Statistical Analysis. Humor. Knicks.

Wednesday, September 17, 2014

Q-Rich Poll

Ok folks, I’m always curious what my readers think, so another poll for y’all. As always I ask that if you have a minute, leave a comment to let us know why you voted the way you did. You can always leave it anonymously, if you wish. Today’s subject: Quentin Richardson.

According to his stat page on Basketball-Reference.com, Q-Rich’s PER his rookie season was an impressive 16.6. He topped that his second year, going up nearly a point to 17.4. Unfortunately, he’s never topped that mark. His third year, a time when many players show marked improvement, Richardson’s PER dropped anemically to 12.5. In his last year for the Clippers, he brought it to the league average (15.1), but even last year on a super-charged offensive team, Quentin only manged a 13.5.

In addition to his decline in production, Richardson also became less durable. Despite missing only 5 games total his first two seasons, Quentin missed 40 in the next two. One of the knocks against him is his reported chronic back pain, something Richardson claims is no longer an issue.

So here’s the poll, and remember, leave a reason in the comment section.

60 comments on “Q-Rich Poll

  1. Michael

    Q has always been an enigma. At Depaul, he was supposed to the be the savior and the best 2-guard in the nation. He never lived up to that. He seems to be satisified in spotting up all the 3-point line (both in college and the pro’s). He is decent at this but what would make him much better is using his size to post-up and driving more. When he does these two things he is very effective. One of the reasons I said he would be below 14.4 is that last year he had a past first point guard and a good set-up man in Johnson to get him the ball. He was probably the third option on the team as well. In NY, he will be the third option, but with ball-hogs Marbury and Crawford he will see the ball less. So, I think he will be less productive. If he wasn’t productive in PHX, then I can’t believe he will be more productive in NY especially with Brown probably slowing the pace down and if they sign Walker he will never see the ball.

  2. Ted

    I don’t know how PER is calculated. If he improved his shot selection and thereby his FG% and 3 point FG%, remember that the Suns offense was designed for them to take the first semi-open look and Q clearly did as he was told, would that drive up his PER? Considering that the only evidence we have was on the Clippers and the run and gun Suns, I’ll wait till he plays under LB before making too many conclusions about him as a player, same goes for Jamal Crawford.
    The bottom line is that if a PER of 13.9 is high enough to get the Suns 60 wins and to the conference finals it’s high enough for me. It is also interesting to note that at about 15 and 10 the Suns other wings, Joe Johnson and Jim Jackson, didn’t do any better. So maybe something about the Suns offensive game plan isn’t suited to the way PER is calculated.
    If Q comes off the bench you’d love for him to be as good as the best team in the league (SA)’s 6th man: Brent Barry, whose PER was only 14.3 last year. In fact the Spurs only have three guys with exceptional PERs and the rest of their team is at or below the league average (Bruce Bowen’s is below 10).
    As Hollinger himself says, PER is a nice statistic but it is not the only factor in determining how valuable a player is. For example, one thing you have to love about Q is that if Frye is going to get minutes you’re going to give up rebounds in the front court so a guy who can grab 6 per in the backcourt is a big help. Q also reduces Crawford’s minutes and adds another scoring option around Marbury which defenses have to account for.

  3. Paul M

    A higher field goal percentage would definitely increase a player’s PER. The PER formula rewards a player for making shots and penalizes for missing them (this is inexact since I don’t have the book around, but that’s the gist of it). So naturally, if Quentin hit more shots, his PER would go up.

    “The bottom line is that if a PER of 13.9 is high enough to get the Suns 60 wins and to the conference finals it?s high enough for me.”
    This seems faulty for a number of reasons. In this example, if Richardson were able to bump up his PSA a bit, maybe the Suns could have won more games. In more concrete terms, if he had fewer games where he shot ~20% and under, that may have been the difference between Phoenix as a 62 win team and Phoenix as a 65+ team. Probably a weak example (this can be applied to anyone), but you get my point. You shouldn’t just accept substandard (at least this year) play from a player just because he was a starter on a good team.

    “So maybe something about the Suns offensive game plan isn?t suited to the way PER is calculated.”
    Not the PER, per se, but Hollinger’s other statistics can lend some light on this issue. The Suns’ great offensive season is pretty clear if you see who they ran the offense through. The two players who shot the most, Amare Stoudemire (25.2 Usage Rate) and Steve Nash (22 Usage Rate) both had PSAs over 1.20. Marion and Joe Johnson both had PSAs at 1.10, and Q had a PSA of 1.04 (above the league average, I believe). Best of all, the team wasn’t really that turnover prone, so Phoenix’s offense was incredibly efficient.

    As for Richardson himself, he seems very hard to read for me. I went with the safe option of 14.5-15.9, but I have no real basis behind this. This season, he was a pretty one-dimensional player, but it didn’t seem like he was asked to do much more (this could also be a reason why his PER was so low, since I can’t think of very many catch and shoot players who have good PERs besides Fred Hoiberg). He’s still a solid rebounder for a two, and his shooting stroke still comes and goes, but I really don’t know how he’ll turn out.

    One thing I didn’t really notice is that Q has some defensie value. Hollinger’s newest article in the NYsun says, “Richardson used to care little about defense, but that changed last season. He took an insane number of charges with Phoenix and was one of the keys to their somewhat-effective defense. That willingness to be physical will quickly endear him to Brown, as will his ability to post up and shoot the 3.” I had no idea.

    Sorry if I went a little nuts. I’m kinda new to this statistical stuff. I’m probably wrong about some things, too.

  4. Ted

    First of all, just because he had a low PER does not mean that he played substandard. Bruce Bowen’s PER was below 10 and I think you would be hard pressed to find anyone who would say he played substandard last year or that he is not a big reason that the Spurs won a championship. There are many parts of the game that PER does not even pretend to address. You can also notice that role players on very good teams often have lower PERs than supporting players on terrible teams.

    I don’t think it seems that flawed to say that if someone was a starter on a team that won 60 games then your team could possibly win with them regardless of any one statistic if he is not asked to play a larger role. Might the Suns have been better off if last season they got a player with a higher PER than Q? Maybe, maybe not. For example, Darius Miles had a PER that was 2 points higher than Q’s, but he can’t hit the three so he might make the Suns far less efficient. Sometimes how well a given player fills their role is far more important than how “efficient” they are statistically. All these statistics can be used to give you a better idea of how good a player is, but they only tell part of the story.

    Also, the Suns won more games than any other team in the league during the regular season winning 65 games would not have gotten them anything. They would have still been the number one seed and lost to the Spurs in the conference finals.

    As for your second criticism, I was simply saying that just as PER does not account for a shut down defender like Bruce Bowen maybe it does not account for the way that the Suns use their wings in their offensive system because all of their PERs are lower than their perceived values. Also, the Suns took the league by storm by jacking up the first semi-open shot, so maybe it is Hollinger’s stat and not the Suns’ 60 win offense which is flawed.

  5. Paul M

    No, I think you’d find a lot of people who think Bruce Bowen is substandard. He’s an absolute trainwreck on offense. His only offensive move is an open shot from the corner. Other than that, he can’t score, never gets to the line, and is an atrocious rebounder for a guy who is 6’7.

    Naturally, he’s not around for offense. No one denies that he’s a great defensive player, but the question is whether his defensive abilities are enough to offset is awful, awful offensive game. The Spur core was good enough to get away with playing someone like him, but that shouldn’t trick you into believing that the spot couldn’t use an upgrade.

    As for you second point… well, first off, Darius Miles isn’t a shooting guard. Second, his three point field goal percentage actually wasn’t too much worse than Richardson’s (Richardson shot .358, Miles shot .348. To be fair, Richardson probably shot twenty times more threes). That’s not that point, anyway. Richardson had limited value last year, and I don’t think people should just heap praises on him just because he was on a good team. The same applies to Joe Johnson, who is a star all of a sudden, in spite of rather average play (I only bring this up because the Hawks are offering him a ludicrous deal). I’m not saying that they did not contribute to the Suns’ 62 win season. All I’m saying is that their contributions should not blind people into overvaluing them (too late in some cases. Yes, I have to rag on that offer).

    Third point is moot since we’ll never know.

    Lastly, you’re right. The PER does not really address man to man defense, which is why Bruce Bowen is anually ranked near the bottom of the league. Still, Hollinger did a study on defensive specialists in his first Prospectus book (it was even on Bruce Bowen), and he noted that his defense contributions may have added 3 or so points to his PER that year (something like that. I don’t have the book on hand), which makes him look better, but he still had an adjusted PER that was worse than the league average. The PER is imperfect. I don’t think anyone disputes this. Some players don’t rack up steals or blocks and are good defensive players, but others swipe the ball very often but have limited defensive ability. Also, the PER doesn’t take into account charges (Why, why, WHY isn’t this recorded?). Even so, it’s still an excellent measure of player effectiveness. Hollinger’s explanations of the formula make intuitive sense, and it measure offensive ability quite well.

    “Also, the Suns took the league by storm by jacking up the first semi-open shot, so maybe it is Hollinger?s stat and not the Suns? 60 win offense which is flawed.”
    I addressed this above, but think about it. The three Phoenix stars all have a PER above 22, they all had a PSA above 1.1, and those three had the highest Usage Rates (meaning that most of the offense ran through them). Phoenix avoided turnovers very well, too. The PER, PSA, and Usage Rates are all Hollinger’s stats, and they explain Phoenix’s season quite well.

  6. Carlos

    It’s hard to predict Richardson’s output, because his shooting % has changed so much over his career. He has a good rebound rate for a guard but a poor assist rate and doesn’t get many steals or blocks either. If he does not shoot that well and doesn’t get that many shots (which happened in Phoenix last year), his PER will be low. If he gets enough shots but shoots terribly, his PER will be even lower (which happened in his third season in the Clippers). The key is that most of his value in a PER-like number will come from his scoring and Q has never been a very efficient shooter (highest PSA of 1,07). To have a PER of 15 or more, he will need to get a good amount of shots (a usage rate of 20 or higher). In a team with Marbury and Crawford, I find that unlikely.

  7. McDaddy11

    But Bruce Bowen is winner, whenever he comes on court he is winner, cause his only job is to play defense, and when some big-shot player has to change his style cause of Bowen, has to take bad shots, he wins. I would like to see someone like that on Knicks lineup. Q-Rich has it all, he can penetrate, he can shoot, he can jump, he is strong, he has potential to play quality d, but with Marbury, Crawford, and now Brown I think his PER will go down. But I dont think that is most important thing, if he can make this team better in any other ways, Brown can help him to be better, same goes for Crawford. Look what he did for Hamilton. I think here he will take less shoots then before, but good ones so he will average somewhere between 10-12, maybe he becomes surprise, never know, but I find it unlikely, there is better chance that he will discover N.Y. night life, and everything goes to hell.

  8. Ted

    What’s your standard if you’re going to call Bruce Bowen substandard? MJ? Kobe? The guy is one of the best on ball defenders in the league and a big part of the league’s best team defense, plus he can hit the three. The reason other statistics exist is because they help a team achieve the ultimate stats: Ws and Ls.

    Bruce Bowen was often charged with shutting down the other team’s best perimeter guy and he filled that role as well as all but a couple guys in the league could. Artest or Kobe could provide that and offensive productivity too, but Bowen fills the role he is given in the Spurs system and they win as a result.

    Bowen is a role player and if you look around the league at the best teams now or historically not every player on their team has a PER of 20. Some guys do little things that don’t show up (or in Bruce Bownen’s case a pretty big things) or sacrifice their own numbers for the good of the team.

    Q didn’t play shooting guard for the Suns either, he played the three. But Miles was the first example of a wing with a higher PER that I saw there are countless others you could use. Miles only shot 0.4 threes and made only 0.1 per game.

    “Third point is moot since we?ll never know.”
    I was just saying that if a couple of regular season games had gone the Suns way they still would have lost to the Spurs. I don’t see how say barely winning a couple games they barely lost would help them in a 7 game series against the Spurs. The Spurs were clearly a better team last year. Of course I can’t go back, change the past, and see what happened.

    “I addressed this above, but think about it. The three Phoenix stars all have a PER above 22″
    You can’t look at Q and say he’s exactly this good of a player because his PER was 13.9. I never said anything about Hollinger’s other stats, and neither did the original post I was responding to. Anyway, if you addressed what I was saying I missed it. The three Suns’ “stars” (I have a serious problem with calling Sawn Marion a star) are not wings in the Suns offense. The role of a PG or a C and the role of a wing are drastically different, so all I was suggesting is that maybe what the Suns’ wings are asked to do is not taken into account in PER. Maybe other stats show it, but what I’m saying is simply that PER is not the end all and be all. Clearly Amare and Nash were better players last year than Q, Joe Johnson, or Jim Jackson, but there are other players around the league with higher PERs who might be more statistically efficient but worse overall players.

    When I said ?Also, the Suns took the league by storm by jacking up the first semi-open shot, so maybe it is Hollinger?s stat and not the Suns? 60 win offense which is flawed.? What I meant was: So maybe we need to use a better stat that takes into account whether or not a player fills the role he needs to fill in order for his team to win.

  9. Paul M

    Okay, I think you’re missing my point here. Not once did I say that Bruce Bowen did not have value. As you said, he takes on tough defensive assignments and generally keeps his man quiet. However, his offensive game is horrid and it is questionable whether his defense is enough to justify having to go four on five on offense every night. Taken as a whole, I don’t think that Bowen’s defensive value negates his “offense,” which is why I called him substandard. He’s certainly better than he is rated in the PER, but his defense would have to be worth 5 points (which is immense in this scale) to even be average. It’s a judgment call either way, and I probably can’t convince you otherwise.

    “Bowen fills the role he is given in the Spurs system and they win as a result.”
    Yeah, Tim Duncan and Manu Ginobili had nothing to do with that.

    “Bowen is a role player and if you look around the league at the best teams now or historically not every player on their team has a PER of 20.”

    Of course not. Last I checked, basketball is still a team game. The Pistons took the title last year without a player with a PER over 20 (or even 19).

    Role players obviously have a place in the league, but you probably shouldn’t give them more credit than they deserve. The Spurs can get away with starting Bowen and still win championships, but that doesn’t mean that he’s beyond criticism all of a sudden.

    “Q didn?t play shooting guard for the Suns either, he played the three.”

    Johnson was small forward for most of the year,
    though he was a shooting guard in some cases. Q spent most of his year as Phoenix’s SG.

    “The three Suns? ?stars? (I have a serious problem with calling Sawn Marion a star) are not wings in the Suns offense. The role of a PG or a C and the role of a wing are drastically different, so all I was suggesting is that maybe what the Suns? wings are asked to do is not taken into account in PER.”

    Why are you so focused on the Suns’ wing offense? That isn’t the only (or even the most important) reason that the Suns were so good on offense. Stoudemire was almost unstoppable beneath the basket, and Nash shot insanely well for a point guard, and those two used the most possessions on offense. The team didn’t turn the ball over very often. The wing offense seems secondary in comparison.

    KB, sorry this got off course so badly. I’m perfectly willing to just drop this, since it’s not really going anywhere.

  10. McDaddy11

    Every team in NBA needs a guy like Bowen, so those 20 million dollars cry-babies gotta play thier best every night, and us mortals can be happy watching good basketball

  11. KnickerBlogger Post author

    The one thing about Bowen, is that some players can give you kind of defense without the offensive blackhole. Ron Artest and Tayshaun Prince actually have a positive affect on their team’s offense. Notice how Bowen doesn’t get a lot of minutes when the other team doesn’t have a dominant scorer at the 2 or 3, because his offense is so weak.

    I think he’s a tremendous defender, and wouldn’t mind having on my team. However his value largely depends on who the opposition is.

    Wait what was this thread about? Oh Q-Rich. Hmmmm to get things back on the right track, I think we can all agree that since his defense is nowhere near the caliber of Bowen, Prince, or Artest that his value can be largely measured by PER?

    He’s only 25 years old & showed potential early on in his career. Q’s eFG has fluctuated from 43% to 50%. So which Q-Rich will we see in a Knicks uniform? Will Larry Brown be able to turn his career back around, was he overrated to begin with, or are his injuries what’s limiting his play?

  12. Carlos

    Even if Q efg% is at last year level (.498 and a 1,04 PSA) he’ll need an usage rate of 20 or higher to get a PER of 15. His numbers are a bit better than Tim Thomas, though. Rebounds a bit better, passes a bit better, shoots a bit worse. If he can improve his defense and shoots reasonably well, he can be a good replacement for Thomas.

  13. Paul M

    Q’s best years came when he was a post up player and not just a shooter. If Brown is willing to use him this way (I doubt he’ll be a big fan of Q’s streaky shooting), he might bounce back. Still hard to say.

  14. Ricky

    I’m predicting Q to achieve a PER around 15. In his first interview after the trade, he claimed to be happy about changing teams mostly because he was asked to play a one dimensional style in Phoenix and his preference is to post up more often. I have no reason to think he’s a liar at this point, so I’ll believe him unless proven wrong. So the bottom line is that we have a tall, strong 2 guard with a proven ability to score in the post (evidenced by the years with high FG%), a stated intent to do so, and a coach who hates the bombs-away approach of shooting 3s all day. Considering those facts, it’s safe to assume he’ll shoot fewer 3s and have a higher FG%. He still shoots well, rebounds above average for his position, passes well and plays with heart. Of course Q has flaws like all players, but after weighing the positive and negative, he seems to be an above average player although not a star, which could realistically result in a PER in the 15-17 range. I’ve done some quick and dirty rankings based on traditional stats (ppg, reb etc) and based on that, Q seems to be a top 100 player in the league. In other words, above average but not great.

  15. Kareem

    This is the point missed the most when PER enters the discussion and with Bowen in particular:

    Bowen does exactly what his team needs: He spreads the floor by being a reasonably good open shooter from 3, does not dominate the ball, and plays great defense. This is infinitly more valuable to a team with other players who are scorers than players with much higher PERs than Bowen. The value of the rare players who do not need to dominate the ball or to shoot much to be effective is the big hole in PER and why players like Bowen, Ben Wallace and Tayshaun Prince should be sending Hollinger back to the drawing board.

  16. Carlos

    Kareem, John knows perfectly well about this aspect of PER, but the lack of boxscore data about individual defense makes impossible to improve a linear weights system like PER on this. Some plus/minus systems (like DanVal) adress this better, but plus/minus is very “noisy” and needs the accumulation of large amounts of data to produce good results. 82games.com is planning to chart individual defense (among other things) next season, so help is on the way.

  17. Kareem

    I hear ya on defense, and plus/minus is certainly not the answer, but my bigger exception is the inability of PER to chart offense within context. To me, if you have effeciant high volume scorers on a team (as is the starting point for most championshop caliber teams), surrounding them players who are effective without having plays run for them or not even touching the ball much at all (and therefore keeping the ball in the hands of the stars) is at a premium. These are the hidden gems of the NBA, and can only come through observation for the time being. I think PER is fun and can be usefull in a limited way, but it’s not a major tool I would use to put a team together.

  18. Ted

    Not to dwell, but I personally don’t have any doubt whether Bowen’s defensive contribution makes up for his offensive shortcomings (my opinion). Although of course it depends on his opponent, the same can be said for just about any player in the league. But how many teams around the league don’t have a scoring threat that Bowen can nutrilize?

    There are better all-around players, no doubt, but he was 20th in the league in 3 point % last year. If you leave him alone he hangs out in the corner and I’ve seen him knock down several threes in a row and carry the Spurs offense for stretches. So not only is he a defensive specialist, but he could also be considered a three point specialist. Bowen also comes at a cheap price and actually took a pay cut when he probably deserved an increase. Teams outside of NY (or any smart team) cannot afford, and realize that they do not need, to pay 10 mill a year for role players or to put all-around stars at every spot in their rotation.

    This certainly did get off track. The reason I brought up Bowen was to question what PER really tells you in terms of whether Q will help us or not. Of course Q will not have the kind of defensive impact that Bowen does and thus if his PER is under 10 it is fair to say he had a terrible season, if it is at 14 it might not. This year he had a low PER but he did what he was asked to do and helped his team win a lot of games. If he had been asked to do something different they might have won more games, but he wasn’t.

    The reason I don’t know how much Q’s PER will effect the Knicks is because I don’t know where he fits in on this club. (If it’s where he fit in on the Suns, not saying it is, a PER of 14 might be fine.) Is he the second best player behind Marbury? Is he going to come in off the bench behind Crawford, TT, and maybe even Houston or Ariza? Somewhere in between?

    The Knicks are really at an exciting point because there is a lot of talent, but only time will tell which players, if any, will step up and fullfill their potential. You can lump Q in with a bunch of young players (and Jerome James) who have talent but still have a lot of room to improve.

    Of course, Q has proven more than any of those players at this point and did just make his first trip into the playoffs, so maybe he is now more aware of what it takes to win and is going to have a very productive year as the Knicks’ second option and therefore have a good PER. I definitely expect it to improve from where it was last year.

    The thing is that if he doesn’t step up and is nothing more than a shooter off the bench with a PER of 14 there are seven other young guys who might step up along with several veterans who have proven themselves to be solid role players and we do have a pretty decent point guard. Basically, if Crawford and Ariza are both outstanding at the 2 amd 3, Q is suddenly not so important. Or if Sweets and Frye are good inside and Crawford and Ariza are decent…or countless other scenerios…. Along with Bowen I also pointed out that Brent Barry (a key reserve for the champions) had a PER just over 14.

    So while I think PER is a very useful statistic I think we can all agree that it is not perfect and that is must be considered a means to an end, not an end in itself. That’s all I’ve been saying.

  19. KnickerBlogger Post author

    I agree that PER is a poor way to build a team, and falls prey to Dean Oliver’s “no holy grail” argument.

    But here is something interesting. Despite being a flawed stat it seems to stay within a player’s reasonable career curve. To use baseball as an analogy, you don’t see the wild swings in PER year to year like you do in batting average (or even ERA). If you look at the PER leaders from last year (and yes I know my PER calculations are off), there aren’t any outliers. Sure Amare, LeBron, and Wade are in the leaders for the first time, but that’s not unexpected considering where they are in their career.

    I’ll say that PER ignores anywhere from 50%-99% of defensive value. However it’s a quick & very dirty tool if you’re looking at someone’s non-defensive value.

    I want to seperate the two issues that we’re discussing. Bowen, Iguodala, Prince, Artest are all underrated because PER doesn’t acknowledge their defensive strengths. But it should pick up their offensive contribution well enough. If Bruce Bowen has a 9.5 PER, then I would agree that he sucks on offense. Being told to stand in the corner and only shoot if you’re wide open should verify this.

  20. Kareem

    No, you’re still missing the point. Bowen is an effective offensive player much the same way Fred Hoiberg or Prince are, and why players with higher PERs like Jerry Stackhouse and Antoine Walker are ineffective. The idea that the first group’s PER is hurt by the fact they won’t take the asinine shots the second group does is crazy. It’s not as if when Bowen doesn’t shoot the Spurs lose a posession… all that happens is Duncan or someone who is a truely effective offensive player gets a look.

    That’s the part of PER that is really screwed up. The idea that low percentage players are rewarded for jacking up shots is beyond comprehension.

    Or, in other words, Bowen going 2 for 5 is prefered to a Stackhouse 8 for 20 any day of the week.

  21. KnickerBlogger Post author

    There is a difference between Hoiberg & Bowen, as shown by the 4 point difference in their PERs. Bowen’s eFG% of 48.3% is just about league average, so he’s not giving you anything special. Additionally he barely gets to the line (1.00 PSA). Compare that to Hoiberg who shoots 51.4% with a 1.15 PSA. Also note that Fred is a superior rebounder(7 to 4.9 REB/48) and passer (4.1 to 2.3 AST/48), and it becomes clear why he’s rated higher.

    As for myself I’ll take 8 for 20, over 2 for 5 any day. And I’ll win 16 to 4. ;-)

    Seriously though, there is something to be said for usage. Give Bruce Bowen 20 shots, and what will his percentage be? If Bowen were on a less talented offensive team, would his inability to creating scoring opportunities for himself & his teammates still be seen as a positive?

  22. Carlos

    Kareem, PER being just one number, it doesn’t tell you much about a player’s game or how does he fits into a team. That’s what usage rate, PSA, etc. are for. Regarding Bowen offensive value as measured by PER, Dean Oliver “skill curves” concept is useful here. He says that as a general rule, a player efficiency will go down when his usage rate goes up. And that’s the difference between Hoiberg and Bowen offensively. Hoiberg compensates his low usage rate with a high efficiency and Bowen doesn’t. Bowen is effective enough on offense for the Spurs, but a player who only shoots when he is wide open should have a higher efg than 48.3%. Also, high usage players have value. Remember, the reason Hoiberg-like guys can afford to only take good shots is because somebody else is taking the other shots.

  23. Kareem

    When did we start comparing Hoiberg to Bowen? This makes no sense. They’re the same kind of offensive player, Hoiberg is just a better version.

    The point, again, is that low usage/high efficiancy players are much more valuable than is accounted for on teams with outstanding offensice players.

    Or, to make it even more clear, you want the best scorers on your team taking the most shots. When less talented offensive players take shots, it takes the ball out of the hands of the better guys. There is no reason to reward this as PER does and is part of the reason it needs reworked if people are going to use it as a meaningful stat.

  24. Carlos

    But Bowen is a low usage/AVERAGE efficiency player!! That’s the reason his PER is so low!! I completely agree that low usage/high efficiency players are very valuable; what I don’t agree is that PER does not account well enough for this value and Hoiberg’s PER is an example.

  25. Kareem

    Yes, Bowen is an average efficiency player. That is why the fact that he is low usage is to his credit. The real problem are players who are average/low efficiency and average to high usage.

    Let’s take the case of Hoiberg and Latrell Sprewell.

    In 03-04 Sprewell’s PER was 14.63 but his PSA was .99 with a USG of 22.6

    Hoiberg on the other hand, Hoiberg’s PER was 13.61 and his PSA was 1.22 with a UsG of 12.5

    So, I’ll state again… with Garnett being a far better offensive player than Sprewell or Hoiberg, why would you want either of them to have a high usage rate? Shouldn’t Hoiberg be credited for not taking posessions away from Garnett AND when he does, making the best of them. This, of course, applies to Bowen’s selective shooting as well.

    The worst players in the game are high usage, low efficiancy players and teams with guys like this have terrible offenses. PER does not at all reflect this as well and is a deceptive stat for that very reason.

  26. Ted

    The Spurs are clearly an exceptional example because they have TD. However, they could have surrounded him with whoever they wanted. They got penatrating playmaking guards, a solid bigman, and 4 guys who, among other things, are all deadeye shooters. So if you try and double up on TD or help out when Parker or Manu blows by his man, you’ll get burned.

    If the question is whether it’s more valuable to have a player like Stackhouse take 12 shots a night and shoot .414 (.267 on 2.1 three point attempts) or Bowen take 7 shots a game and shoot .420 (.403 on 3.1 3 point attempts) or Hoiberg taking 3.7 shots and hitting .489% (.9 of 1.9, .483, from downtown). Considering what Bowen brings to the table defensively and that all three have plenty of other scorers around them I don’t think that this is much of a question.

    I guess a better question is who’s more valuable to a lesser team.

  27. Kevin Pelton

    I don’t think you can ignore the pressure it puts on other players to have a low-usage player in the lineup.

    If opponents can double-team off of the low-usage player, it hurts the efficiency of players like Garnett and Duncan. If a slightly less efficient Bowen using more possessions makes all his teammates slightly more efficient, that’s a good thing for the Spurs offense.

    Now Bowen is slightly different in this regard because his one offensive skill, 3-point shooting, makes it harder for defenders to double off of him than, say, Chris Duhon.

    Teams with low-efficiency, high usage players certainly aren’t always bad offensively. After all, Minnesota with Sprewell and Dallas with Walker were two of the NBA’s best offenses two years ago, while Chris Webber was never a model of efficiency for the high-octane Kings.

  28. Kareem

    To me, it’s irrelevant as to how valuable these guys are to lesser teams. The goal isn’t to put together lottery squads where players like Stackhouse are top options or Bowen gets 12 shots a night. This is why any smart GM sees the huge value in players who bring something to the table besides shooting a lot.

    Or, to answer your question another way, Bowen, Iguodala, and Prince are just as valuable to a lottery squad because at least they are one piece to the championship puzzle their team should be trying to put together. High usage/Low Percentage players only impede these lesser teams progress and congest cap space.

    In other words, any of the top teams can use defensive minded players who do not shoot a lot and are reasonably high percentage players, yet no one needs Cybertwan, Jalen Rose, Stackhouse or other high “Usage” players hijacking their offense and getting torched on D. Really, this is very much what seperates the good teams from the bad ones in the NBA.

    The smart GMs know there are SO many “scorers” who can get their shot ineffeciantly in the NBA, CBA, USBL, and in leagues around the world that guys like Sprewell are virtually worthless offensivly.

    I don’t see how this even possibly confuses a guy as smart as Hollinger is supposed to be. If a player has a lower PSA than his teammates around him, why would any team want this guy shooting?

    – On the issue of alieviating offensive pressue, although I see your point about how low usage players can put stress on an offense I don’t see how guys who are not efficient can help. It’s not like by virtue of Sprewell jacking up bad shots that Garnett is going to get double teamed any less.

    Any low usage player who can handle themselves on defense, has a good PSA and can space the floor is very valuable to a team with excellent scorers, and, again, this is not reflected in PER because of the emphasis on point production and usage.

    It doesn’t make Bruce Bowen better than Tracy McGrady, or even an all-star, but it sure makes him a hell of a lot better than a great majority of the players Hollinger has ranked ahead of him.

  29. Kevin Pelton

    Kareem, you bring up one of my latest theories. I think that whether a player is any good is less important to his teammates than whether he’s considered good by opponents.

    In other words, you’re not the coach, someone else is. And they probably don’t know or care about Latrell Sprewell’s True Shooting Percentage. As a result, they don’t double off of him the same way they do Trenton Hassell. And that’s what matters for the defensive coverage Kevin Garnett faces. *shurg*

    There’s a great 82games column in here, but I’m not sure I want to give front offices any extra insight. …

  30. Eddie

    Kareem, I agree with you completely that low usage, high efficincy shooters who can defend a little are more useful than their PER would indicate if they’re surrounded by quality scorers (as an example Steve Kerr and John Paxson on the Jordan Bulls were much more valuable than their PER), but I disagree with you up to a point about high usage, low efficiency players. These guys do have their uses. The most prominent of these players are probably Allen Iverson, Antoine Walker and Jerry Stackhouse. Carlos mentioned Dean Oliver’s “skill curves” and how generally efficiency goes down as usage goes up (seems logical) but these guys curves aren’t like that. They’re just as (in)efficient taking 8 shots a game as they are taking 25. However, their ability to carry a larger than average load without their efficiency collapsing allows the more “normal” players the luxury of picking and choosing their spots therefore increasing their efficiency. Players like this are definitely not perfect and can only be effective in the right situations like Iverson had on the 2001 Sixers when they surrounded him with hardass defenders who could rebound an above average percentage of Iverson’s misses or do things on offense that Iverson couldn’t (Todd MacCulloch scoring in the post for instance). In other situations though, mainly ones where there are other good scorers on hand, these guys can either be useless or counterproductive. For example Walker in Dallas in 2004 (I shudder to think how amazingly great that offense could have been without him) and Iverson and Stackhouse fighting each other for control on the mid 90′s Sixers. If I have a point here, and I may not, it’s that GM’s really need to think about how the pieces on a team are going to fit together and not just always get the “best” overall player, which is probably where you were coming from too. Oh well. By the way, as I am writing this, I just hear that the Heat (currently possessing two high usage players in Shaq and Wade) have just traded for Antoine Walker. Irony. Sorry for the long post.

  31. D.J.

    I am that rarest of rare commodities, a Clippers fan. I followed Q’s career closely with the Clips. When he and D Miles played together during their first two seasons, they fed off of each other’s energy and created an excitement and enthusiasm the franchise had never seen previously and has never seen since. Q is excellent in transition, rebounds well, and spots up pretty well. His post game, however, is his biggest strength, and one that was completely underutilized last year in PHX. Q is a modern day Dantley or Aguirre on the block.

    On the flipside, his ball handling is awful, and on defense he appears to have two left feet. I love his energy and enthusiasm, but the Clippers made the right move in letting him to go PHX, and PHX obviously was anxious to jettison him after only one season. Maybe his style will fit better with the Knicks, although the question always becomes one of enough basketballs (and shots) to go around. He could end up as the go-to guy, or he could end up pissed off at Steph because he never sees the rock. In any event, his defense will always limit him to being a good player on a bad team.

  32. D.J.

    PS: I would consider Q to be the antithesis of Bruce Bowen. Hopefully the Bowen discussion only arose in terms of PER and how it is calculated, not in an effort to compare the two players.

  33. Kareem

    Although I agree that players don’t always know who to respect on the court, through observation I think it’s pretty obvious the Stackhouses, Sprewells an Ron Mercers of the world do not take any pressure off of their teams top players when they’re in the game. Last I checked, Kevin Garnett still had 3 guys running at him last playoffs and the Maverick’s offense ran no more productivly when Stack was paired with Nowitzki.

    For that matter, how often has a high usage/low efficiancy player been a featured scoring option on a championship team? How about a second option? Or third. If the answer to this question is “once” or “never” then well…

    Don’t you think that whatever pressure Sprewell takes off of Garnett (assuming he even does, which I think is a false premise) that it’s outweighed by the posessions Garnett is losing due to Spree’s poor shot selection? Wouldn’t Garnett, Cassell and Szerbiak be capable of creating better scoring opportunities for the Wolves if they had the use of Spree’s wasted possesions?

    Also, to clarify, this is NOT the DH vs. a bad defensive shortstop argument. It’s the player with a low OPS who gets a ton of RBIs vs. a high OPS/low RBI player argument. Players who shoot a lot and are inefficiant are taking posesions away from their team much the way a player in baseball who doesn’t get on base much is taking outs away. In both cases, they are hindering their teams offense by generating superficially impressive personal stats at the expense of the team.

    Look, it’s one thing to construct a stat to give a reasonable idea of who the best guys are, and I think PER more or less does that. The problem is, we all already know who the best players are in the league. PER’s greatest failing is in secondary and role players and how they fit into a winning team. Without being able to properly acount for defense or shot selection (amongst other things), what the hell is the point of even using it in an educated discussion?

  34. KnickerBlogger Post author

    The difference between the low OPS/high OPS guys is that they’ll continue their trend both given the same ammount of plate appearances. The same just isn’t true in basketball. If you gave Fred Hoiberg 10 more shots, his eFG%/PSA would plummet.

    Look at Big Ben.

    fG% PSA
    .518 0.99
    .578 1.08
    .503 1.02
    .490 0.94
    .531 1.04
    .481 0.97
    .421 0.88
    .453 0.92

    On al accounts for the first 6 years of his career he was an efficient scorer. But what happened to make his FG% drop about 60 points? He went from shooting about 7.5 shots a game to almost 12.

    As for the baseball analogy, here’s a better one. Bowen and the like are lefty mashers. They can hit left handed pitchers, and if their manager only uses them thusly, they’ll put up good numbers. Next to guys that play everyday their numbers will look superior. However if you give them 600 ABs, their averages will plummet.

    Sure they have value, but I’d agree that on some teams guys with high usage rates are needed. Not every team has a Shaq or Garnett to take an ishload of shots at a high rate. But I’d also say that many of these players are overrated (like Riley calling Walker one of the best players in the league).

  35. Kareem

    The latest baseball analogy is more apt, however, I think we’re pretty much at a philosophical stalemate on this one. I think it’s pretty clear there are any number of players who are high efficiancy who could be high usage simply by jacking up bad shots. So let me ask you this…

    What championship teams, or even top offenses, have benefited from high usage/low efficiancy players?

    I can’t think of any players who are high usage/low efficiancy I would consider effective players or who have benefited their teams offenses or have been a top option on a championship team and if you look at the top offensive teams, there’s a noticable absensce of low efficiancy players.

  36. Simon Ramirez

    Depends on what you mean by “benefitted”.

    Derek Fisher was the 3rd leading scorer after Shaq on the Lakers championship teams of 2001 and 2002 with an efficiency of only 13 or so.

    Likewise, in 1999, Sean Elliot was the third leading scorer for the Spurs with an efficiency rating of only 10.

    The efficiency rating seems to directly spiral downwards when you increase the efficiency of those at the top of the order.

    Shaq and Kobe, as well as Robinson and Duncan had players with otherwise poor efficiency ratings around them. Although their stats may have been inflated because of this, the efficiency wasn’t.

    We all know that it is still dominant big men that win championships, unless your team has Michael Jordan on it. Because of that, the efficiency vs. championships debate is always skewed, because big men will almost always have higher efficiency ratings than guards, unless that big man is Antoine Walker.

  37. Kareem

    And you think that Fisher and Elliot were high usage? I don’t get it.

    Aren’t these classic examples of players of mediocre offensive players along the lines of Bowen playing a low usage/conservative offensive game for the best of the team and letting the stars score?

  38. Eddie

    What about Isiah Thomas? During the Pistons run of title contendership from 1987 to 1991 he had PSA’s of 1.06, 1.04, 1.06. 1.00 and 1.01 which were poor considering the high offense era he was playing in to go with some pretty high usage rates yet the Pistons won by surrounding him with outstanding role players (Rodman, Laimbeer, Dumars etc) who could defend, rebound and score more efficiently. Asides from Vinnie Johnson, Isaiah always was the least efficient scorer on those teams but he took the most shots.

  39. Kareem

    I’ll agree with Isiah being high usage/low efficiancy. That’s one; and I might add, getting him to curtail his shooting was part of what pushed them over the top. Also, that was a great, great, defensive team so I don’t know how much we can tie their success to their offense and Isiah.

    Now, correct me if I’m wrong, but under Hollinger’s method, Isiah, who as we just established was not particuarly efficiant and also who had to shoot less for his team to succeed, would have a higher PER if he shot more on those teams?

  40. Eddie

    Kareem, I’m not sure that Isaiah curtailing his shooting was necessarily key to the Pistons championships. Going by http://www.basketball-reference.com his per 48 minute field goal attempts were pretty consistent from 1985 to 1990 with a high of 22.0 and a low of 20.1. Yes, they did win with defense, but their offensive efficiency ranks in their championship seasons were 7th and 11th which aren’t bad. As for your question about PER I’m not sure how its calculated but Isaiah’s highest PER was in 1985 which was not due to shooting efficiency but a great assist rate (17.5 per 48 minutes). I’m pretty sure, however, that his PER would only have been higher if he’d actually make a reasonable percentage of the extra shots he’d be taking.

  41. Kareem

    My first question would be this… What proof do we have that Isiah having a high “usage” helped the Pistons if his other teamates were more efficiant? Besides the fact that this lukewarm example is the best we could come up with of a high usage, low efficiancy player on a championship squad or top offensem how are we determining that this was a benefit to his team?

    This sounds like some stat mumbo jumbo to me. I mean, we’re all operating under the assumption that high usage even with mid to low efficiancy is good… even with very little evidance. Then how, may I ask did we come to this conclusion? I mean, if Isiah is the best example, it seems like a flimsy premise at best for EVERYONE to be buying in to.

  42. Eddie

    My answer to your first question would be that Isaiah taking so many shots allowed his teammates to be more efficient as they were better able to pick their spots. Second I haven’t really been trying to say that these guys are good per se, just that they can have their uses and aren’t as bad as you seem to think. If I was trying to build a team from scratch I would definitely not build it around a player like this. I’m just saying that if you do have a player like this on your team there are ways to use it to your advantage. I also agree with you that these guys are almost always overrated. Isaiah was not one of the best players of the ’80′s, Antoine Walker should not be a multi time allstar and AI is not one of the very best players in the NBA. Coming full circle, I’d say that PER sums up all these players worth quite nicely.

  43. KnickerBlogger Post author

    There seems to be two opinions on this. The first is that any player with low efficiency has a negative effect on a team’s offense. The second is that a player with a high usage takes the burden off his teammates that are unable to create on their own, keeping them at a high efficiency.

    Until someone does a study on it, and proves it once and for all, I think we’ll be arguing until the cows come home.

  44. Kareem

    We are certainly arguing in circles at this point.

    However, if usage is consistent through a players career, yet is valuable only within certain context, how can it be factored into PER as a universal positive?

    Are you telling me that Antoine Walker, a guy who shot 42% from the field, and 55% from the line while being the 4th best option on one of the best offenses of all time should be rewarded for having the 9th highest usage in the league? Who was he taking the “offensive burden” off of? Nash? Nowitzki? Jamison? Finley? Daniels? Are you fucking kidding me with this nonsense? Did Sprewell take a burden off of Garnett? Or was it Cassell? Or Szerbiak?

    There is NOTHING to show the benefit of a low efficiancy/high usage player when constructing a good offense. In fact, in his own book, Hollinger even says of usage “Among other things, Usage Rate allows us to identify who is shooting the ball too much…” NOW WHY REWARD A PLAYER FOR SHOOTING TOO MUCH?!?! ARGH!!! THIS IS INSANE WE EVEN HAVE TO DISCUSS THIS!!!

    Usage is factored in by Hollinger to have a stat that passes the giggle test. A stat to make sure guys we ALL know are over-rated but still GOOD NBA players like Jalen Rose or Jerry Stackhouse are not rated behind guys we ALL know are marginal NBA players like Fred Hoiberg, Bruce Bowen or Mike James. If Hollinger came up with a stat like that we would ALL know how silly that would look, right guys? A lot of people who are leery of the stat nerds, like for one, the good folks at ESPN, would laugh a book based on a new statistic like that out of the room right?

    I guess the part that is more disapointing to me isn’t that Hollinger wants an over-simplified stat that looks good but has serious, serious, flaws, but that a lot of you just ate the shit burger up without even questioning it.

    Or, in the words of our own KnickBlogger

    “Until someone does a study on it, and proves it once and for all, I think we?ll be arguing until the cows come home.”

    Exactly. And as opposed to using a catch all stat that its own creator admits clearly needs some work in almost every serious discussion, to the point where people use it as a defense/invalidation of players wouldn’t it be preferable to:

    1. Study/Argue the stat further

    OR

    2. Not use the stat until these kinks are worked out.

    salem

  45. KnickerBlogger Post author

    Kareem, I’m just curious what you think about this. If the 2001 Sixers replaced Iverson with the 2001 version of Hubert Davis, would they have been better?

    ## eFG TS% Usg
    AI .45 .52 33.5
    HD .54 .54 13.8

  46. Kareem

    If what you’re trying to say is that Allen Iverson is better than Hubert Davis offensivly or otherwise I will agree. I will also say that Tim Duncan is a better basketball player than Tim Conway and that usage has value as an offensive measure if you can figure out how to use it or don’t apply it universaly as PER does.

  47. Eddie

    Kareem, I have to agree that we are definitely arguing in circles at this point so this will be my last say on this matter. I have not been trying to say that these guys are ideal players, just that they have their uses in the right situations. I apologise if I have not been clear on this. I specifically mentioned in an earlier post that Walker was counterproductive on the 2004 Mavericks because there were many better offensive options available and he was taking shots off them.

  48. Kareem

    Eddie:

    We are now getting somewhere. I agree with you that these players have their uses in the “right situations.” I can’t see anyone making the argument that some players high usage is anything more than chuckers taking a bunch of bad shots. Hollinger even says as much.

    So how can usage be applied as a universal positive in PER? By that same train of thinking, why should we use PER if it’s fundamentaly flawed? Lastly, do you not think it’s weird that no one else has picked up on this or questioned it?

    KAJ

  49. KnickerBlogger Post author

    Kareem, what stat should I use to replace PER? Everytime I mention a player, maybe I should use his eFG, TS%, and a handful of per minute stat adjusted for the team pace (and if I’m going across years league pace)? Such as …. “The Grizzlies want Damon Stoudamire (.470 .510 9.9 22.2 0.0 0.2 6.2 7.6 4.3 9.0 13.2 2.6 1.0 0.5 3.4 3.4 26.1) …”

    In baseball, people use OPS all the time, which is also a fundamentally flawed stat. But when you hear someone has a 700 OPS, it give you an idea what kind of value you’re talking about. Additionally, I don’t think usage is such a large part of PER to totally invalidate it. Sorting the stat page by usage, doesn’t show a huge bias.

  50. Kareem

    I don’t know, what did you do before PER came into your life? It’s not a personal attack against you for using it, but I really don’t think perpetuating a flawed, catch all stat without examining it further is wise for intelligent discussion. I mean, I’ve literally seen people using PER as a way of supporting an argument their making. Like “What are you talking about? X has a PER 3 points higher than Y???”

    What about… “The Grizzlies want Damon Stoudamire.” Or… “How do you think Quentin Richardson will perform for the Knicks this year?”

    As far as sorting the stat page by usage not having a bias… maybe I’m misunderstanding the way you’re using bias, but it has Lee Nailon ahead of Steve Nash and Ben Wallace.

    Although I think OPS is frequently misused, at the very least it’s just adding two stats together that make perfect sense: On Base PLUS Slugging. Regardless of whether people think On Base% is more valuable or not, it’s at least a stat that makes perfect sense. Combining bad shooting with volume shooting as a positive makes no sense, and that’s why PER gives you twisted results. And if you look closely at PER you’ll see other major problems with it beyond the fact it doesn’t account for defense and what was previously stated.

  51. Carlos

    Man, this is still going on ???. Why can’t we agree on something here like … A) PER has flaws. B) High usage/low efficiency and low usage/high efficiency players are hard to evaluate properly because their usefulness is dependent on context. C) We should try to build a better tool than PER. Is it that hard? Kareem, I personally use the PER to get a quick and dirty idea of the productivity of a player as compared to others or to himself. Most of the time it does its job reasonably well. Personally, I’m not very interested in building a better PER because a single number simply can’t contain all the relevant information. It will always be imperfect and its current imperfections are not very important to me. If you want, you can come to the Apbrmetrics board and check the discussions on the subject.

  52. Kareem

    Carlos, I agree with A,B (somewhat), and C. However, I think it’s not a quick and dirty tool, but an overly simplistic number that summarizes players in a misleading way. The more we use it, the more we are ignoring defense and shot selection… which are two of the most important attributes a player can have.

    KAJER

  53. Kareem

    KB,

    You know, there’s nothing wrong with talking through an issue… I can tell the whole thing makes you uncomfortable, but I mean, when we started this, there were people (not naming names) that were claiming Usage is valuable without even looking at context. The more everyone can understand the stats they’re just spewing out as fact, the more educated a conversation you can have.

    Super Nintendo Chalmers

  54. James

    Quentin Richardson disappeared in the playoffs.

    I saw the playoffs, and he didn’t exactly excel. But did he disappear? I happen to bump into his stats page recently, and thought this interesting:

    b Fact: His scoring average dropped from his season average of 15ppg to 12ppg.

    This is the most common knock against Richardson’s playoff performance. But get this, Richardson’s fg% and 3pt% actually ROSE. His ft% dropped 10 percentage points, accounting for over half a point per game. He also stopped being a totally ‘jacker’ by cutting his 3pt attempts by almost 3 (!). Right there is the drop-off in his points production. Not only that but Amare Stoudamire, Steve Nash, and Jim Jackson were taking significantly more shots. Richardson score less by shooting less.

    So superstars are supposed to improve their personal stats in the playoffs? Yes, but even in the regular season, Richardson was only the FIFTH scorer on the team (on a team with an inflated scoring average). Judging his production in terms of scoring average is spurious, particularly when his average dropped by only 3ppg which is well within error.

    More to the point, Richardson’s scoring and rebounding averages dropped in March and April, following and during some mild injuries. His playoff averages were actually HIGHER than his March/April numbers. His Nov-Feb numbers were also before Jim Jackson joined and took 8ppg. His numbers were much higher (very good) against Memphis in the first round.

    b Fact 2: his rpg dropped by 1 in the playoffs.

    The total suns rebounds dropped by 3 rpgs, and Stoudamires rose. Richardsons OFFENSIVE rebounds (which aren’t padded by gimme defensive rebounds only contested by your teammates) actually ROSE.

    Ah, enough for now. Point is that Richardson wasn’t exactly the star on his team, and using his stats against his playoff performance isn’t necessarily fair.

  55. Jesse

    Tried to read all the replies but it’s a lot to process. I have to go with the argument that Richardson was in a position where he was asked to take as many reasonably open 3′s as possible. Given this situation, his FG % isn’t as bad as it seems. The point of this was obviously to get other players open, because while his % wasn’t great, he took so many shots that he remained a real scoring threat.

    I think the Knicks will use him very differently. Brown is not a coach that likes players to chuck. Look how few 3′s the Pistons took last year. Given this, I think his FG % will go up, and therefore his PER.

    Richardson is a good player, above average, but not great. The question is whether we need him to be great, or consistent.

  56. Pingback: Damon Stoudamire

  57. Pingback: Retin-a

Comments are closed.