Saturday, while we dropped. That's because they were eight or so more possessions more efficient than we were, and that's a noticeable difference.

If this team finds a way to put a string of games together where we're turning it over 12-15 times a game, we'll make a jump of about 50 spots because we're in the green (and that's good) in a ton of numbers inside our individual Pomeroy profile.

Will Lavender = Closet mathmatical genius!

[Will Lavender]

Will Lavender = Closet mathmatical genius!

I'm awful at math. My ACT score was just a little higher than what my five-year-old could pull off.

Basketball stats are a little easier to digest than calculus. It's the difference really between reading stereo instruction manuals and Archie comics.

[blewis18]

I'm awful at math. My ACT score was just a little higher than what my five-year-old could pull off.

Basketball stats are a little easier to digest than calculus. It's the difference really between reading stereo instruction manuals and Archie comics.

I know exactly what you mean. I don't know if you remember the thread where I figured up what all of the final four teams in the past 5 years had as a Pomeroy rating, but that was fun to me. Make me do resarch on anything other than basketball and I look at you like you've got something wierd on your face.

[Dwight Schrute]

I just want to say that I guess dyslexia kicked in or something, and I read the headline "RIP Pomeroy" and got a bit sad thinking he died.

[ukcats1]

The low rankings didn't keep us from beating UNC :icon_biggrin:

Dwight I hope you have recovered from the shock

[Will Lavender]

BTW, here's Pomeroy's most recent blog entry:

http://kenpom.com/blog/

Looks like as a state we're #2, and the UK page is #3 behind Duke and UNC. We need to get on this, Cat fans.

[AugustaDan]

Re: the Pomeroy numbers, schedule has nothing to do with it. The numbers are adjusted for competition.

What's hurting us there is the fact that we turn it over all the time. We don't play very fast, we turn it over a lot = low offensive efficiency. The difference between 14 turnovers in a game and 20+ is a world; for example, Carolina jumped about 15 spots after the loss on Saturday, while we dropped. That's because they were eight or so more possessions more efficient than we were, and that's a noticeable difference.

If this team finds a way to put a string of games together where we're turning it over 12-15 times a game, we'll make a jump of about 50 spots because we're in the green (and that's good) in a ton of numbers inside our individual Pomeroy profile.

Here's my take on the rankings at this point.

RPI: No comment.

Pomeroy: Don't really know what's going on with that system, but I do know that the early season numbers have been ludicrous. Part of this is that the teams aren't well connected, i.e. there aren't enough games mixing enough different opponents for it to converge meaningfully. That's not all of it though. I won't go into it, but my faith in the system has been shaken. I expect odd ratings based on aberrant results at this point in the season, but I've seen, with the Pomeroy ratings, aberrant ratings based on normal results. That is troubling.

Sagarin: Sagarin uses some initial set of ratings at the beginning of the season and isn't based solely on this season's results. That drives ratings early in the season to fit closer with preconceived expectations.

[Will Lavender]

Pomeroy: Don't really know what's going on with that system, but I do know that the early season numbers have been ludicrous. Part of this is that the teams aren't well connected, i.e. there aren't enough games mixing enough different opponents for it to converge meaningfully. That's not all of it though. I won't go into it, but my faith in the system has been shaken. I expect odd ratings based on aberrant results at this point in the season, but I've seen, with the Pomeroy ratings, aberrant ratings based on normal results. That is troubling.

Hard to be too down on the system when the teams at the top are all really good teams.

But I surely didn't expect UK to be 65th on December 9. Not when we've won 8 games and lost 0. I'm not sure, but I'm almost 100% positive we weren't 65th last year at any point in the season and we had a major turnover problem all season and were a truly abysmal team by the end.

My theory is that we just do a few things that the Pomeroy numbers punish you for, and we've done them consistently in each of our 8 games. Turnovers specifically, but also things like free throw percentage and turnovers forced.

I think it can be said that we've won so far based on individual talent, while our team is pretty raw. The first half of the UNC game showed what we can do, but we're still confused much of the time on both ends of the court.

[NC Cat]

Our pathetic strength of schedule is hurting us in these rankings. After playing UCONN and UL that will increase some going into SEC play.

I tend to believe it helped us as a team, however. With so many young players and a whole new system it was important to get some early games against weaker foes. Even then, we could have lost 3 or 4 of them. And now we're getting back-to-back games against ranked opponents and that will pay dividends down the road.

And if you want to see a weak schedule, take a look at KU's. The Jayhawks won't see a ranked opponent until January 9, a full month from now. Will not playing top teams pre-conference hurt them later on? Seems like it almost has to.

[JFCats22]

I tend to believe it helped us as a team, however. With so many young players and a whole new system it was important to get some early games against weaker foes. Even then, we could have lost 3 or 4 of them. And now we're getting back-to-back games against ranked opponents and that will pay dividends down the road.

And if you want to see a weak schedule, take a look at KU's. The Jayhawks won't see a ranked opponent until January 9, a full month from now. Will not playing top teams pre-conference hurt them later on? Seems like it almost has to.

KU's schedule is a TOTAL joke... not sure how they get away with that!

They have a cakewalk schedule this entire year. They only play 3 teams that are currently ranked in the top 25 (Texas, Tenn, and Texas A&M). If they can get past Texas twice (reg season and conference tourney) they have a chance at an undefeated regular season.

[AugustaDan]

Hard to be too down on the system when the teams at the top are all really good teams.

But I surely didn't expect UK to be 65th on December 9. Not when we've won 8 games and lost 0. I'm not sure, but I'm almost 100% positive we weren't 65th last year at any point in the season and we had a major turnover problem all season and were a truly abysmal team by the end.

My theory is that we just do a few things that the Pomeroy numbers punish you for, and we've done them consistently in each of our 8 games. Turnovers specifically, but also things like free throw percentage and turnovers forced.

I think it can be said that we've won so far based on individual talent, while our team is pretty raw. The first half of the UNC game showed what we can do, but we're still confused much of the time on both ends of the court.

I would argue that the simple fact that Ohio St is ranked 31 places ahead of Florida makes the ratings invalid. Look at their results and try to piece together an argument that justifies their relative rankings.

You can look through the top teams and see some rankings that just don't make sense.

There's a weakness in the system (no way to mitigate the effect of blowouts) that combines with the relative lack of interconnectedness of the teams that yields bogus ratings.

Think about it this way. Of the teams that we know are the best teams, most of their common opponents are teams that we know are bad teams. For example, the only team UK and Tennessee have played in common is UNC-Ashville. We know that Tennessee's 70+ point blowout doesn't mean that they are substantially better than UK's 30+ point blowout, but in Pomeroy's system, that difference is huge. For most all the teams, the basis of differentiation at this point is how much they won by in blowouts.

I would say that Pomeroy's ratings won't start making a lot of sense until we are a few games into conference season.

[BigBlue75]

Just to stay on topic, I have a comment/question.

I know as fans we all have certain areas we like to focus on during the season, be it stats, rankings, coaching decisions, etc. and that's fine. Not trying to make an issue of that.

I guess I need someone to explain to me why there always seems to be so much concern over this RPI/SOS stuff each season? I have friends who literally live and die with those numbers and some get downright angry sometime. I just give those numbers a cursory glance at best and honestly, don't worry about it. For me, in the end, to win it all, you have to beat the best to be the best, regardless of what the numbers spit out by some formula are. I guess that's a simplistic approach, but there it is.

Am I missing the point? Maybe it's because I'm not a number-cruncher.

[AugustaDan]

Just to stay on topic, I have a comment/question.

I know as fans we all have certain areas we like to focus on during the season, be it stats, rankings, coaching decisions, etc. and that's fine. Not trying to make an issue of that.

I guess I need someone to explain to me why there always seems to be so much concern over this RPI/SOS stuff each season? I have friends who literally live and die with those numbers and some get downright angry sometime. I just give those numbers a cursory glance at best and honestly, don't worry about it. For me, in the end, to win it all, you have to beat the best to be the best, regardless of what the numbers spit out by some formula are. I guess that's a simplistic approach, but there it is.

Am I missing the point? Maybe it's because I'm not a number-cruncher.

Some people like to talk about momentum and intensity and wearing lucky socks and others like to talk about the numbers. It's just a different place to begin a discussion.