After reading the Landy for CV trade rumors it had me wondering where Scolandry actually rank when it comes to bang for your buck, which basically is the amount you pay for the production you get. Basically what I did was get each teams PER from their PF's using 82games.com, the 2 main players at the PF position using the depth chart on Yahoo.com (I realize this could be off, I put question marks for the ones I was unsure of), and the salaries of each of those players from Hoopshype.com. I ranked each team based on the PER and the salary (remember, lower is better), and averaged those together to get a final ranking. Here is how it all played out: Rank Team Players PER 08/09 Salary 1 Houston Rockets Scola/Landry 18.6 $6,142,800 2 Portland Trail Blazers Aldridge/Frye 18.4 $7,795,169 3 Utah Jazz Boozer/Millsap 21 $12,391,398 4 Milwaukee Bucks Mbah a Moute/Villanueva 16.8 $4,205,050 5 Atlanta Hawks Smith/? (Jones) 18.2 $10,797,581 6 Memphis Grizzlies Arthur/Warrick 16.2 $3,096,261 7 Washington Wizards Jamison/Songaila 20.5 $14,157,285 8 New Orleans Hornets West/? (Wright) 18.2 $11,732,100 9 Golden State Warriors Wright/? (Randolph) 16.1 $4,206,600 10 Sacramento Kings Hawes/Thompson 15.7 $4,074,600 11 Dallas Mavericks Dirk/Bass 23.2 $18,904,713 12 Phoenix Suns Amare/? (Amundsom) 19.6 $15,868,131 13 Miami Heat Haslem/Beasley 17.6 $10,889,960 14 Cleveland Cavaliers Wallace/Varejao 22.5 $20,284,480 15 Detroit Pistons Prince/McDyess 16.4 $10,495,794 16 Philadelphia 76ers Young/Evans 15.6 $6,608,120 17 Oklahoma City Thunder Green/Joe Smith 15.6 $8,082,640 18 Boston Celtics Garnett/Powe 19.8 $25,547,581 19 New Jersey Nets Yi/Anderson 12 $4,211,280 20 Los Angeles Clippers Randolph/? (Skinner) 16.8 $15,928,941 21 Indiana Pacers Murphy/Foster 16.9 $16,301,984 22 San Antonio Spurs Duncan/? (Bonner) 17.8 $23,557,893 23 Minnesota Timberwolves Jefferson/Smith 15.9 $13,300,000 24 Charlotte Bobcats Diaw/May 15.4 $11,661,026 25 Chicago Bulls Gooden/Thomas 15 $10,849,880 26 Los Angeles Lakers Gasol/Odom 17.8 $26,480,312 27 Orlando Magic Rashard/Battie 17.5 $22,984,000 28 Toronto Raptors Bosh/? (Humphries) 16.2 $17,610,581 29 New York Knicks Harrington/Thomas 15.1 $15,275,650 30 Denver Nuggets Martin/Kleiza 14.8 $16,006,311 It could definitely be a little off because teams that are not paying much for their PF's are pushed up the list due to the low pay. It's nice seeing the Rockets in 1st, and by a healthy margin (their average rank is 6th, the next best is 7.5, the worst is 25). After looking at this it definitely makes me shy away from the trade, even though we do get a backup center out of it. I'm not going to say this absolutely proves the Rockets have the best PF tandem in the NBA, or it is 100% accurate (many things can change, like Kaman or Camby taking over for Skinner at the PF position, that can alter this), but it sure makes you feel good about what we do have. Pugs
I like both Scola and Landry. There is no doubt that they are both a great value. However, basketball is all about match-ups and when you take out one the other is basically giving you the same game. There is redundancy. Both of them have a good offensive game that requires them to be close to the basket. Since Yao is more efficient when he is closer to the basket, it can get pretty packed in the paint. It would be nice if we could get a power forward that can shoot outside. Scola and Landry give you similar defense as well. If one of them is getting torched that night, the other one will usually have the same experience. You need to have a change of pace between the two. I hope we get a PF that is tall and can shoot outside, which fits CV.
I took the PER and ranked it, so the team with the best PER got a 1, the team with the worst got a 30. I then took the Salaries and ranked them, the lowest salary got a 1 and the highest got a 30. The Rockets were 7th overall in PER and 5th overall in Salary, so their final value was 6th. After getting a final value for each team, I ranked them, the lower your final value the better. The Rockets were the lowest, the second lowest was Portland at 7.5. I thought about doing Salary/PER to get a value on the money spent, but when I did that it was pretty much in order by the amount of money spent, the less a team spent the higher the PER value. Pugs
Per basiclly only measures offense. Just for offense alone, yes. But if defense is also considered, No way Rockets PF Tandem is the best in NBA.
doing PER for a team position is silly. Why is 82games doing that? PER is an individual stat meant to normalize all individuals around 40mpg. A team stat should just use straight up numbers or normalize to a full 48mpg. I mean, as far as I know, Hollinger does not do a PER for team stats. Not here for sure: http://insider.espn.go.com/nba/holl...//insider.espn.go.com/nba/hollinger/teamstats plus, apparently 82games is putting a lot of Duncan's stats at the center position, because 17.7 PER for the Spurs at PF is odd considering Duncan is a 24.6 and Bonner is a 17.4. Did you really want to compare the Spurs that way?
PER is not normalizing to 40 minutes/game. It's a per-minute stat, period. It's no more a per 40min stat as it is per 35min, or per 25min, or 1min, or per 48min, etc. 82games is just taking the PER formula for assigning credit/debit to various box score stats, and applying it for each player-position over all the team minutes. They could just as easily have used the more simplistic NBA EFF formula for assigning credit/debit to the stats. They'd probably get similar results. It makes sense to me.
When healthy - Boozer and Milsap It's tough to judge though as we don't see the other PF tandems nightly like we see Scola and Landry.
This year due to Boozer injury the Duo is scoring only 16.5 PPG + 9.76RPG + 32.2 Min. Houston Duo 19.9PPG + 12.8RPG + 48.4MIN.
I'm not familiar with Yahoo's depth charts, but I would double check that they correspond to the positional assignments used at 82games. For example, according to Yahoo Tim Duncan is the Spurs starting PF, and Bonner is their center. So, they have Duncan/Tolliver for their two PFs. 82games shows, correctly, that Bonner gets 42% of minutes at PF, Duncan gets 18%, and Thomas gets 17%. Tolliver hardly even plays.
durvasa, ok. understand. that said, shouldn't the 82games.com minutes for each position be exactly 48, (with a fraction more for overtime.) What's the point of running any team stat through a per minute calculation?
All teams don't play exactly the same number of minutes, or even the same minutes per game (due to overtime periods). So, they add up all the stats in accordance to the PER calculation for each position over all the team minutes and games, they divide by total team minutes played, and they adjust for pace. Then they normalize around 15. Here's the PER calculation steps, according to B-R.com. Instead of doing it for all individual players, they are calculating PER for 5 virtual "players" on each team, representing the composite production of every player at each position. These virtual players play every minute of every game. So, all 5 virtual players on a given team play identical minutes, but across different teams the minutes can vary a bit. [rquoter] All calculations begin with what I am calling unadjusted PER (uPER). The formula is: uPER = (1 / MP) * [ 3P + (2/3) * AST + (2 - factor * (team_AST / team_FG)) * FG + (FT *0.5 * (1 + (1 - (team_AST / team_FG)) + (2/3) * (team_AST / team_FG))) - VOP * TOV - VOP * DRB% * (FGA - FG) - VOP * 0.44 * (0.44 + (0.56 * DRB%)) * (FTA - FT) + VOP * (1 - DRB%) * (TRB - ORB) + VOP * DRB% * ORB + VOP * STL + VOP * DRB% * BLK - PF * ((lg_FT / lg_PF) - 0.44 * (lg_FTA / lg_PF) * VOP) ] Most of the terms in the formula above should be clear, but let me define the less obvious ones: factor = (2 / 3) - (0.5 * (lg_AST / lg_FG)) / (2 * (lg_FG / lg_FT)) VOP = lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA) DRB% = (lg_TRB - lg_ORB) / lg_TRB After uPER is calculated, an adjustment must be made for the team's pace. The pace adjustment is: pace adjustment = lg_Pace / team_Pace Now the pace adjustment is made to uPER (I will call this aPER): aPER = (pace adjustment) * uPER The final step is to standardize aPER. First, calculate league average aPER (lg_aPER) using player minutes played as the weights. Then, do the following: PER = aPER * (15 / lg_aPER) [/rquoter]
I don't see where the issue is. PER essentially measures total box score production per minute, so actual number of minutes played does not matter. The minutes listed on 82games are the total number of minutes an individual plays at a particular position on average.
hey, I'm going to answer this, but know I don't really mean to make a big deal. I thought some of the PERs seemed way off, and noticed Hollinger doesn't use PER for team stats. JeopardE, I was referring to the team stats per position, not individuals. by definition, 82games.com is making each team field each position for 48 minutes, plus overtimes. My point was simply why do per minute stat on a team (just to account for OTs?) when (not counting OTs) they all play 48 minutes per position. Then durvasa pointed out that team schedules aren't the same, so some teams have more games played than other, until the end. I would agree with using PER (as much as I ever do) if 82games.com would realize the fallacy of having strict PG-SG-SF-PF-C lineups for all 48 minutes. In reality, some teams would have less than 48 minutes at certain positions, no? especially at the center position.
It's not a fallacy. Every team plays 5 players on the court. And those 5 players are assigned positions, 1 through 5. Their role in a given offense or defense may not be prototypical, but it doesn't matter. I mean, if a team played did not play a "true center", does that mean they're getting extra production out of their PF spot? Of course not. You call one of the bigs a center, and whatever he's doing is what the team is getting out of the center position.
Even the most unconventional of lineups would still have designated positions - e.g. Al Harrington playing center for the Warriors last season, or, to bring the point home, Ron Artest playing the power forward position. It's actually a good idea for 82games to do this, as you can immediately see how effective a particular player is when playing a specific position -- for example, McGrady has been much more effective at SG than at SF, and Scola has been much more effective at his natural PF position than at C.
Strict 1-2-3-4-5 roles is not reality. What's wrong with saying their are 2 SGs out there? a 1-2-2-3-4 or 1-2-2-4-5 is a common offense. What's wrong with saying their are two 4s out there, too? Especially if it forces the other team to pull their center, like Dallas made us do by pulling Mutombo out, anytime we brought him in. imo, a prime example is the fallacy involves moving the majority of Duncan's offensive stats to the center position, which made the Spurs PF PER drop way down. When in reality, the Spurs have the best PER PF in the game. imo, you're not really saying much with positional stats if the Spurs PER (17.8) for PF ranks 12th. All you'd be saying is that Duncan is not a PF. okie-dokie then.
