663 careers statistically ranked 1952-2011

[Statman]

This is not false, at least not demonstrably so with the information given. What that actually shows is that offenses function far more optimally than the layman may suppose. (If everyone at various usage rates the past 30 years actually settle on very similar %s) What this leads to, in a roundabout way, is that simple usage is a good predictor of actual (not just perceived, but actually "real") scoring ability. There's no doubt that Mo Cheeks, as a career 15% usage guy, was effective, he was just only really effective as a 4th (or 5th) option. Isiah Thomas, with a 25% career usage, may not have been the most effective 1st (or 2nd) option, but he still is a 1st or 2nd option.

Love to see some debate spring up again.

The quoted paragraph sums up pretty much what I was thinking about the Thomas/Cheeks comparison while reading this thread. Thomas was a higher usage guy because he was capable (in my eyes) of being a higher usage guy on a good/great team. Coaches (and I think Daly was a GREAT coach) allowed him to be because they felt it made their team better. The TEAM results seem to indicate that - even if Thomas had highish totals of TOs and lowish TS% - his TEAMS he ran the offense for were better than average in both categories (TO% & TS%) with defenses having to focus on Thomas and less on his teammates. His higher usage allowed his role playing teamates to play their roles to the best of their abilities. The proof was in the pudding.

And honestly - Cheeks had some GREAT teammates around him - superstars Malone & Erving - good players Toney, Jones, Dwkins, etc. - and won one championship. He probably was the perfect type of PG to team with Erving/Malone/Toney. Thomas had some nice teammates - but Dumars was arguably his best teammate - and he was no Moses or Dr. J. Everyone knew Thomas is what made that team go - and they won two championships. His ability to run the team and be high usage ALLOWED (in my eyes) Detroit team full of good role players to actually not suck (technically speaking) offensively. I don't know if the Sixers would have been as good with in his prime Thomas as their point instead of Cheeks - maybe not - but I am completely certain that those championship Piston teams would have not gotten a sniff of the finals with in his prime Cheeks at their helm instead of Isiah.

Look at that - I made a post without referencing any of my ratings - and people tell me I focus too much on numbers.

[Crow]

The 1982-83 Philadelphia 76ers had 3 starters over 25% usage and another at 20% in the regular season. In the playoffs the 20%er backed off to 19% and Cheeks pushed up to almost 20%.

The 1988-89 Detroit Pistons were pretty similar. Just 2 at near 25%, but 2 more over 20% and another guy at 20% off the bench in the regular season. Close to the same in the playoffs though the leading 2 guys increased their usage and 1 guy slipped back a bit under 20% to 18%. 1989-1990 was pretty similar.

It looks like champs often have 4 guys over 15 minutes per game at or over 20% usage. 7 or 8 of the last 10 champs had or nearly had that level of depth among its high usage players.

Having or nearly having 5 rotation guys at or over 20% usage made these 2 earlier champs even more diversified, rarer and maybe more similar to each other on usage distribution than different despite the PG usage difference and 3 vs 2 over 25% usage.

Among the 4 conference finalists Dallas was the only team with 4 guys over 15 minutes per game at or over 20% usage in the regular season and playoffs. The others had 3 in each regular season, but Chicago was actually a bit short of 3 in the playoffs and OKC was clearly short (not enough shots left for Harden).

Does this matter much? Based on checking 10 years of champs and this season a bit further and doing some other work before I think it probably does. With 4 guys over 20% usage you can probably keep 3 on the court almost all the time if you want to and go with 4 at times, especially at the end of games. With just 3 available there is going to be pretty significant time with just 2 out there. 2 over 20% usage seems pretty different than 3 to me in terms of threat posed to the defense, especially if it is a good defense, but maybe not everyone thinks that a 20% usage player as the 3rd guy is noticeably different in offensive impact from 18% or 16% usage players. The issue could be researched further at the level of individual ORAPM and at lineup level to add more data to the discussion.

It may be that is a bit more wiggle room in a playoff series or a single game for the average recent history usage level of the 4th or 5th option if those players are 'higher usage capable with efficiency', perhaps demonstrating that in other parts of their careers. A vet with high usage experience in such a role could step up in some circumstances and take and make more baskets when needed. If you fell short of having 4 over 20% usage, this would probably be the wise plan B.

I am sometimes critical of high usage players based on their individual stats but in the end the team performance matters more to me. If the team can make it work well enough to win a title then at least a high usage player was not an insurmountable problem and maybe they were a help or major help. In some cases it might be hard to get folks to agree exactly where they fell on that range. Their own efficiency is part of the story but the efficiency of the team with them on the court needs to be checked too. High team efficiency for all team minutes is not necessarily enough to vindicate a high usage player unless you take that case as being an example of good overall rotation management with the star carrying weaker performing offensive units he tends to play with so stronger offensive player mixes without him can excel against more subs instead of being an example of a high usage star perhaps getting too much credit.

[Mike G]

Well, I've dug up an old file called 'Isiah effect' -- I used to do these occasionally -- in which every teammate player-season with-and-without Isiah in consecutive seasons is compared for various productions and efficiencies.
There are about 70 players who played with and without Isiah, and some of them have more than one entry in the spreadsheet. Rick Mahorn, for example, has his 1985 Bullets stats compared to his 1986 Pistons'; and his '89 Pistons to his '90 Sixers rates.

Weighting by minutes in their with-Isiah seasons, I get these average per-36-minute rates and shooting%
eff% = Pts/(FGA*2 + FTA)

Code: Select all

per 36     mpg    eff%   Sco    Reb    Ast     PF    Stl    TO    Blk    T
without   26.7   .522   14.25   6.92   2.24   3.47   .90   2.09   .77   24.13
w.Isiah   24.8   .517   13.84   6.75   1.86   3.48   .75   1.81   .81   23.22

Not surprisingly, alongside a scorer/distributor of Isiah's standing, players score and assist noticeably less. Also play fewer minutes.
I am surprised that he generally didn't raise teammate shooting% from their previous/next teams' environments.
Turnovers are reduced by 13% when running with Isiah. Steals, for whatever reason, by 17%.

[bchaikin]

And honestly - Cheeks had some GREAT teammates around him - superstars Malone & Erving - good players Toney, Jones, Dawkins, etc. - and won one championship... Thomas had some nice teammates - but Dumars was arguably his best teammate - and he was no Moses or Dr. J.

you bet thomas had some nice teammates - 3 hall of famers (same as cheeks). while the 76ers with cheeks had malone, erving, and barkley, the pistons with thomas has dumars, adrian dantley, and dennis rodman...

dantley was a great offensive player. prior to coming to the pistons in 86-87, for the utah jazz from 79-80 to 85-86 he had the highest scoring average in the league at 29.6 pts/g. no one else was even close (gervin at 26.0 pts/g since 79-80 was next best). as a matter of fact, at that time (after 85-86), dantley had the 4th highest scoring average in league history among players aged 23-29, behind only chamberlain (39.6 pts/g), baylor (30.8 pts/g), and jabbar (29.6 pts/g). that's some pretty select company to be associated with, and he was a significantly better shooter than all 3 players in that age range...

dantley was also an incredibly efficient player on offense - his 2.75 pts/0ptposs was 3rd highest over those 7 years. but from 81-82 to 85-86 isiah thomas got just 1.79 pts/0ptposs. that means for every time dantley cost his team a team possession (worth approximately 1 point) by missing a shot rebounded by the opponent, missing a FT rebounded by the opponent, or by committing a turnover, he scored 2.75 points. thomas scored just 1.79 points. dantley was far far more efficient on offense than thomas was...

yet in 86-87 and 87-88 combined thomas on offense took more shots than dantley did (dantley was there 2 and 1/2 years), in total and also on a per minute basis - did that make a whole lot of sense considering each's offensive efficiency? thomas also shot much more than 6 of his teammates that shot better overall those 2 seasons than he did. dantley, dumars, laimbeer, salley, rodman, and mahorn all shot better than thomas did, and all were more efficient on offense (pts/0ptposs) than was thomas...

those 2 seasons thomas took the 15th most FGAs among all players in the league (not just PGs but all players) and finished with the 5th most assists. 3 of the 4 PGs who had more assists (stockton, doc rivers, terry porter) each took 1000+ less shots than did thomas those 2 years. those 2 years thomas shot a 51.5% ScFG%, less than just what the league average PG shot (a 52.1% ScFG%)...

and any idea that thomas, as a high touches/min PG, had to take more shots because his teammates were incapable of creating shots for themselves is patently false. dantley got just 0.96 touches/min on offense his 2+ seasons with detroit, but in his 3 years with utah prior to being traded to the pistons he got 1.25 touches/min. so dantley, one of the most efficient high scorers the game had ever seen, saw a large 23% drop in his touches/min, primarily because thomas shot the ball so much, and despite the fact that he had shown the ability to score very efficiently with significantly higher touches/min on offense...

in 88-89 and 89-90 thomas again took far more shots than any other pistons player (and the most of any PG in the league), but shot just a 50.6% ScFG%, again less than what just the league average PG shot (a 52.2% eFG%). he threw for the 6th most total assists but lead the league in turnovers. stockton threw for 800+ more assists than thomas did these 2 years but with less turnovers. kevin johnson threw for 400+ more assists with less turnovers than thomas. thomas' offensive efficiency was simply not very good these two years, just 1.71 pts/0ptposs, well below that of the league average PG...

dumars got to detroit in 85-86 and was still there when thomas retired. he was all-D 1st team 4 times (including 88-89 and 89-90 when they won titles) and all-D 2nd team 1 time from 88-89 to 92-93. that means he was a dominant defender for 5 straight years. he scored 16 pts/g over his career and was a simply awful rebounder for an SG, so you can see that he was named to the hall of fame primarily because of his excellent defense...

rodman was with detroit from 86-87 to 92-93. he was all-D 1st team each year, including the 2 title years, and a top 6 rebounder on a per minute basis while shooting a 60% ScFG% those 2 seasons. so he was a stud defender, a stud rebounder, and efficient on offense with his few scoring attempts (just 9 pts/g the two title years)...

bill laimbeer was one of the league's top Cs from 82-83 to 90-91. those 9 seasons he grabbed the 2nd most rebounds among all Cs (next to moses malone) and scored the 5th most total points, and rarely missed a game (just 2 regular season games over those 9 seasons). so for pretty much every game over 9 years laimbeer gave the pistons close to 15 pts/g and 11 reb/g...

good players Toney, Jones, Dawkins

in the 5 years caldwell jones played for the 76ers (77-78 to 81-82) and the 4 years darryl dawkins played for them (78-79 to 81-82) they combined for just 10 pts/g and 9 reb/g. so why do you call jones and dawkins good players but laimbeer a role player?...

His higher usage allowed his role playing teammates to play their roles to the best of their abilities.... His ability to run the team and be high usage ALLOWED (in my eyes) Detroit team full of good role players to actually not suck (technically speaking) offensively.

so any idea that the pistons got to the ECFs and won titles with isiah thomas and a bunch of role players is not true. he played with 3 HOF players plus one of the top Cs at that time...

Everyone knew Thomas is what made that team go - and they won two championships.

sounds pretty clear cut the way you tell it. so tell us - how did the pistons win 2 titles? the fact is that in 88-89 and 89-90 combined in the regular season they ranked 2nd in defense (1.026 pts/poss allowed, only utah was better), but just 8th in offense (1.085 pts/poss scored). in the playoffs they were - by far - the best team defensively (just 1.017 pts/poss allowed, the next best team was at 1.042 pts/poss allowed). they ranked just 5th on offense in the playoffs. so they clearly won the two titles more because of their defense than their offense...

and their best defenders were rodman, dumars, salley, and laimbeer - the pistons players other than thomas to play the most minutes for them in the playoffs those 2 years...

Thomas was a higher usage guy because he was capable (in my eyes) of being a higher usage guy on a good/great team... I am completely certain that those championship Piston teams would have not gotten a sniff of the finals with in his prime Cheeks at their helm instead of Isiah.

completely certain - based on what? cheeks was the PG of the team with the 3rd highest W-L record over a full decade, the 76ers from 78-79 to 87-88 (one title). how could he be the starting PG for a great team over such a long time and win a title but not some other team? from 83-84 to 89-90 dennis johnson was the starting PG of the boston celtics who had the 2nd best winning percentage during that time (and one title). are you completely certain D.J. could not have lead those pistons teams to titles too? i ask as D.J.'s touches/min those 7 years were actually lower than those of cheeks from 78-79 to 87-88 with the 76ers...

correction: rodman was all-D 1st team from 88-89 to 92-93 with the pistons, not in 86-87 or 87-88...

[Mike G]

[Cheeks] was the PG of the team with the 3rd highest W-L record over a full decade, the 76ers from 78-79 to 87-88 (one title). how could he be the starting PG for a great team over such a long time and win a title but not some other team? from 83-84 to 89-90 dennis johnson was the starting PG of the boston celtics who had the 2nd best winning percentage during that time (and one title)...

The Celts actually won 2 titles after acquiring DJ.
Once the Pistons had acquired the deepest team in the league, it didn't really matter much who was taking the shots. They gave up Dantley to get Aguirre, a less efficient scorer. Why'd they do it? Apparently AD was not their ideal go-to scorer.

Rather like the '83 Sixers, the Pistons of 1989 and 1990 were close to unbeatable. Their '89 playoff record was 15-2, and the next year they went 15-5. (Five of 7 losses to the Bulls).
They could do quite well with Vinnie and Dumars in the game and Isiah on the bench. These 3 account for all but about 900 guard minutes over 2 seasons, including 23 games they missed.
In 1988-89, Dumars missed 13 games. They were 54-15 when he played, 63-19 overall, thus 9-4 without Joe.
In 1989-90, they were 54-19 when Dumars played and 5-4 when he didn't.
In 1990-91, Isiah would be out for 34 games. They were 31-17 when he appeared, and 19-15 without him.

One reason for the relative ease of their playoff runs (once past Jordan) was that all 3 guards tended to step up their games, and none more than Isiah.

In the 1989 season, Vinnie shoots the most (per minute) but the worst; Dumars the least but the best; Isiah in between.
Then in playoffs, Vinnie outshoots Isiah.

In the 1990 season, Vinnie shoots least of the 3, and still worse.
In playoffs, Isiah just took over. From .501 TS in the season, to .550 -- behind only Salley and Rodman.
In the Finals, Isiah hit .628, and 27.6 ppg. Vinnie next at .583, Dumars .522.

It seems guards have to take a certain portion of the shots. Among these Pistons' guards, Isiah tended to have about average TS% during the season, and better in the playoffs.
http://bkref.com/tiny/YHuRX

Cheeks and Thomas both made 3 trips to the Finals. Here are their composite per36 rates, standardized to 100 pts and 44 reb/tm/G:

Code: Select all

Finals    G   Min   Pct.   Sco   Reb   Ast    PF   Stl   TO   Blk    T
Cheeks   16   37   .538   14.3   2.5   6.9   2.4   2.5   2.6   .1   25.5
Thomas   16   37   .563   23.5   4.8   7.8   3.5   2.1   3.8   .3   35.7

Equal games and minutes, but the similarities end there.

[Mike G]

I'm not trying to be difficult, but that doesn't really address my concern. FT% isn't affected by team environment or league environment. There is no defense being played against a FTA, and there is no "pace" during a FTA. So the reasons for adjusting FG% or eFG% aren't applicable to FT%.

There's no "pace" while a FGA is in the air, either. If it goes in, the other team takes the ball out of bounds. If it bounces off the rim, both teams try to rebound it. Same with FTA.
The clock stops for a FTA, but that's just so players can line up along the lane, without time running off the clock. Since no time is running, the pace of the game is not affected. It resumes when the ball goes back in play. It's just a pause button.
The conditions that created the FTA or the FGA have much to do with what the defense has been up to. A player's FT are his FTA * FT%. You don't get FT without FTA.

Would it be any different if a foul led to one FT worth 2 points (or 3)? Stopping the clock doesn't affect Pts/G on either side. It may just lead to a high-% shot with no time elapsed.

[Statman]

you bet thomas had some nice teammates - 3 hall of famers (same as cheeks). while the 76ers with cheeks had malone, erving, and barkley, the pistons with thomas has dumars, adrian dantley, and dennis rodman

You actually think Moses/Dr. J/Sir Charles compared to Dumars/late AD/& early Rodman is a wash?

All the things you wrote about AD/Dumars/Rodman/ & Laimbeer are true. I was a HUGE Pistons fan - watched hundreds of their games through those years. I know all the stats like everyone else here. The AD at Detroit was NOT the AD in his prime at Utah - he still was one of the best in offensive efficiency in the league, BUT he was not a good defender, and the rest of his skills had fallen off a bit (rebounding, passing, ability to play large minutes, etc). He was not nearly a HoFer at Detroit. He was a ROLE player in that he had ONE ability - to post up and score efficiently. Don't ask him to defend, to create his shot off the dribble or run an offense, or to rebound. With all the solid arguments than AD should have had more shots in Detroit than Isiah, the fact it - they didn't win a championship until AD was gone, with Isiah still trowing up shots and getting assists.

Rodman - he was the ultimate ROLE player because he rebounded and played defense better than anyone - but that's ALL he did - he wasn't gonna do much on the offenive end but rebound and finish occasionally on the break. Dumars was a ROLE player who became an All Star - but he seemed to thrive as a partner to a high usage PG like Thomas. Laimbeer - rebound, piss people off, scrap, and hit his face up jumpers. I'll admit, he became more than a role player also.

But - we ALL know these good players with specific good/great skills - they WERE NOT NBA superstars to the level of Moses Malone, Dr. J, or Sir Charles.

good players Toney, Jones, Dawkins

in the 5 years caldwell jones played for the 76ers (77-78 to 81-82) and the 4 years darryl dawkins played for them (78-79 to 81-82) they combined for just 10 pts/g and 9 reb/g. so why do you call jones and dawkins good players but laimbeer a role player?...

Dawkins was pretty much 13/8 those 4 years (eyeballin' it) - not sure how you have him combined with someone else at 10 & 9 - it's OK - I was talking about Bobby Jones anyway, one of the very best defenders in the NBA in his time.

His higher usage allowed his role playing teammates to play their roles to the best of their abilities.... His ability to run the team and be high usage ALLOWED (in my eyes) Detroit team full of good role players to actually not suck (technically speaking) offensively.

so any idea that the pistons got to the ECFs and won titles with isiah thomas and a bunch of role players is not true. he played with 3 HOF players plus one of the top Cs at that time...

Maybe I should have highlighted "good" when I called them role players. Let's just say - none of the players Isiah had as teammates during those championships would have been classified by anyone as being anything near the talents of superstars the level of Moses Malone, Julius Erving, or Charles Barkley.

Everyone knew Thomas is what made that team go - and they won two championships.

sounds pretty clear cut the way you tell it. so tell us - how did the pistons win 2 titles? the fact is that in 88-89 and 89-90 combined in the regular season they ranked 2nd in defense (1.026 pts/poss allowed, only utah was better), but just 8th in offense (1.085 pts/poss scored). in the playoffs they were - by far - the best team defensively (just 1.017 pts/poss allowed, the next best team was at 1.042 pts/poss allowed). they ranked just 5th on offense in the playoffs. so they clearly won the two titles more because of their defense than their offense...

and their best defenders were rodman, dumars, salley, and laimbeer - the pistons players other than thomas to play the most minutes for them in the playoffs those 2 years...

Just like I said - he made that team GO. I didn't say he was the defensive anchor - but he ran the show, and was considered by his teammates, fans, coaches, and media as the star of the team. Cheeks was considered AT BEST to be the 3rd best player on any team he was on - champion or not. Who knows - maybe BOTH classifications of either player were wrong - BUT to act like it's absurd that a statistical ranking system like Mike G's has Thomas ahead of Cheeks, when pretty much EVERY fan that ever watched during the 80s would agree with Isaih being ahead of Cheeks, just because you don't think he scored efficiently enough seems a bit tunnel visioned to me.

Thomas was a higher usage guy because he was capable (in my eyes) of being a higher usage guy on a good/great team... I am completely certain that those championship Piston teams would have not gotten a sniff of the finals with in his prime Cheeks at their helm instead of Isiah.

completely certain - based on what? cheeks was the PG of the team with the 3rd highest W-L record over a full decade, the 76ers from 78-79 to 87-88 (one title). how could he be the starting PG for a great team over such a long time and win a title but not some other team? from 83-84 to 89-90 dennis johnson was the starting PG of the boston celtics who had the 2nd best winning percentage during that time (and one title). are you completely certain D.J. could not have lead those pistons teams to titles too? i ask as D.J.'s touches/min those 7 years were actually lower than those of cheeks from 78-79 to 87-88 with the 76ers...

I am certain based on my 30 some odd years of watching & playing basketball. Am I supposed to think it possible that Cheeks wins ONE championship with the sixers as a low usage PG - and he could replace a HIGH usage PG for a Pistons with no one on their roster that can compare to Moses Malone, Julius Erving, & Charles Barkley - and they would win a championship? Nope - I don't believe it's possible. I also don't believe DJ could have done it either.

For what it's worth - I do think it's possible (I'm not certain though) that the sixers win zero championships with a high usage PG like Isiah instead of Cheeks - or the Celtics don't win as many championships with Isiah in place of DJ. Or, maybe they do..... I dunno really.

BTW Bob - you do realize you are arguing with guys that almost ALL think Cheeks was way undervalued as a player - we just don't think he was a top 5 PG of all time type. Mike G's ratings probably has Cheeks ranked higher than most all known basketball "historians".

[Mike G]

It's funny that Bob's arguments are based entirely on regular season numbers, and it's the playoffs where Cheeks really excelled. As did Thomas. I was firmly convinced, through most of the '80s, that Isiah was a ball-hogging gunner -- until the '88 playoffs, and thereafter.

If I just count playoff games without additional weight, Jordan ranks 4th all-time, after Wilt, Kareem, and Malone. Robinson, Barkley, Kobe and Garnett pass Bird. Magic drops below them plus Erving, Pettit, Ewing, Dirk, and LeBron. Russell drops below all those plus Oscar, Moses, Stockton, Baylor, and West -- to #24.

Without boosting the importance of playoff minutes, Cheeks' stats (relative to his opponents' basically) would rank him at #157 all-time. Because of his superlative performances in the big games, he leapfrogs over all these players with more impressive totals/rates: D Harper, J Drew, Price, Smits, Murphy, Twyman, R Lewis, J Kerr, McDyess, Finley, Rice, Gallatin, Macauley, Daugherty, Brandon, E Jones, C Robinson, Terry, Mullin, Randolph, Blaylock, Greer, Willis, Ilgauskas, Reed, Richmond, Thorpe, Chambers, Bosh, Bing, Buck W, W Davis, A Miller, Camby, Shareef, Chet Walker, Jamison, Wilkens, King, Haywood, Marbury, Yao.

In his playoff minutes, Cheeks was about 1.05 times as productive, relative to opponents, as he was in regular seasons. Of all players with at least 4200 playoff minutes, the only ones with higher po/rs are Reggie, Jordan, Olajuwon, Horry and, at the tip top -- Isiah Thomas (1.10).

The top 100 players in minutes played average a career po/rs of .960
The next 100 average .942 ; #201-300 avg .938 ; then .932, etc

Without extra weight on playoffs, Isiah still would rank #56. Because of his playoffs' extra weight, I put his career over those of Dwight, Brand, Bellamy, Gasol, Gervin, Johnston, Allen, McAdoo, Mourning, Reggie, Carter, and Wilkins .
(Versatility Index also boosts him ahead of Issel, McHale, Payton, Reggie, and Iverson.)

Playoff/regular season ratio serves as a proxy for 'big game performance', 'clutch play'.
Playoff minutes register team success, and the lack thereof indicates relatively inflated stats, when playing for sub-par teams.

[AYC]

Mike, I still think you're avoiding my point. I don't think all the reasons for normalizing eFG% from era to era apply equally to FT%. When Wilt shoots 50% from the field in a league where the avg eFG% is 43%, that's far more impressive than doing that when the league avg is 50%. We adjust for era because we want to recognize the fact that he was far above his peers. The fair assumption is that the type of league he played in affected his production. Maybe defense was better back then (not likely in the 60s, but possibly true for the late 90s); or maybe the quality of his teammates and coaching was worse; or maybe a league averaging 120+ TSA/G wasn't too concerned with good shot selection (that's where pace arguably comes in). Later in his career, as the Lg pace slowed a little, coaching improved and Lg FG% went up, Wilt became far more efficient; I don't think that's entirely due to the change in his specific role; in general, the entire league was becoming more efficient at the same time.

FT% is different. There is no opposing D at the line. The clock isn't ticking. Coaching isn't an issue; neither is the quality of your supporting cast. At the FT line, BBall is an individual sport. And that's why the variance in eFG% is far greater than the variance in FT% over the course of Lg history. eFG% has ranged between .279 and .501, while FT% has ranged between .641 and .771; that's nearly an 80% increase in eFG%, vs a 20% increase in FT%. Take away the league's first two seasons and FT% has been between 70.3% and 77.1% since 1949. So, unlike eFG%, adjusting FT% doesn't seem to be necessary.

[Statman]

FT% is different. There is no opposing D at the line. The clock isn't ticking. Coaching isn't an issue; neither is the quality of your supporting cast. At the FT line, BBall is an individual sport. And that's why the variance in eFG% is far greater than the variance in FT% over the course of Lg history. eFG% has ranged between .279 and .501, while FT% has ranged between .641 and .771; that's nearly an 80% increase in eFG%, vs a 20% increase in FT%. Take away the league's first two seasons and FT% has been between 70.3% and 77.1% since 1949. So, unlike eFG%, adjusting FT% doesn't seem to be necessary.

I disagree. Everything else equal - a guy shooting 75% from the FT line in an NBA where the average is 68% is accomplishing more from the line than a guy shooting 75% in an NBA where the average is 75%. He's helping his team more.

Also - I would expect FT %s to have changed through the years somewhat - softer rims (better for dunking), better arena conditions (lighting, temp, no fans smoking), etc. Shooting a FT in a 2011 NBA game is not the same as shooting one in a 1955 NBA game.

But, I digress. I think the important thing is how efficient and proficient the player is at scoring relative to the league. Pretty much for efficiency - how his TS% is relative to the league. How he gets his TS% - whether from mainly the field, threes, or free throws - doesn't matter that much to me. When it comes to free throws though - the MAJOR factor isn't the player's FT % related to the league (which hasn't changed THAT much through the years) - but his FREQUENCY at scoring points at the FT line relative to the league. TS% incorporates that (combining twos, threes, & FTs) - that's what I use (well, my version of TS%). Unless I'm completely confused - that's what Mike G uses too (I think his version is the same as mine).

[Mike G]

Everything else equal - a guy shooting 75% from the FT line in an NBA where the average is 68% is accomplishing more from the line than a guy shooting 75% in an NBA where the average is 75%. He's helping his team more.
... I think the important thing is how efficient and proficient the player is at scoring relative to the league. ..

Again, why does it matter what the league is doing in other games?
In a league where avg eFG% is .487, Joe Dumars shoots .487 (.552 TS%), scoring 19.3 Pts/36.
In the same league, Orlando Woolridge shoots .498 (and .565), getting 26.3/36 min.

Dumars shot the league avg eFG%, but it was .016 higher than his teammates, and it was .023 higher than opponents.
Woolridge shot .011 higher than Dumars and the league, and some .046 higher than his 1991 Enver teammates; but .026 worse than their opponents.

In both cases, the era and the league's averages have nothing to do with what goes on in Pistons games and Nuggets games.
Any adjustment made relative to league averages fails to address the fact that the players are scoring in radically different environments. Namely, for one of the best and one of the worst defensive teams of all time.

After accounting for team and opponent rates, I call both of them 21.4 Pts/36 scorers in 1991.
Then in 1992 and most of '93, Woolridge played for Detroit. Lo and behold, in the same environment, he and Dumars scored at almost the same rate.
http://www.basketball-reference.com/teams/DET/1992.html

[AYC]

Mike, if player A shot 75 FT% playing for the '68 Sixers (.635 Tm FT%) would he be more valuable than player B shooting 75% on the '68 Royals (.762 Tm FT%)? If so, you're saying player A is more valuable merely because he played alongside Wilt when he was at his most inept from the line. Again, I think it makes sense to adjust for team eFG%, but not FT%.

I will concede there are two ways to look at it. If you simply want to show a player's performance relative to his peers, using eff% or TS% is fine. But if you want to adjust stats based on how Lg or Tm environment (it doesn't matter which) affected individual performance, you should leave FT% untouched for the reasons I stated above.

[Mike G]

I don't think 75% FT, in and of itself, has more value in one environment than in another.
If you play alongside Wilt, and he's hogging the ball, and you're not getting as many FTA, your FT% is of less significance.

... you should leave FT% untouched ...

Somewhere you got the idea I've adjusted players' actual shooting percents ?
All I've done is multiply their scoring rates by (Eff%/.527), along with other factors.

A player generally earns his FTA by being aggressive with the ball. He converts them to points according to his FT%.
Are you saying a player's points by FTA are not indicative of his scoring efficiency?
I'm saying it's a high-% shot attempt, when you get to the line.

[Statman]

Everything else equal - a guy shooting 75% from the FT line in an NBA where the average is 68% is accomplishing more from the line than a guy shooting 75% in an NBA where the average is 75%. He's helping his team more.
... I think the important thing is how efficient and proficient the player is at scoring relative to the league. ..

Again, why does it matter what the league is doing in other games?
In a league where avg eFG% is .487, Joe Dumars shoots .487 (.552 TS%), scoring 19.3 Pts/36.
In the same league, Orlando Woolridge shoots .498 (and .565), getting 26.3/36 min.

Dumars shot the league avg eFG%, but it was .016 higher than his teammates, and it was .023 higher than opponents.
Woolridge shot .011 higher than Dumars and the league, and some .046 higher than his 1991 Enver teammates; but .026 worse than their opponents.

In both cases, the era and the league's averages have nothing to do with what goes on in Pistons games and Nuggets games.
Any adjustment made relative to league averages fails to address the fact that the players are scoring in radically different environments. Namely, for one of the best and one of the worst defensive teams of all time.

After accounting for team and opponent rates, I call both of them 21.4 Pts/36 scorers in 1991.
Then in 1992 and most of '93, Woolridge played for Detroit. Lo and behold, in the same environment, he and Dumars scored at almost the same rate.
http://www.basketball-reference.com/teams/DET/1992.html

Quick answer - in probably not quite right terms - it "normalizes" the stats better imo.

I COMPLETELY agree with your specific team adjustments - I do the same - BUT I normalize to league average FIRST - THEN adjust for team quality, pace, & defense. Otherwise - Dumars probably isn't being helped even enough when you just look at just the Pistons (at their peak) & their opponents - since the average quality of the teams in those games were well above average (a great Pistons team vs. about league average opponent). Dumars is being judged in a stats pool of, say, a 1.04 factor - while a player on a BAD team would be judged in a stats pool of about 0.96 or so. If you normalize to league first - THEN adjust to team quality/pace/defense - the playing field is more level.

Anyway - that's why I do it the way I do - since I want to create an entire adjusted statline. Also, for example, if a guy plays on a team that rarely turns the ball over (few opponent steals) - why should his adjusted steals totals be lower because his team avoided steals real well? Adjusted stats to league and then ratings to team pace/defense/quality helps eliminate that problem (decent looking adjusted statline and ratings) for me.

[Mike G]

I normalize to league average FIRST - THEN adjust for team quality, pace, & defense. Otherwise - Dumars probably isn't being helped even enough when you just look at just the Pistons (at their peak) & their opponents - since the average quality of the teams in those games were well above average (a great Pistons team vs. about league average opponent). Dumars is being judged in a stats pool of, say, a 1.04 factor - while a player on a BAD team would be judged in a stats pool of about 0.96 or so. If you normalize to league first - THEN adjust to team quality/pace/defense - the playing field is more level.

Hmm, well you could scale player rate vs league average rate, then scale team and opponents' rate to league rate, and what you'd end up with is player rate vs opponent rate. I just skip the redundancy.

In the example I gave, Dumars' raw Pts/36 was 19.3; adjustments boost that to 21.4.
This is equivalent to his being not the 20th best scorer of 1991, but about 12th.

I've adjusted players' Sco rates only to that of their opponents; assists to Tm+Opp points; rebounds also to both, but more to opponent rates.