PER is what it is; a statistic. A stat that is clearly defined and meant to be used as a tool in combination with others. Generally, people that think it should be more or don't understand what it means waste time arguing about it. Honesty, it's just a mathemetical formula. Why the hell people think that's supposed to actually be an all-encompassing player evaluation and then complain that it falls short is beyond me.
Efficiency is far from being an all-encompassing measure of a players value. No where does PER nor its definition, explicitly or implicitly claim to measure a players total value, worth or abilities. Instead it simply measures the efficiency per minutes of individual players. This And this. /Thread
This is the problem with formulas like PER. People like Francis actually have the audacity to write this.
So tell me how it's useful. Don't just tell me what Hollinger tells you and then type /thread. That doesn't make you look as smart as you think it does.
agreed, people try to simplify something that can't be quantified so simply. PER doesn't take into account system, which offense your team runs which dictates where and how you get the ball, where your teammates, and what your options are to do what you want to do. If Pau Gasol was in a triangle offense and he wa the number one option surrounded by shooters and slashers, maybe he would lead the league in per and have a triple double threat. IF bynum gets traded to Orlando, he is the centerpiece there and will have more shots, maybe less efficiency, hgiher turnovers, defenses will make him a primary target, etc. The whole game and per changes. IT doesn't include the chess element of basketball. That is why over 50 percent of the players still refer to Kobe as the best, while PER and all those peg Lebron as the best player for years now. Heck advanced stats had him as the leading 4th point scorer when he was a cav. If you saw that you woudl think he was some clutch god, and we all know what happened not to mention his very own teammate says when we need a bucket, clutch time, we goto wade. I think we try to use PER as much as we can as fans and writers, but Gms, Coaches, Players don't think much of it in it actually determining a players capability. So yes, Bynum is a better player then dalembert, but his role, his system, etc all play parts in his statistics so it doesn't show an overall players capability to determine if he is 'better'
if we're going to argue the merits of a per minute stat, you kind of have to use per minute numbers. bynum is averaging 17/14/2 per 36 mintutes. dalembert is averaging 12/12/3 per 36 minutes. and dalembert is averaging half the turnovers (3.0 to 1.6) and scoring far more efficiently (63.5 TS% to 53.9 TS%). add it all up and i'm not sure how one would argue they haven't been similar players on a per minute basis. if someone wants to say bynum is playing better, he should either start doing more good things or fewer bad things. or more likely, just wait for dalembert to stop being far more efficient than he's ever been.
then please, explain to us how bynum has been better on a per minute basis this season. him being better last year or the year before will not suffice as an answer. nor will "you have to watch the games" suffice as an answer. break it down for me using these first 12 or so games. as far as i can tell, you've got all the answers.
Why don't you look at actual minutes played per game? Is that allowed? You really think 17/14 is similar to 12/12? When your team actually depends on you to score and defenders actually care about you, you tend to turn the ball over more. Are you really trying to make this case?
the quotes were of what Durvasa's and AstroRocket's statements not of Hollinger. Furthermore the /thread comment was in reference to those two succinctly and IMO correctly defining PER and the topic of the thread. Not sure how you think my being impressed with their statements has any relevance to how smart I am. PER is useful if you are interested in measuring a players rate of production. Its useful if someone wanted to see how efficiently a player produce taking into account, minutes played, fg/ft attempts as well makes, fg/ft% etc, rather than simply the total amount or avg sum of rebounds or points per game. If care of neither those things, then I guess it holds no value for you. PER is simply a statistic measuring efficiency. IF ppl misuse or misinterpret the data, that reflects not on the data but on the mistaken person. According to PER, Dalembert has minisculely "out produced" or rather was fractionally more efficient per minute thus far than Bynum. Bynum - 347 min 164pts 138rbs 20blk 53%fg Dalem - 254 min 83pts 87rb 21blk 58%fg comparing just these raw stats by a per min rate, its evident both end up with very similar measures. Of course Bynum is still clearly the more valuable/better player than Dalembert because he produces more, due to being able to play more minutes and at the same time not doing so at a lower rate. But PER doesn't disagree or even measure that, nor does it imply that Dalembert is the better player, it simply measures the rate of their prodcution. Thats it.
how is anything useful then? how are points or rebounds or assists useful? assuming we agree that people like to have sports arguments (the reason this site exists), then having something to base them on is kind of necessary. so far stats seem to be the most accepted thing to base them on. "i think he's better" "no i think he's better" would be kind of pointless. so we argue homeruns, rbi's, touchdowns, interceptions, rebounds, points, etc. PER is just one fairly comprehensive way of accounting for all the things we currently put down in the box score. turnovers hurt you, steals help, rebounds are good, missed shots hurt you. all with fairly straight forward ways of weighting them. then, in probably PER's best aspect, it account for pace, minutes, and league conditions, making it a great way to not only compare players on different teams with different roles, but also players across eras (the fact that the best PER's ever include wilt in the early 60's, jordan in the late 80's, and lebron from the current day speaks to it's adaptability). if people don't want to accept PER, great. i wouldn't suggest just picking up the PER rankings and saying those are absolutely the way the players should be ranked. but i would suggest it will probably get you closer than simply picking up the PPG rankings, the RPG rankings, or most any other stat. so that would seem to be a good purpose. a quick and easy way to establish the hierarchy of the league at any given moment. and as for the "you have to watch the games" crowd, i always love the insinuation that "stats people" don't watch games. as if they just found a spreadsheet full of numbers one day and said "wow, these look neat, i want to talk about these in depth, what are they?" most stats people watch plenty of basketball, i know hollinger does. it would seem hard to even get into the stats if you didn't love the game in the first place. i also love the undying faith the "you have to watch the games" crowd has in its own omniscience and impartiality. people don't watch every single game, your brain doesn't remember every single thing it sees, and you have your own biases. only liking the stats that agree with your preconceived notions and discarding those that don't doesn't really speak to a mind that wants to be open to a debate, that actually wants to think about things. when a stat disagrees with your world view, perhaps you should wonder if your world view is actually correct. if you are actually considering things from different angles. if, my heavens, you could actually have been incorrect because you simply couldn't observe every little thing going on for all players at all times. but nah, i'm sure because you watch the games you couldn't be wrong.
Anyone who says: "so far through 12 or so games, bynum has not outproduced dalembert. if people want bynum to be considered better than dalembert, then i suggest he start outproducing him." Must not be watching any games. OR looking at stats. Congrats.
it certainly is. it plays into something called Estimated Wins Added, somewhat equivalent to WAR in baseball i believe. bynum would be winning that due to playing more minutes. playing more minutes at a high level like a 20 PER is better than playing fewer minutes because your team doesn't have to give those minutes (13 more per game in bynum's case) to somebody else who sucks or at least spend money on someone who doesn't suck to take those minutes. but that wasn't the argument in this thread. it was "oh my gosh, how can a per minute stat say bynum and dalembert are about the same per minute," which isn't at all crazy based on how they have played. people might as well get pissed off at batting average every time some scrub is outhitting some star after the first month of the season. if we're going to pretend that these 12 games are going to hold up for the whole season, then do i think 17/14/2 is similar to 12/12/3 when the 12/12/3 is much more efficient? why wouldn't i? if bynum has to turn it over way more and shoot crappier to get his points, should that be considered impressive? maybe it he had to take care of the ball and be efficient, that 17 would be 13? or maybe he's just having a bad first few weeks and at season's end he'll be getting 17 on very high efficiency like usual and dalembert will not be near as efficient going forward, PER will reflect this, and everyone will be able to sleep easy knowing that PER appropriately captured the fact that when andrew bynum started actually playing like andrew bynum and samuel dalembert started actually playing like samuel dalembert, it considered them to be the miles apart that we all know they are. or maybe bynum will shoot 20% for the rest of the season and people will continue to be mystified that PER has the audacity to think that isn't great.
do you feel stats are just random numbers placed on a box score at the end of the game? because outside of point guard assists in utah, i'm almost certain they are not. they just reflect what you've done. since you've watched all the games, how have you felt about the 3 times per game that you have seen bynum turn it over. have you felt it helped the lakers when he missed the 60 free throws you saw him miss? and did you not think the 21 blocks in only 254 minutes that you have watched from dalembert were pretty good? or that the 33 shots you saw him make on only 57 attempts weren't a good use of rocket possessions? all of those things happened. they are happening better for dalembert than they pretty much ever have and in some areas are worse for bynum than they've ever been. when that happens, they are apparently pretty similar on a per minute basis. if you disagree, then i'm not sure what you think could possibly happen that would make them similar. andrew bynum is better because andrew bynum is better doesn't cut it. the reason he's better is because he usually is far more productive, which the nonrandomly generated evil stats have concurred with quite strongly. for 12 games he hasn't been. we all agree it will almost certainly not last. but we should also all agree that for 12 games, that's what has happened.
Bynum is definitely the superior player, and he's had some monster games that show it. He's a better scorer and rebounder. He's a larger, more intimidating defensive presence. He's also had some down performances this year, but overall he's played pretty good. A 20 or 21 PER is a fair assessment of what he's done, and it will I think almost certainly go up as the season continues. Dalembert, for his part, has been quiet in several games (where he hasn't played many minutes), but he's been very good in 3 or 4 (where he's played more minutes). His 58% shooting from the field and 89% shooting from the free throw line, along with his high offensive rebound rate and relatively low turnover rate, plus his high rate of steals and blocks, has given a significant boost to his PER. If he actually sustained that production over an entire season, he would be playing at a career best level and deserving of his 21 PER or whatever it is. But its much more likely that he doesn't sustain those rates, and his PER will therefore go down.
The only reason you guys are even comparing Sam Dam to Bynum is because of this stupid PER. I guarantee you that if PER did not exist and we just watched the games and checked the box scores and more importantly saw the impact that Bynum has compared to Dalambert nobody and I mean nobody would even put these guys in the same sentence. This is the problem when you have measurements like PER. People like Francis claim that he is just as productive because some formula created by an ESPN journalist. How anyone can defend such utter nonsense is beyond me.
No, the reason I'm comparing them is because you brought it up. Bynum is a better player. PER doesn't sway me to think differently. Hollinger isn't writing articles saying the Lakers should trade Bynum for Dalembert. So, why are you hung up so much on this?
The only reason I brought it up was to prove that PER has some major flaws. If that isn't evident enough i'm not sure what is. The reason why this dialogue has lasted so long is that people are actually attempting to argue that this year Bynum and dalambert have been fairly equal.
I think PER is quite valuable to compare similar players, but I don't like using it a way to compare 15 minute bench player against 30 minute starter. The bench player won't get as exhausted as the starter, hence they are often called "energy guys", "quick hustle off the bench" or whatever. I don't like stats adjusted for minutes in general. Because players on the bench are generating nothing (moral support maybe...).
