'Cuse Vs. KY on Hollinger Player Efficiency Rankings(PER)

[arbitragegls]

ESPNinsider Hollinger has taken his NBA PER rankings and done so for college players. What I find surprising is that 'Cuse has 6 players in the Top 20 in the BE while KY has 3 players in the
Top 20 in the SEC. Also Hollinger equates each college to others so PER can be compared.
Anthony Davis is far and away the leader in this comparison. However, as shown in chart below 'Cuse has more players and averagely better player ratings than KY. Surprising that the EYE Test and hoopla around the KY players may be just that...they are good but so is the 'Cuse team...equal or better using this metric.

Hollinger would indicate that a team with higher PERs is the team that should win a game compared to competition. You will be somewhat surprised when you look at this metric that includes per minute results of scoring, defense, assisting, rebounding, TOs etc.

Bottom line is 'Cuse's 1-6 players are very good and that the team is led again by Dion, Kris, BT, Scoop and Fab according to these rankings. KY will give 'Cuse a very tough game but I believe that 'Cuse wins.

BE Position:
4. Waiters 25.47
8. KJ 22.38
13. Triche 21.27
18. Scoop 21.14
20. Fab 20.96
33. Fair 19.43

SEC Position:
1. Davis 35.22
9. Jones 24.27
13. Gilchrest 22.06
21. Lamb 20.47
32. Miller 18.23

 

[KingOtis]

I think our rotation plays right into Hollinger's formula, as players are able to be productive with less minutes. UK also has a younger team.

I don't put a lot of stock into what Hollinger says. I think PER, for the most part, tells us what we already know. His top ten tends to mirror the top ten in scoring, with a few exceptions. Volume shooters like Iverson get downgraded. And then you get strange results like Brian Cardinal. If PER dictates that you need to get a guy like Cardinal and give him a bunch of minutes, then PER is flawed.

Would be interested to see how the Mizzou players stack up, though. They have the most efficient offense in the NCAA, so most of their players should be at or near the top of the Big 12. If not, then something is wrong.

 

[orange2win]

According to Hollinger's Player Efficiency Ratings, we have only one player ranked in the Top 100. Something's wrong here.

http://insider.espn.go.com/ncb/holl.../insider.espn.go.com/ncb/hollinger/statistics

 

[arbitragegls]

According to Hollinger's Player Efficiency Ratings, we have only one player ranked in the Top 100. Something's wrong here.

http://insider.espn.go.com/ncb/hollinger/statistics?&action=login&appRedirect=http://insider.espn.go.com/ncb/hollinger/statistics

This is Holliger's ranking on BE conference only(he does conferences also...you are correct in Top 100 only Waiters from 'Cuse and Davis from KY are in Top 100. Interesting that the two top teams only have 1 player each in the Top 100.

 

[arbitragegls]

I think our rotation plays right into Hollinger's formula, as players are able to be productive with less minutes. UK also has a younger team.

I don't put a lot of stock into what Hollinger says. I think PER, for the most part, tells us what we already know. His top ten tends to mirror the top ten in scoring, with a few exceptions. Volume shooters like Iverson get downgraded. And then you get strange results like Brian Cardinal. If PER dictates that you need to get a guy like Cardinal and give him a bunch of minutes, then PER is flawed.

Would be interested to see how the Mizzou players stack up, though. They have the most efficient offense in the NCAA, so most of their players should be at or near the top of the Big 12. If not, then something is wrong.

You are correct. Mizzou players rank 1,4,6,19,26,and 45 in Top 50.
Link below:
http://insider.espn.go.com/ncb/hollinger/statistics/_/group/8

 

[Knicks411]

I actully think volume scorers are overrated; I think I read somewhere that as long as you shoot something like 34% or better, each shot you take increases your PER.

PER isn't perfect, and it only really covers offense anyway. Cardinal had a pretty high PER onbe year ( I don't think it was top 10 or anything), but he was legitimately awesome that year. He lead the league in true shooting%. He didn't play a lot of minutes, and that was obviously a fluke, but you should be able to realize that.

 

[crazyqx83]

I don't follow the logic. Sure, SU has players ranked highly in the Big East but look at the comparison...

UK's #1 player is FAR AND AWAY better in PER than SU's #1.
UK's #2 is marginally better than SU's #2
UK's #3 is slightly better than SU's #3
SU's #4 is slightly better than UK's #4
SU's #5 and #6 are marginally better than UK's #5

So SU doesn't find an advantage until each team's #4 player. How would you categorize this as an advantage for SU?

 

[cusetown1]

I don't follow the logic. Sure, SU has players ranked highly in the Big East but look at the comparison...

UK's #1 player is FAR AND AWAY better in PER than SU's #1.
UK's #2 is marginally better than SU's #2
UK's #3 is slightly better than SU's #3
SU's #4 is slightly better than UK's #4
SU's #5 and #6 are marginally better than UK's #5

So SU doesn't find an advantage until each team's #4 player. How would you categorize this as an advantage for SU?

Yeah we know, Kentucky's god blah blah blah.

 

[KingOtis]

I actully think volume scorers are overrated; I think I read somewhere that as long as you shoot something like 34% or better, each shot you take increases your PER.

PER isn't perfect, and it only really covers offense anyway. Cardinal had a pretty high PER onbe year ( I don't think it was top 10 or anything), but he was legitimately awesome that year. He lead the league in true shooting%. He didn't play a lot of minutes, and that was obviously a fluke, but you should be able to realize that.

Do conventional stats contain flukes?

I'll take Iverson over Corey Maggette every day of the week and twice on Sunday.

 

[Knicks411]

Do conventional stats contain flukes?

Sure. Like remember the year Brian Cardinal lead the league in true shooting%? That was a fluke. He isn't really the best shooter in the league. But he made 44% of his 3's that year, for whatever reason. Also, I'm a little confused about the Cardinal thing; the one year he had a high PER he was 32nd int he league in PER. it isn't like it was saying he was one of the best players in the league or anything. In his career, he has 2 years where his PER was above league average, and one year was a 28 game season. Both times he had true shooting percentages over 60%. I am not quite sure what the issue with Cardinal is. Guys sometimes have fluky years where they make a ton of their 3's.

I'll take Iverson over Corey Maggette every day of the week and twice on Sunday.

As would I. (And as would PER for that matter, which has Iverson with a higher career mark. And since Maggette is a defense liability and Iverson could play some D, the gap would be even bigger).

And as for the OP, I am not sure
A) what PER is worth on the college level, and on any l evel in just predicting a game, and
B) how those rankings show us to be better.

And I don't want to come off as a slave to PER. It's fine, for wh at it is. It doesn't really look at defense at all. It's mainly an offensive stat, and even in that, you always need to consider how players interact with one another on the court and all that good stuff.

 

[Loose Nukes]

This is not how college basketball match ups should be measured.

 

[KingOtis]

Hollinger was pimping Cardinal as I remember, which is part of the reason why he was given a ridiculous contract. And isn't true shooting percentage another one of Hollinger's stats? Make no mistake, ESPN only gave Hollinger a platform because of Bill James and Sabermetrics. That doesn't mean we should embrace his faulty numbers.

Maggette's PER has dropped over the years because he's been given more PT. But several years ago, he was another guy Hollinger was pimping because of his flukey PER. In addition, Maggette was actually a solid defender his first couple of years with the Clippers. He later sacrificed that for offense.

 

[Knicks411]

Hollinger was pimping Cardinal as I remember, which is part of the reason why he was given a ridiculous contract. And isn't true shooting percentage another one of Hollinger's stats? Make no mistake, ESPN only gave Hollinger a platform because of Bill James and Sabermetrics. That doesn't mean we should embrace his faulty numbers.

True shooting percentage was definitely not created by Hollinger. I remember reading about it when I was like 9 or 10. (So this would be like 1993). It's not some convoluted stat like PER. It adjusts shooting% for 3's and free throws.

And in 2004 Cardinal had a good year, it was a fluke in retrospect, but what else would you say about a guy who made 44% of his 3's

Maggette's PER has dropped over the years because he's been given more PT. But several years ago, he was another guy Hollinger was pimping because of his flukey PER. In addition, Maggette was actually a solid defender his first couple of years with the Clippers. He later sacrificed that for offense.

Maggette's career high in PER was 20.3 (Which happened in 2010.) Iverson's career number is 20.9. So Maggette has never had a single season PER better than Iverson's career average. I don't think PER was ever saying Maggette was a better player than Iverson.


And as I said, I don't worship at the altar of PER. It's a pretty good offensive stat, I'm sure there are better ones out there now. It has flaws (though despite what you said, i t favors volume scorers) But no one says field goal% is a bad stat if someone who isn't considered one of the better shooters in the league is first in it. (Let alone 32nd, like Cardinal was in 2004, which was his best year).

 

[jordoo]

I feel that these uber stats are barely more revealing than the old basic stats.

 

[moqui]

I decided to take a look at how the compare position by position, rather than just by rank order. For those purposes, I used CJ as the starting big forward for Syracuse, rather than Rak, since he plays starter minutes.

Syracuse has a big advantage among the guards - cumulative 15.44 among the three rotation guards in each lineup.

Kentucky has a similarly big advantage in the frontcourt, cumulatively 18.78 and 19.1 when you just take the 4 and 5.

Then, Syracuse has another huge advantage off the bench, with a cumulative 18.82 margin. If they ever played, however, barring foul trouble, Southerland for Syracuse and Wiltjer and Vargas for UK would not get many minutes (last night, for example, they played 4 and 2 minutes respectively).

Taken altogether, UK starters are up 9.58 on the SU starters. Taking the top 8 rotation players, SU has the advantage of 9.24. If you just take the top 6, who would likely get the bulk of the minutes in any real matchup, the advantage is UK + 3.34.

In a real game, I think UK would dominate on the glass. SU's chances would come down to our guards' ability to get turnovers and runouts and to extend out on Lamb & Miller to minimize their 3 point threat.