PER Stats After 5 Games

[B-Roy]

You forgot to include his "low basketball IQ". That adds credibility to any Outlaw post.

Do you disagree with me? He had a pair of <20% shooting nights sandwiched between his good games.

 

[BlazerCaravan]

And KingSpeed, do you really believe what you said? That Travis has been the 2nd-best player this year (so far), based upon his PER through 5 games? I understand if you were skillfully using hyperbole, but something tells me you believe that and are using the PER to back it up.

To be fair to Kingspeed, four things are leaning in Outlaw's favor:

1 - a couple of really good games inflating his PER
2 - a lot of mediocre and bad games from other (non-Roy) players depressing their PERs
3 - a little bit of actual improvement
4 - PER doesn't really weight "driving instead of shooting a wide open three at the end of the shot clock" heavily enough

 

[PapaG]

Do you disagree with me? He had a pair of <20% shooting nights sandwiched between his good games.

So, does that mean that everybody else on the team outside of Roy has been consistently bad?

 

[B-Roy]

So, does that mean that everybody else on the team outside of Roy has been consistently bad?

I never said Outlaw was playing the worst on the team. As far as I'm concerned though, Aldridge, Oden, Joel, and Webster have all been consistently more useful, especially when adding in defense.

 

[BlazerCaravan]

Outlaw's always been that way, though - wildly inconsistent - averaging out to, well, average (15.1 last year) which is totally respectable for a bench guy. All you can really say about the PERs so far is that Outlaw's had a couple of really good games early. In fact, that goes for everyone in a general sense: after five games, the PER is representing what we've seen. Some good games from some players, some bad games from others, some average games. The PER is a symptom, and later in the season will hold much more weight than it does now.

...though, this does bring up an interesting idea: a zeitgeist PER, a rolling 10-game average PER for every player. It'd require a computer to churn out the stats, but I think it's be a great way of measuring a sort of "temperature" for the players... a sort of biorithim of past performance. It'd also be sweet to see that graphed out independent of the overall average PER of a player.

 

[BrianFromWA]

isn't PER a proprietary formula, though? I mean, I have a spreadsheet of each game's box score, with advanced stats and metrics cobbled in and player-specific totals...so I have a lot of the tools. I don't think PERs something I can just get the formula for and plug in, can I?

 

[BlazerCaravan]

isn't PER a proprietary formula, though? I mean, I have a spreadsheet of each game's box score, with advanced stats and metrics cobbled in and player-specific totals...so I have a lot of the tools. I don't think PERs something I can just get the formula for and plug in, can I?

Yeah, I think it is, but you can probably formulate a rough PER-like advanced stat with a little thought as to the conceptual impact of the numbers. The trick of PER is to find the number of possessions in a game, and divide that by 48 minutes to get a Possessions per minute average. Then, multiply that by how many minutes each player played to get the "per possession" divisor for each player. If a player played 28:48 minutes in a 100 possession game (60 possessions), and scored 20 points, he gets .333 points per possession.

 

[Minstrel]

isn't PER a proprietary formula, though? I mean, I have a spreadsheet of each game's box score, with advanced stats and metrics cobbled in and player-specific totals...so I have a lot of the tools. I don't think PERs something I can just get the formula for and plug in, can I?

You can, if you're willing to do a lot of calculations (or research) to figure out league-wide rates for things:

http://en.wikipedia.org/wiki/Player_Efficiency_Rating#Calculation

Hollinger has published his formula in a book, I believe. In some form or another anyway, because it's been out there for a while. That's how Basketball-Reference.com has the PERs for players.

 

[Boob-No-More]

What? Outlaw has been our second best player this season? You wouldn't know it from all the haters in here.

Come on Speed, you know better than that (don't you). PER is HEAVILY weighted towards offensive production. Outlaw's scoring has been fine, as usual, but what about the rest of his game? All this stat does is tell you Outlaw has been our second best player half of the time he's on the court - the half when we have the ball. While PER does include basic, easily measured defensive stats (steals and blocks) it makes no attempt to include a players ability (or inability) to guard his opponent. If you include defense, Outlaw definitely has NOT been our second best player. PER doesn't tell you how many times he gets lost on defense and allows his man an easy, uncontested basket. In spite of his scoring, Outlaw' defense would need to improve a lot before I'd be willing to call him our second best player.

Just for the record, I don't consider myself an Outlaw hater. I'm more of an Outlaw agnostic. In my opinion, he does some things well and other things poorly. He's a good offensive spark of the bench against the other team's second unit, but when he plays extended minutes against the other team's starters, his flaws become all the more obvious and are easily exploited.

He really is a modern day Vinnie Johnson - and I'm fine with him in that roll. Johnson was great coming off the Piston's bench to give them bursts of instant offense. His last year in Detroit, he was forced into the starting roll when Isiah was injured and it truly exposed his weaknesses. Like the Microwave, if Travis' minutes are limited to the low 20s, he's fine. It's when he starts getting more than about 24 minutes in a given game that it causes problems, because that means he's playing extended minutes against the other team's starters who will exploit his weaknesses.

I'd trade Outlaw in a minute if I felt it would help this team. Of course, that would depend on what we got in return. But, I was never in the "he should be renounced" camp. He does have value as a 6th man, and it would be foolish to just give that away for nothing.

BNM

 

[MARIS61]

It's not Miller's fault. It's everybody else around him (even though they all had better +/- stats).

If the Blazers were a marching band, only Andre would be in step with the music.

 

[jlprk]

...though, this does bring up an interesting idea: a zeitgeist PER, a rolling 10-game average PER for every player. It'd require a computer to churn out the stats, but I think it's be a great way of measuring a sort of "temperature" for the players... a sort of biorithim of past performance.

Efficiency is just as good as PER when you are comparing teammates. It's true that unlike efficiency, PER includes an individual's production in proportion to his whole team. But efficiency does that too (in effect) when comparing teammates, since they have a common divisor and it's unnecessary to standardize them among teams (they're on the same team). So to compare Blazers against other Blazers, here is efficiency for the last 10 games. 2009-10 is too young, so I'll link to the last 10 games of 2008-09, including playoff games.

http://www.hoopsstats.com/basketbal...d-trail-blazers/team/profile/09/25/13-6-2-eff
(scroll to bottom)