clock menu more-arrow no yes mobile

Filed under:

Stampler's Take: John Hollinger's PER is dead

Okay, okay, message received loud and clear. I'm still right about Bruce Bowen though.


Wow, what a gathering. I'm humbled that so many of you have shown up, on such short notice, in the off-season no less, for this solemn occasion. And you're all wearing such fine suits, even in this interminable heat. That kind of gesture really means so much. I really wasn't expecting this kind of turnout.

But that's kind of the point, isn't it? Death, like life, is all about the unexpected. No matter how much you think you know what's coming around the corner, no matter how much you plan and prepare, there is so much that takes us all by surprise every day the sun rises anew. The next day isn't promised to anyone though, and so here we are gathered, to pay our final respects to PER.

Heaven knows we've had a good run, PER and Spurs fans. Me perhaps more than anyone. I admit, without a speck of humility, that I'm guilty of running PER into the ground, of overworking it, of quite frankly abusing it to the bitter end for my own selfish needs. What can I say? PER made Manu Ginobili look awfully good over the years, and how can anything that makes our patron saint look good be a bad thing?

The eternal lesson, as always: Everything in moderation. (sigh) Everything in moderation.

Not only was I living in denial these past few years, naively using PER when better, more accurate statistical metrics for NBA players were developed for the public at large to learn and use, but I guess I was just trying to breathe oxygen into its lungs well after the doctors declared that its relevance was gone forever.

PER was more than a simple statistic invented by a bicycle-riding bald man who now works for the enemy. It was my companion, my shoulder to cry on, my confidant, my crutch. It was my everything. I mean no disrespect when I say it was more of a friend to me at Pounding the Rock than any of you.

So here lies PER on PtR (2005-2013). We had so many good times together, howling with derision at fans of Ray Allen, Reggie Miller, Monta Ellis and Vince Carter. Oh, how we laughed and laughed. It brings a smile to my face just thinking about those days. Still, time marches on for all of us, unrelenting as Skip Bayless' love for Tim Tebow.

Time has run out on PER, and henceforth I will no longer use it on this site. Godspeed, old friend. If you see pitcher wins and batting average in the afterlife, give them a high five for me.

*   *   *

Okay, so what's the point of this silliness? Well, the other day PSherman42 took me to task for my "sobering" prediction of what the 2013-14 season (and beyond, possibly) would bring for Kawhi Leonard. That column was written, naturally, as a response to J.R. Wilco's original positive article about Kawhi's goal to be a superstar.

So what you have here is a response to a response to a response. A prouder moment for the internet you will not find anywhere.

Believe it or not, but I wasn't the least bit hurt or defensive by the post. In fact, I'm glad somebody wrote it. I'm all for someone watching the watchers, as it were, and it's about time I was called out for leaning too much on a stat like PER that's run its course. I've long railed on the likes of Mike Wilbon:

Darren (Milwaukee, WI)

Mike, I get how you're more "old school" than "new school" when it comes to stats, but why are you often so dismissive of people who are trying to analyze sports in new and different ways? Rather than calling it all "junk", doesn't it make more sense to acknowledge that sometimes our eyes lie to us and we can benefit from having more information available?

mike wilbon
(1:48 PM)

Does it make sense for you to acknowledge that sometimes stats lie...or more important than "lie" that they don't explain everything. There's no metric to tell you how a 76-yard run on 3rd-and-six late in a game demoralizes a defense or how much it lifts the team that scores...there's no metric to measure leadership or emotional steadiness. So, different strokes for different folks.

Wilbon, Bruce Jenkins and Jason Whitlock are so militantly anti-thought when it comes to advanced, objective analysis that they bully anyone who tries to educate them by dismissing them as "nerds." It's like "Revenge of the Nerds" come to life, only the "Ogre" character isn't some jock but rather one who writes about them.


Well, it'd dawned on me that I'm being a hypocrite by relying on PER almost exclusively and not looking at newer, better stuff. I freely admit I have much to learn, especially if I want to write about this stuff for a living. There are very few things more useless than a journalist who thinks he knows it all and doesn't have to learn or research anymore. Somewhere along the line, fellas like Wilbon, Jenkins and Whitlock all lost their way. That's not gonna be me, and if it ever feels like it is, please, let me know about it, preferably in a respectful way.

*   *   *

With that being said, this particular section in PSherman 42's column note struck me as the lone sour note in an otherwise excellent post.

PER doesn't measure defense

PSherman42: The only two defensive stats that are taken into account are steals and blocks, something which even John Hollinger, the creator of PER, has acknowledged as a flaw:

Bear in mind that this rating is not the final, once-and-for-all answer for a player's accomplishments during the season. This is especially true for players such as Bruce Bowen and Trenton Hassell who are defensive specialists but don't get many blocks or steals.

There are players that get big steal and block numbers (like Monta Ellis and Deandre Jordan) despite being subpar overall defenders. Think about that for a second: PER actually rewards a guy like Ellis on the defensive end more than Kawhi Leonard.

J. Gomez: After reading that, I went and looked at Bruce Bowen's career PER. It was 8.2. What a scrub. So the contributions of low-usage, defensive-minded players, which is what Kawhi was last season, will never be appropriately gauged by PER.

What Sherman and probably even Gomez, a longtime Pounder, likely don't remember is that my very first guest post at PtR, way back during the 2005-06 season, was partly a rant about how Bruce Bowen is overvalued by Spurs fans. Since then, my contention has only grown more emphatic: Bowen is the most overrated player in Spurs history.

While I think there are indeed flaws with PER, it's "failure" to protect guys like Bowen and Hassell isn't necessarily one of them. Those guys don't do well in PER because they simply weren't very complete ballplayers. They're just a different version of a Matt Bonner, a specialist, someone who needs to be on the floor with a bunch of good players before he looks competent.

I won't pretend to analyze Hassell because he wasn't a guy I followed much, so let's focus on Bowen here. He couldn't dribble. He couldn't pass. He wasn't much of a rebounder. He could shoot from, literally, one spot on the whole floor. He couldn't create a shot for himself or draw fouls, which was okay because he was a terrible foul shooter anyway. Defensively he did his job very well, but he didn't cause much chaos by turning defense into offense because he didn't create any live-ball turnovers (i.e. steals and blocks). Mainly, his role was to funnel wing scorers to their weak side, to the baseline, and toward the gaping maw of Tim em-effin Duncan.

*   *   *

Honestly, I think Hollinger is just being diplomatic in a way. Politically correct to less-talented players. Being a "defensive specialist" doesn't get valued by PER (and other metrics, as we'll see) because it shouldn't be. It's akin to baseball fans complaining about why middle relievers don't rack up a bigger WAR for the season when really the difference between the best seventh-inning guy in the league and the worst is like maybe 1.5 wins over a 162-game season.

Bruce Bowen was the basketball equivalent of Javier Lopez for the San Francisco Giants. Very useful when you need to get Joey Votto, Chase Utley, Ryan Howard or Adrian Gonzalez out in a big spot, but he's not gonna give you seven shutdown innings every fifth day against a full lineup of lefties and righties. Lopez is the quintessential LOOGY, and Bowen was the NBA DOOGIE (defense-only-offensive-game-is-excrement).

Pop and countless other coaches always talk about what nonsense the concept of individual defense is. Sure, each guy on the floor has assignments, rules to follow -- play over the screen, hedge hard on pick-and-rolls, switch everything, close out hard on this guy but let that guy have the open shot, etc. -- but for a defense to really work, all five guys have to work in sync because superstar players are just too talented for any one defender to stop them consistently.

Was Bowen a great wing defender? Obviously. One of the best ever. But he was only as good as the defenders behind him -- Tim Duncan, and dudes like Rasho Nesterovic, Robert Horry and Nazr Mohammed -- allowed him to be. If Bowen had his entire career in front of DeMarcus Cousins, then nobody would've even noticed his defense because it simply wouldn't have mattered. Having Timmeh the Golden God behind him allowed Bowen to gamble, to get right up on a guy and crowd their jump shots, because getting blown by to the rim wasn't a bad thing, as long as the dribble drive was gonna come from the direction the Spurs wanted it to.

Bowen had one specific skill, and it worked well within the Spurs system because he had not only one of the best defensive big men ever behind him (actually, two of the best for one season), but on the other end of the floor the Spurs had three superstar offensive players who were all incredibly efficient. That doesn't mean that on his own he was ever much of a player, or particularly valuable. I mean, let's be real, a team of five Tony Parkers or five Manu Ginobilis would absolutely destroy a team of five Bruce Bowens. The score would be something like 115-50 in a 48-minute game.

And before you counter with the argument that it's better off that, in Bowen, the Spurs had a guy who could focus on defense and didn't need the ball since they already had three high-usage guys, I'll respond that there's a difference between a player who doesn't need the ball and one who can't do anything with it when he finds it in his hands. Kawhi Leonard doesn't need the ball, but he sure as heck knows what to do with it. He's already a far more valuable player than Bowen ever was, and he does a ton of things on both sides of the floor that Bowen never could. If the Spurs had Leonard during the big three's peak seasons, they'd have likely run off five or six straight titles.

Being a great individual defender is awesome (and I dare say, necessary) if you're a big. Not only are you shutting down your man, but you're essentially shutting down everyone else too by protecting the rim. And you're probably pulling down a healthy share of rebounds (cue Pop's voice in your head "a defensive possession isn't over until you get the board"). But a wing defender who doesn't cause turnovers and just funnels people toward the big who does the hard part? Sorry, unless you're bringing something else to the party, you're just a guy.

*   *   *

While I agree that my PER argument wasn't the best way to make the case that Leonard's progression might underwhelm next season, the larger point I was trying to make was that his development may not be as rapid as we want, not necessarily because of some shortcoming he has but more because he'll still be playing with a couple of star teammates who have the ball a lot. It's not his fault, it's not anyone's fault. It's just how it is.

There's a reason that superstar wing players like Michael Jordan, Kobe Bryant and LeBron James have never played with stud point guards. It's because, for all intents and purposes, they were the point guards, and their backcourt mates were short guys who could knock down open jumpers and otherwise just stayed out of the way.

As long as Leonard has Parker dominating the ball, he's not going to realize his full potential as a player. He'll have a few more plays called for him this season, but it won't be a dramatic increase, and I expressed in my last post my concerns about Leonard either as the ball-handler or the finisher on a pick-and-roll. He can run ISO stuff, but I'm less convinced he can ever turn into an effective pick-and-roll guy.

I'm still quite ecstatic that he exists and that he's one of ours, and figures to be for years to come.

*   *   *

So what stat should I use now?

Wins Produced?

Consider me skeptical about a stat that had Leonard and Danny Green as more valuable than Parker the past two seasons. In fact, Wins Produced kind of hates Tony Parker. It states that Brent Barry was worth more to the Spurs than Parker in each of the Spurs past two championship seasons and has Parker as something of a non-entity during his injury-plagued 2009-10 season. Wins Produced, on the other hand, likes Ginobili quite a bit, so I'm willing to give it a chance. Bowen, incidentally, only had three seasons as a Spur where he was worth more than six wins, according to Wins Produced, and never more than 7.7. He was particularly weak in the last two seasons they won titles.

Win Shares Per 48 minutes?

Looks good to me! I especially like the part with all the "Ginobili"s, "Duncan"s and "Robinson"s. Also, four Spurs in the top-13 all-time? Pretty sweet. Pretty sweet. Can't complain about any of that.

Here's the thing though... Leonard had a very respectable .171 WS/48 as a rookie, and then it dropped to .166 last season. Not a strong case in his favor. Kawhi lagged fourth on the Spurs in that stat last season, well behind Parker, Duncan and Tiago Splitter.

Incidentally, Bowen's career WS/48 is a pitiful .086 and his single-season high as a Spur was 1.11 in 2004-05, which Wins Produced lists as one of his worst seasons. Whatever he contributed, WS/48 can't find it.

Adjusted +/-?

Well, the names up top certainly look like the right guys. Once we get into the second tier though, I grow dubious. Nene? Jamal Crawford? Vince Carter? Oy vey. Also, you'll note that the name right above Leonard's is none other than George Hill, which I found interesting.

This is another stat that hates Bruce Bowen. Pick a year. It is not Bowen's friend.

It looks like there is no perfect, all-encompassing stat. At least none that I can find. But I promise, earnestly, to be curious and open-minded and to keep looking. You're all more than welcome to make arguments for your stat of choice in the comments. I'll read it all.

But before I go, let's take a look at one final stat. Defensive rating. As basic as it gets. How many points a player allows-per-100 possessions. Surely if there's any stat created to illustrated Bowen's value, it's this, right?

Duncan is second all-time, at 95.37. Robinson is fourth, at 95.65. Ginobili is 28th, at 99.69.

Bowen is 86th, at 102.25. One spot below...

wait for it...

Pete Maravich.


More from Pounding The Rock: