RealGM 2017 Top 100 List #75 (Chris Bosh)

Moderators: Doctor MJ, trex_8063, penbeast0, PaulieWal, Clyde Frazier

trex_8063
Forum Mod
Forum Mod
Posts: 12,731
And1: 8,360
Joined: Feb 24, 2013
     

Re: RealGM 2017 Top 100 List #75: RUNOFF! Greer vs Bosh 

Post#41 » by trex_8063 » Fri Dec 29, 2017 6:35 pm

dhsilv2 wrote:
trex_8063 wrote:
dhsilv2 wrote:I do'nt deny there was a glut of talent at the 4 during his era, but that goes back to if rules/style were good for 4's or the 4's were just strong during that era.


It's not just the 4's. Bear in mind, All-NBA honors doesn't require one SF and one PF for each team; it can be two SF's or two PF's (more recently, I think they just require three frontcourt players----from any position---given the blurring of the lines between positions these days). That's why I was also listing the Lebrons, Durants, Pierces, etc of the game.


Yes and no. The all nba selections tend to of the 6 spots go 3-3 on small forwards and power forwards, though certainly not always.


True that they are not always an equal 3-3 split ('13 is a recent example of it going 4-2). But important to note that even though it usually goes 3-3 overall, it's frequently NOT 1-1 within each individual team. There were a whole string of years recently where the 1st Team went to two SF's (usually Lebron and Durant).


dhsilv2 wrote: They do separate centers though, so it's not 3 front court guys like the allstar has been. This is a point I've brought up elsewhere in that there is center bias in these awards.


It sort of depends on who's classification of position you trust. There are some years where guys like Tim Duncan or Pau Gasol are playing more at center (according to bbref's play-by-play data), but received All-NBA honors as a forward. Sometimes bbref's play-by-play data lists someone as more commonly playing C, but still designates them a PF, too. So it gets muddy.
Look at '15 as an example: DeMarcus Cousins (definitive C) gets All-NBA 2nd Team. But next to him on that All-NBA 2nd Team----also designated C, which is corroborated by bbref's play-by-play data [just slightly: 52% C to 48% PF]----is Pau Gasol. On the 3rd Team is DeAndre Jordan getting honors at C (and obv he is definitively a C).......but also receiving All-NBA 3rd Team honors (designated as F, even though bbref's pbp data indicates he played 100% as a C) is Tim Duncan.
So the positional designations are very very murky to say the least.


dhsilv2 wrote:The case for Greer is consistency and longevity.


I agree.

dhsilv2 wrote:The 7 2nd team all nba's.......helps make that case


More or less agree; though....

dhsilv2 wrote:The argument against him is a poor peak.


......also weaker era.


dhsilv2 wrote: along with the work I did on PER*minutes vs peers or at least I hope so.


fwiw, I did a "total career value" metric based on PER, WS/48, and minutes played (for both rs and playoffs), where the measure is set to assess how much value over a [approx] replacement level player an individual provides over his career.

A "replacement level player" was designated as PER 13.5 and .078 WS/48 for the rs sample; PER 12.5 and .064 WS/48 for the playoffs. So for Player A for the rs it basically boils down to:

{[(aPERrs - 13.5) * y] * rs minutes played} + {[(aWS/48rs - .078) * z] * rs minutes played}

Where aPERrs is Player A's rs PER
y and z are modifiers set such that, for example, 10,000 minutes at PER of 15.0 would equal the same value as 10,000 minutes of WS/48 of .100.

Then the same is done for the ps sample with Player A's ps PER and WS/48 (vs 12.5 and .064 as the new values for replacement level--->recognizing that most players dip a bit in the playoffs), except that the playoff minutes are multiplied by 3.25 (putting higher emphasis on playoff minutes).

The rs total and ps totals are added for the overall score. fwiw, Chris Bosh (as of the end of last season) ranks 59th all-time by this metric; Hal Greer ranks 166th.
I did a scaled version (based on my scaled PER and WS/48 standard dev studies from awhile back----can provide links later if you haven't seen them) where standard deviations above replacement level are utilized instead of just raw values. That version drops Bosh down to 63rd all-time, moves Hal Greer up to 162nd; so still a big separation.


Not necessarily saying I think Hal Greer is a poor candidate at this stage, though I admit I have no intention of supporting him any time soon (unless he comes up in a runoff against Mel Daniels perhaps......sorry pen :( ); but suffice to say I've a variety of analytics and broad-based methodologies which have led me to think more highly of any number of other potential candidates, and to believe that mainstream ("status quo", if you will) rankings of Hal Greer have perhaps overrated him.
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
dhsilv2
RealGM
Posts: 51,013
And1: 27,513
Joined: Oct 04, 2015

Re: RealGM 2017 Top 100 List #75: RUNOFF! Greer vs Bosh 

Post#42 » by dhsilv2 » Fri Dec 29, 2017 6:46 pm

trex_8063 wrote:fwiw, I did a "total career value" metric based on PER, WS/48, and minutes played (for both rs and playoffs), where the measure is set to assess how much value over a [approx] replacement level player an individual provides over his career.

A "replacement level player" was designated as PER 13.5 and .078 WS/48 for the rs sample; PER 12.5 and .064 WS/48 for the playoffs. So for Player A for the rs it basically boils down to:

{[(aPERrs - 13.5) * y] * rs minutes played} + {[(aWS/48rs - .078) * z] * rs minutes played}

Where aPERrs is Player A's rs PER
y and z are modifiers set such that, for example, 10,000 minutes at PER of 15.0 would equal the same value as 10,000 minutes of WS/48 of .100.

Then the same is done for the ps sample with Player A's ps PER and WS/48 (vs 12.5 and .064 as the new values for replacement level--->recognizing that most players dip a bit in the playoffs), except that the playoff minutes are multiplied by 3.25 (putting higher emphasis on playoff minutes).

The rs total and ps totals are added for the overall score. fwiw, Chris Bosh (as of the end of last season) ranks 59th all-time by this metric; Hal Greer ranks 166th.
I did a scaled version (based on my scaled PER and WS/48 standard dev studies from awhile back----can provide links later if you haven't seen them) where standard deviations above replacement level are utilized instead of just raw values. That version drops Bosh down to 63rd all-time, moves Hal Greer up to 162nd; so still a big separation.


Not necessarily saying I think Hal Greer is a poor candidate at this stage, though I admit I have no intention of supporting him any time soon (unless he comes up in a runoff against Mel Daniels perhaps......sorry pen :( ); but suffice to say I've a variety of analytics and broad-based methodologies which have led me to think more highly of any number of other potential candidates, and to believe that mainstream ("status quo", if you will) rankings of Hal Greer have perhaps overrated him.


The issue with this method is that during this era there were imo rules/styles of play that greatly undervalued or reduced the value of a guard imo. While I hate metrics that normalize positions (see Hollinger's Value add type metrics with PER), I think there is real validity with Greer and guards of that era. This is likely partially why I likely would have West higher than some (possibly in my top 10). I certainly could be wrong in doing this mental gymnastics, but I don't believe I'm completely off base.
trex_8063
Forum Mod
Forum Mod
Posts: 12,731
And1: 8,360
Joined: Feb 24, 2013
     

Re: RealGM 2017 Top 100 List #75: RUNOFF! Greer vs Bosh 

Post#43 » by trex_8063 » Fri Dec 29, 2017 7:36 pm

dhsilv2 wrote:FYI, he was for most of his career listed as a center. If Amar'e Stoudemire can get multiple all nba center selections, I don't think the voters would have had much of a problem moving him over there during some down center years either.


Most years, Bosh played a bit of both (or sometimes THREE) positions, though history seems to want to consider him among the PF's.

And again, the positional designations are murky......
In '06, bbref designates him a "C", even though their pbp data shows him as playing C just 49% of the time (33% PF and 18% of the time at SF!). How he was playing that little of center is somewhat hard to figure, as there isn't a consistent center in the rotation: they were sort of inconsistently platooning Antonio Davis, Rafael Araujo, Loren Woods, and Pape Sow at the C; Matt Bonner may also have seen some very limited time at C.

In '07, bbref again designates him a "C" even though their pbp data shows him as playing 51% of the time at PF (plus 4% of the time at SF). And if you look at the roster, you'll indeed note that Rasho Nesterovic was the primary staring C (21.0 mpg in 80 games), and Andrea Bargnani played a bit of both (25 mpg playing 51% C/49% PF, according to the pbp).


So idk. Yes, he played a lot of C, but that doesn't necessarily mean he'll be designated so in the eyes of the media.
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire

Return to Player Comparisons