APBRmetrics

LINK AT THE BOTTOM OF THE PAGE!!!

Each player is ranked according to individually estimated contribution, or Value Share.

My concept of Value Shares is based on linear weights including Rosenbaum's Alternate Win Score, and the Value Over Replacement Player (VORP) metric. Because each of these metrics seem to lack especially in measuring individual defensive effectiveness, through my research I've implemented my own tweeks, most importantly to factor in defensive value not captured in the box score.

The Value Shares formula is a counting stat with a high correlation with wins, it comprises of other details including my own recalculated Usg%, but significant adjustments are the reassessed box stats.

The first reassessment is the time period regulation, which covers among others things the change in pace, generally from decade to decade although I made some adjustments to mirror major NBA rule changes. It's fundamentally impossible to even start with comparing players across from different time periods without first forming a neutral period to serve as the common denominator. Generally, RPG, FTA in the 60s is higher than it is today, possibly due to pace. I didn't get into the "why" much, although there are very good explanations there. I simply took account of the league averages. The most important feature of the time period regulations, is of course which time period do I denominate all the box stats? In concluding my research, I found the early 90s to be the most neutral time period in NBA history to this point. By neutral I mean, the league averages and pace were fairly moderate, and "top-to-bottom" there was generally high level play from all positions.

That brings me to my next reassessment, the league quality adjustment. There are many factors that have affected the NBA quality of competition in my estimation: Expansion, Demographics, Rule Changes, etc. In my selection of the 90s as the most neutral period, the rules in particular allowed for a more balanced game. The lane dimensions compared to the early 60s, physicality compared to today's NBA (hand-checking, defending around screens, etc.), Rules governing offensive/defensive play in the paint, all allowed for a more quality game. If you were a poor defensive player, you were more likely to get exposed and exploited, if you were a great offensive talent, you were more likely to showcase your skill. In measuring League Quality, I used ADJP48 for average players, generally per decade for the seasons available, and average player WS/48 for the seasons where ADJP48 is not available.

Lastly,

*I prorated the adjusted box stats for strike-shortened seasons
*All Season stats totals include post season stats.
*I made slight adjustments in 2pt FG, FGA for pre-3pt NBA periods
*I made slight adjustments for Assists totals for pre-NBA/ABA merger seasons, it's my belief that scorekeepers were less generous during these periods
*70s ABA league quality = 60s NBA league quality
*Used multiple statistical models to estimate unavailable Box Stats
*GP are calculated as Total Minutes/36 = GP
etc.

https://docs.google.com/spreadsheets/d/ ... oE/pubhtml

Surprises or highlights for you?

I found many interesting results, and thanks for the question bro.


1.) Bill Russell, a player ranked in many top 10s ranks 67th on my list.

Russell finished with 170.02 Total Value Shares, a career ranking 32nd all-time. Dwyane Wade will likely overtake Russell next year.

I had to estimate blocks, steals, etc., prior to applying the adjustments. Russell and Ben Wallace (#63) come out even more similar than I expected.

Wallace's career isn't quite as extensive as Russell's, Wallace finished with 133.40 in Total Value Shares (#63)

Wallace adj. Career Stats

FG% = .491
RPG = 12.02
ORPG = 4.38
APG = 1.74
SPG = 1.74
BPG = 2.42
TOV = 1.25
PPG = 7.97

Russell adj. Career Stats

FG% = .504
RPG = 12.11
ORPG = 3.10
APG = 3.55
SPG = 1.39
BPG = 2.33
TOV = 2.34
PPG = 10.60

Very similar.


2.) Chris Paul, still an active player obviously, ranks as the 2nd best PG of all-time. Paul currently trials Magic by a fairly wide margin however.

Paul has a USG% = 27.26
FG% = .484
APG = 11.1 to only 2.59 TOV per game
19.16 PPG/36

In a neutral period, given my research, Paul numbers are good enough to earn 3 MVPs.

Paul is 36th in Total Value Shares 163.97, passing Havlicek and Jerry West this season.

3.) Dale Ellis 1989 Season.

Steph Curry's 2016 season (RS. PS) adjusted = 299 3PM (1st all-time) @ .436 pct
Curry's 2015 season adjusted = 223 3PM (4th all-time) @.433 pct

But Ellis' 3pt efficiency stands out @ a time when perimeter defense and offensive schemes made it much more difficult.

Ellis '89 season adjusted = 180 3PM (17th all-time) but @ a remarkable .527 pct

I'd probably still go with Curry's season as the best long range shooting season in NBA history because of the sheer volume, but Ellis' '89 season deserves strong consideration.


4.) Shaq's 2000 season (RS, PS) registers as the most dominant season of all-time. Over Wilt's '62.. which I thought would've been higher than what it was. Chamberlain's '62 season ranks 15th all-time once adjusted.

Shaq finished the 2000 season (RS, PS):

PPG = 29.93
RPG = 12.47
BPG = 2.58
FG% = .603
In 4217 total minutes for a Total Single Season Value Shares = 26.58, topping Michael Jordan's 1990 Season = 25.66 Value Shares.

These where just the four things that immediately jumped out ...

All interesting comparisons.

... In measuring League Quality, I used ADJP48 for average players, generally per decade for the seasons available, and average player WS/48 for the seasons where ADJP48 is not available. ..

What is ADJP48?
Isn't avg WS/48 just .100 (or .099) all the time?

*70s ABA league quality = 60s NBA league quality

Don't you think ABA quality improved during the interval 1970-76? It may be Spencer Haywood's 1970 ABA was more like the inaugural 1968 version than it resembled post-'71, when the ABA not only dominated the draft but actually had contraction in # of teams.

*GP are calculated as Total Minutes/36 = GP

What about calling these 'equivalent games' ? or some other term not already in use?

Bill Russell, a player ranked in many top 10s ranks 67th on my list.

This is startling, and it's also surprising since a couple of years ago you ranked him 8th.
What accounts for the downgrade?

Sorry if this is too many inquiries at once. But you have more choices in what you reply to!
The Bill Russell downgrade prompted me to look at his totals, as you have adjusted them. And rebounds are a big part of his statistical record.
Here are the top 12 career rebounders in chronological order. Showing official (RS+PO) rebounds, the D-rell adjusted total, and the ratio of the 2.Gilmore gets dinged by his ABA time, I suppose. How does Kareem do so well? Did you get a satisfactory estimate of opponent rebounds before 1971?

After my league adjustments, which account for pace among things, my assessment of Centers between 1950 - 1980 shows the NBA progressed in quality each decade. I indexed each decade from 1950 for 1990s league levels. This factor results in a significant decrease in rebounds:

Bill Russell's career RPG is reduced from 22.5 to 12.11 in a neutral era. It's important to pool players in a neutral era, because now I can compare his rebound stats in a more fair and accurate way. By comparison, Dikembe Mutumbo's career RPG=12.18 but in 300 less games (RS & PS). I estimated Russell, Lucas, and Chamberlain ORB = 25-26% * TRB, possibly a bit too conservative for top rebounders (I may adjust the ORB levels without changing TRB). Russell finishes with 3.1 ORPG, compared to Mutumbo @ 4.07.

The formula I used to estimate blocks before the box stats:((((Sqrt(Calc. GP/TRB)-1.6)*TRB)*1.13888888888889)+(49))*0.76)))

Russell's top two shot-blocking seasons after the adjustments:

'62 = 205
'59 = 202

Career Blocks = 3186 10th all-time.

In a neutral era Russell is exposed as a subpar offensive player and a top 10 defensive player.



I'll give more of a response when I get some more time

D-rell wrote: The formula I used to estimate blocks before the box stats:((((Sqrt(Calc. GP/TRB)-1.6)*TRB)*1.13888888888889)+(49))*0.76)))

Well. It seems my formula is (sqrt(Reb) - 1.6) also
The other stuff I don't understand, and extra parentheses?
Anyway, from anecdotal evidence, I arbitrarily bump Russell's Blk/36 up to 4.0
Otherwise, Jerry Lucas is about as good. Never heard of him blocking shots.

Meanwhile, Kareem thrived in an era of 2nd-string centers promoted to starters, during the hyper-inflationary early '70s. Almost half his opponents were expansion teams for a while. He put up his biggest numbers then.

3.) Dale Ellis 1989 Season.

Steph Curry's 2016 season (RS. PS) adjusted = 299 3PM (1st all-time) @ .436 pct
Curry's 2015 season adjusted = 223 3PM (4th all-time) @.433 pct

But Ellis' 3pt efficiency stands out @ a time when perimeter defense and offensive schemes made it much more difficult.

Ellis '89 season adjusted = 180 3PM (17th all-time) but @ a remarkable .527 pct

You really think perimeter defense was greater back then?
I see Curry hitting from 30+ feet and think there is nobody in my memory who could do that with any consistency. He does this because modern defenses are geared to defending the arc. In 1989, as I recall, anyone could get a 24' shot at any time.

So you've deducted 103 of Steph's 402 3FG this season, an effective tax rate of 25.6%, due to lax (relative to early '90s) perimeter D that allows him open looks at 30' ?
And Dale Ellis '89 hit 162/339 from the arc; you've expanded that to 180/342, relative to some neutral era?

I can appreciate the scaling of rebounds to the # of rebounds available. But my eyes and brain tell me guys are shooting the long ball like never before; teams are collecting players with this skill; and this isn't a trend that is due to defenses now failing -- inexplicably -- to chase shooters off the arc.

Mike G wrote:
I can appreciate the scaling of rebounds to the # of rebounds available. But my eyes and brain tell me guys are shooting the long ball like never before; teams are collecting players with this skill; and this isn't a trend that is due to defenses now failing -- inexplicably -- to chase shooters off the arc.

I think it makes much more sense just to scale scoring & TS% in general (general pace & defense), but not threes or free throws or 2pt FGs.

It's like if you scaled Babe Ruth's home runs back in 1927 to today's game - and credit him for 127 home runs that season, obviously horribly skewing his relative value. Just scale his OPS & call it a day, that just seems like it'd make much more sense.

Statman wrote:

Mike G wrote:
I can appreciate the scaling of rebounds to the # of rebounds available. But my eyes and brain tell me guys are shooting the long ball like never before; teams are collecting players with this skill; and this isn't a trend that is due to defenses now failing -- inexplicably -- to chase shooters off the arc.

I think it makes much more sense just to scale scoring & TS% in general (general pace & defense), but not threes or free throws or 2pt FGs.

It's like if you scaled Babe Ruth's home runs back in 1927 to today's game - and credit him for 127 home runs that season, obviously horribly skewing his relative value. Just scale his OPS & call it a day, that just seems like it'd make much more sense.

Bonds was probably a million times better, Athletes evolve over time.

Sports are pointless to watch if one can't accept that every few years someone better than your hero appears (yes, even if the stats relative to era aren't comparable). Babe Ruth's stats should be scaled for the putrid era he played in with no talent, or volume population-wise, so his homerun total for 1927 is irrelevant in my opinion.

I don't mind if people here disagree, unless it is one of those 538-type guys who should know better.

Mike G wrote:

3.) Dale Ellis 1989 Season.

Steph Curry's 2016 season (RS. PS) adjusted = 299 3PM (1st all-time) @ .436 pct
Curry's 2015 season adjusted = 223 3PM (4th all-time) @.433 pct

But Ellis' 3pt efficiency stands out @ a time when perimeter defense and offensive schemes made it much more difficult.

Ellis '89 season adjusted = 180 3PM (17th all-time) but @ a remarkable .527 pct

You really think perimeter defense was greater back then?
I see Curry hitting from 30+ feet and think there is nobody in my memory who could do that with any consistency. He does this because modern defenses are geared to defending the arc. In 1989, as I recall, anyone could get a 24' shot at any time.

So you've deducted 103 of Steph's 402 3FG this season, an effective tax rate of 25.6%, due to lax (relative to early '90s) perimeter D that allows him open looks at 30' ?
And Dale Ellis '89 hit 162/339 from the arc; you've expanded that to 180/342, relative to some neutral era?

I can appreciate the scaling of rebounds to the # of rebounds available. But my eyes and brain tell me guys are shooting the long ball like never before; teams are collecting players with this skill; and this isn't a trend that is due to defenses now failing -- inexplicably -- to chase shooters off the arc.

I disagree..

I think perimeter defense was much better in the 80s and early 90s, because there was much less official regulations involved in perimeter play. Great perimeter defenders such as Cheeks, Dumars, Jordan, Dennis Johnson, Moncrief were allowed by scope of the rules to maximize their skill on the defensive side of the ball. Later, to some extent, Payton, Kidd. But by the 2004-05 had, much like the NFL ten years earlier, shifted dramatically to building advantages for offensive players. Obviously, coaching philosophies grew to exploit these rule changes, and it's given birth to the environment to today where Curry, Lillard, Irving, are virtually unarguable by perimeter defenders.

I'm careful to not engage too much in an imaginary realm where the physical translations of players into other era's take precedent over what I'm attempting to do which is simply to index each players value to reflect the league wide statistical changes over the various time periods in NBA history. My theory, is that by accurately adjusting the stats, I can for the most part account the changes pace and rules, etc.

Curry, with just 533 games (P-36, RS, PS), is still arguably the greatest shooter all-time even factoring in the rule changes covering perimeter defense. But by neutralizing the time periods, I think the skill of Ellis emerges more clearly. I score Ellis with 1641 career 3pt, @ .436 pct. Curry currently post 1087, @ .434. But Curry's range, by my eye test, dwarfs Ellis'. However defenders were much more empowered in Ellis' day, from the back court on down.

Statman wrote:

Mike G wrote:

I think it makes much more sense just to scale scoring & TS% in general (general pace & defense), but not threes or free throws or 2pt FGs.

It's like if you scaled Babe Ruth's home runs back in 1927 to today's game - and credit him for 127 home runs that season, obviously horribly skewing his relative value. Just scale his OPS & call it a day, that just seems like it'd make much more sense.

If you neutralize the pace, and the ratio of 3pts goes up, then 2pt FG go down.

I've actually done the baseball numbers, and of course factored a regression to mean to provide a ceiling that prevents something as impossible as 127 HR's. Obviously, if a player is skilled at the long ball, you'd have to adjust his Walks, many of them intentional.

But some notable findings in my MLB adjustments.

1. I found today's era to be neutral enough to denominate all player stats
2. Ruth had an enormous propensity to strike out, especially in comparison with Bonds who had much more discipline at the plate
3. Ruth played in an era before relief pitchers were popular. Pitchers were overworked and fatigued by the late innings; no fresh relief specialist
4. Ruth played in the live ball era, allegedly the baseball was doctored at the center to induce more offense. The live ball numbers across the league were insane.
5. Ruth played in a Yankee stadium with left field distance favorable to both he and Gehrig. Bonds played in predominantly San Fran parks notoriously tough for lefties

There's much more to point out, including that Ruth played in an era prior to the color line in which Blacks, Hispanics, and Asians competed separately.

But... While Bonds adjust as the better overall player, and they're more so equal as hitters

Bonds adj. OPS = 1.057
Ruth adj. OPS = 1.067

Because although Bonds played in a higher quality era, and in tougher parks for his handedness. Bonds ultimately played in the era where it was easiest to Homer, the "steriod" era.

Bonds adjusted career HRs = 664
Ruth adjusted career HRs = 651


Even considering Ruth's dominant stint as a pitcher with Boston, Bonds' ability to hit, field, and run the bases makes him, in my opinion, regardless of era the greatest baseball player of all-time. Follow by Clemens, Maddux, and Willie Mays.


I know this is a NBA forum, but was too tempted to respond to the baseball comparison.

D-rell wrote: Bonds ultimately played in the era where it was easiest to Homer, the "steriod" era.

Bonds adjusted career HRs = 664
Ruth adjusted career HRs = 651

...Bonds' ability to hit, field, and run the bases makes him, in my opinion, regardless of era the greatest baseball player of all-time.

Does everyone in the "steroid era" get the same deduction, whether or not they were steroid users?
664/771 = .861, so he gets 14% of his HR chipped off?

Bonds got to the postseason 7 times in 20 seasons. In these, he hit .245, slugged .503
His RS numbers were .298 and .607
That's quite a downturn in 151 at bats.

He was in 2 winning playoff series, losing 7.

Each player is ranked according to individually estimated contribution, or Value Share.

the league first kept track of minutes in 51-52. from 51-52 to 58-59 a full 4.5% of the regular season minutes of league history (5152-1516) was played, yet of your top 400 career ranked players there are only 6 players from the 1950s listed (mikan, lovellete, johnston, arizin, cousy, twyman) - that's just 1.5% of the top 400 players, 3 times less than the minutes played. why?...