Starting 5 PER Comparison

[e_blazer]

I don't know how valid this is from a statistical point of view, but I thought it would be interesting to compare the total PER for the starting 5 players for this season's Western Conference playoff teams.

Blazers
Oden23.14
Roy 21.36
Aldridge 18.30
Miller 18.18
Batum 17.31
Total: 98.29

Suns
Lopez 17.26
Stoudemire 22.69
Hill 14.07
Richardson 16.63
Nash 21.67
Total 92.72

Nuggets
Nene 18.98
Martin 14.93
Anthony 22.29
Smith 15.15
Billups 20.25
Total 91.57

Spurs
Duncan 24.79
McDyess 14.17
Jefferson 13.18
Ginobili 22.54
Parker 16.49
Total 91.17

Jazz
Okur 15.65
Boozer 21.42
Kirilenko 18.25
Korver 13.99
Williams 20.62
Total: 89.93

Lakers
Bynum 20.26
Gasol 22.97
Artest 12.12
Bryant 21.95
Fisher 9.27
Total: 86.97

Mavericks
Dampier 14.05
Nowitzki 23.01
Marion 14.79
Butler 13.77
Kidd 17.20
Total: 82.82

Thunder
Krstic 13.77
Green 13.82
Durant 26.23
Sefolosha 9.29
Westbrook 17.90
Total: 81.01

Obviously, given where the Lakers are at today, there are multiple factors that are more important than this simple total PER statistic, notably health, coaching, depth, cohesiveness, etc. But it does point out, I think, why giving up on Oden is a very bad idea. Healthy, he is the difference maker that puts the Blazers into contending status. You could probably argue either way whether he would continue to put up such a high PER over the course of a season if healthy. Still, if you plug in either Camby at around 14 or Joel at around 11, the Blazers fall into a middle of the pack starting 5. IMHO, you just have to give Oden the benefit of the doubt and sign him to an extension. The upside is just too dramatic to pass on.

 

[Blaze01]

Well clearly, that NBA title is ours...I mean after all, our PER is the best and that is what wins games....

Sarcasm aside, yes of course you don't give up on Oden, his age, his potential, you have to ride it out for a few more years at least before you pull the plug...I completely agree...

 

[handiman]

I would venture a guess that those stats are quite misleading for this year's Blazers due to the extensive number of games missed from injuries. No one's PER goes down when they're sidelined, but someone else's likely goes up... At no point did all 5 "starters" play together, let alone 3 or 4 of them playing as well together as those stats indicate.

 

[axs88]

It would also be interesting to add up the VA (Value Added) stat of the top 8. VA (also by Hollinger) is based on PER but accounts for the total minutes played by the player that season (Roy finished 19th (all-star level PER but with injuries) and Aldridge 36th (below all-star level PER but plenty of games and minutes played)).

LINK

 

[Mediocre Man]

Clearly the Blazers were the best team, and would win every title until Roy, Oden and Aldridge retired.


As for your original reason to post this, I agree. Oden is a beast when he is healthy, but is 30 games of 23.14 PER better than 80 games of 19.5 PER?

 

[THE HCP]

I would venture a guess that those stats are quite misleading for this year's Blazers due to the extensive number of games missed from injuries. No one's PER goes down when they're sidelined, but someone else's likely goes up... At no point did all 5 "starters" play together, let alone 3 or 4 of them playing as well together as those stats indicate.

Or you could be positive.

 

[Mediocre Man]

Or you could be positive.

Or choose to live in reality, as he has?

 

[THE HCP]

All the cat who started this thread was saying is that we had some pretty damn good PER #'s by our starting 5.

 

[noknobs]

All the cat who started this thread was saying is that we had some pretty damn good PER #'s by our starting 5.

All handiman did was explain why said PER #'s could be misleading... no need for the Positive Police to give him a citation... I don't think 'being positive' is one of the rules of the forum, and I don't think his post was that negative anyways, just pointing out facts.

 

[THE HCP]

And all I was saying was that anytime somebody brings up some positive stats or outlook on things, people are quick to jump 'em. Almost go out of there way to be negative. Just acknowledge the guy did some interesting research. Agreed that PERs can be misleading in both directions as are +/-.

 

[e_blazer]

My main point is that while the Blazers have good players at every position,Oden is the player that potentially puts the Blazers into true contender status...nothing more than that. Sometimes posting here is like dropping raw mat into a shark tank.

 

[mook]

Thanks for posting this, e_blazer. I enjoyed reading it--certainly not a way I've thought about this before. Perhaps a more accurate measure would be "starting 5 + 6th man". The Lakers look pretty mediocre in PER, but you add his 15.9 PER as a bench player and suddenly they go up a lot higher.

I'd probably also add a multiplier of 1.3 or 1.4 to anyone with a PER over 21. A single guy with a PER of say, 24, usually has a bigger impact on a game than two guys with PER's of 18.

Then, of course, weighing in total minutes played.

Is it all just kind of a silly math game? Sure. But so is fantasy basketball. And really at some point all basketball stats (aside from "championship won") are math games.

It's either this or talk about the Lakers and the Celtics winning a championship.

 

[hasoos]

Or does it send another message. Like PER be damned, it is the defense you play that determines what you win.

 

[blazerboy30]

Clearly the Blazers were the best team, and would win every title until Roy, Oden and Aldridge retired.


As for your original reason to post this, I agree. Oden is a beast when he is healthy, but is 30 games of 23.14 PER better than 80 games of 19.5 PER?

Depends which 30 games they are. Oden posting a > 23 PER for the entire playoffs would have been more valuable than somebody posting 80 games of 19.5 PER.

 

[blazerboy30]

Or does it send another message. Like PER be damned, it is the defense you play that determines what you win.

How do things look (I'm too lazy) if you include a defensive rating such as: "Team_Total_PER / defensive_rating"?

 

[Erroneous Subterfuge]

Depends which 30 games they are. Oden posting a > 23 PER for the entire playoffs would have been more valuable than somebody posting 80 games of 19.5 PER.

except we all know which games Oden played and they weren't the playoffs. because he was hurt. again. so if we're talking about the regular season, cause we are, i'll take 80 games with 19.5PER instead of 30 with 23.5PER eight days a week

 

[Draco]

It would also be interesting to add up the VA (Value Added) stat of the top 8. VA (also by Hollinger) is based on PER but accounts for the total minutes played by the player that season (Roy finished 19th (all-star level PER but with injuries) and Aldridge 36th (below all-star level PER but plenty of games and minutes played)).

LINK

That was my first thought as well, I would love to see this.

I wonder what it would look like though, since PER is a Player Efficiency Rating it seems like the total should just be the overall team's efficiency?

 

[jlprk]

PER is a per-minute stat. So we have 5 players who do well per minute. Problem is, ours don't last as many minutes as the other teams' best 5s.

The 2 reasons are injuries, and subbing under the misguided belief that Pritchard has formed a deep team.

 

[Ed O]

I would venture a guess that those stats are quite misleading for this year's Blazers due to the extensive number of games missed from injuries. No one's PER goes down when they're sidelined, but someone else's likely goes up... At no point did all 5 "starters" play together, let alone 3 or 4 of them playing as well together as those stats indicate.

This is, I think, just right.

Players' PER is heavily influenced by stats that overlap with other players'... a team can only score so many points a game, so a scorer going to a good team will probably see his PER go down. Same with rebounding.

There ARE stats that are complimentary (assists being the obvious one, reduction in turnovers being another) but it seems pretty clear to me that the starting five kinda fattened their PER up on the fact that they weren't all playing together.

With the being said (as a defense of proper evaluation of PER), I agree that it's a reminder what a force Oden can be, and how good this starting five will probably be next year

Ed O.

 

[e_blazer]

I understand the argument that players' PER stats are influenced by who they're playing with, but I'm not sure that I see that it would have made a huge difference in the Blazers' totals if this squad had played together the entire season. Roy's stats have been pretty consistent his entire career, as have LMA's. Correct me if I'm wrong, but I think Oden's high PER is a reflection of his high shooting percentage, number of blocks and rebounds, not so much his scoring. I don't see how his PER does anything but go up with a guy like Miller feeding him the ball as opposed to Blake kicking it back outside. Batum's numbers don't really relate to offense either, so it's not as if he would have been taking shots away from Roy or Aldridge.

Bottom line, I think we have a starting five that's pretty darned good. Health is the primary issue.

 

[handiman]

I don't know if PER bears this out, but Roy, Aldridge, and Miller all got off to pretty slow starts this year. I'd be curious to see what their collective numbers were while Oden was in the lineup. I'd be shocked if it wasn't significantly lower. It's basically a mutually exclusive relationship with that particular statistic.

 

[PtldPlatypus]

Just on a whim, I decided to use whatifsports to play this year's Blazers--assuming good health--against this year's Lakers. For the depth chart, I set Miller/Roy/Batum/Aldridge/Oden as starters with Bayless/Webster/Camby as the primary reserves. I made no changes to the Lakers depth chart. So far, I've simulated it 25 times, with the Blazers as the road team each time, and their simulator has Portland winning 68% of the time.

FWIW.

 

[jlprk]

Sample size only 25. I'm sorry, but you'll have to run it more like 2500 times. Looking forward to hearing from you.

 

[BrianFromWA]

I am as well, but since his computer's going to be bogged down until the 2014 Finals...it was good knowing you, Platypus!

 

[Nikolokolus]

Clearly the Blazers were the best team, and would win every title until Roy, Oden and Aldridge retired.


As for your original reason to post this, I agree. Oden is a beast when he is healthy, but is 30 games of 23.14 PER better than 80 games of 19.5 PER?

I think I know the answer to this one: "The answer is hell no Alex, for $500"