RealGM Top 100 All-Time (2017)

[dhsilv2]

love the change!

[SactoKingsFan]

Sounds good.

Sent from my ONEPLUS A3000 using Tapatalk

[Dr Positivity]

What do you guys think of going back to the nomination system? This would cut down the votes to 10 candidates, and may prevent voters like the Cousy group from throwing away their vote for many threads in a row

[Doctor MJ]

What do you guys think of going back to the nomination system? This would cut down the votes to 10 candidates, and may prevent voters like the Cousy group from throwing away their vote for many threads in a row

I think the nomination system is a good one.
I think it's best to limit the number of candidates to no more than 5 or 6.
But the Alt vote system should mean no one is throwing their vote away. Just vote for Cousy and then vote for a strong contender.

Also, changing rules at this point seems likely to cause confusion that might reduce participation further.


Sent from my iPhone using RealGM mobile app

[Clyde Frazier]

Read on Twitter

These are the worst kind of fans... ugh

[trex_8063]

fyi, just made the following adjustments to the vote tallying protocol (it doesn't change anything you guys do, so don't worry), more or less just hammering out some specifics:

EDIT (updated 9/9/17): Beginning in thread #41, we will continue using the ranked choice vote as we have been as a means of narrowing the field down to THREE candidates (or potentially 4-5, in the event of a tie for 3rd place, or a 3-way tie for 2nd, etc). We will then automatically enter a run-off [which will be open to ALL registered project participants, even if they had not cast an initial vote in that particular thread] between those 3+ candidates----EDIT (9/11/17); UNLESS it is possible to narrow to ONE candidate who obtains a true majority----where participants are asked to specify with ONE of those candidates they favor (at least cursory reasons for one's pick is required). EDIT (9/11/17): In the event the run-off does not yield a majority for one player, we will enter a second run-off between the top 2.
EDIT of 9/10/17: In the event that the ranked choice system does not clearly narrow it to 5 or fewer candidates, the number of 2nd ballot votes received will be used as the means of determining the run-off candidates. Example: there are 11 total votes cast, Player A and Player B each received two 1st ballot votes, and then seven other players received one 1st ballot vote each. Players A & B will automatically be included in the run-off, as they had the most 1st ballot votes. Which 1 [or more] of the other seven candidates is to be included will be determined by which of them has the most 2nd ballot votes.

Summary of changes:
*Although it's highly unlikely at this stage that anyone would every be able to obtain a true majority in the initial vote [before run-off], I thought I better allow for the vague possibility (so that's what the first edit is). If that actually is possible, we will not have a run-off at all.

**Since we're doing run-offs between three (or more) candidates, it occurred to me that it might not be appropriate to award the spot to the guy who has the most votes in the initial run-off, but continue doing run-offs until someone has a majority. It didn't matter in this most recent #41 thread: Gervin (10) had more than Miller (5) and Pierce (3) combined. But what if out of those 18 votes it had been 8 for Gervin, 6 for Miller, and 4 for Pierce? There, Gervin has the most, but he doesn't have a majority (hypothetically, all four of Pierce's supporters would choose Miller if given a choice between Miller and Gervin). So that's what the second edit addresses. We'll run-off until a majority for ONE player is reached.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

[dhsilv2]

fyi, just made the following adjustments to the vote tallying protocol (it doesn't change anything you guys do, so don't worry), more or less just hammering out some specifics:

EDIT (updated 9/9/17): Beginning in thread #41, we will continue using the ranked choice vote as we have been as a means of narrowing the field down to THREE candidates (or potentially 4-5, in the event of a tie for 3rd place, or a 3-way tie for 2nd, etc). We will then automatically enter a run-off [which will be open to ALL registered project participants, even if they had not cast an initial vote in that particular thread] between those 3+ candidates----EDIT (9/11/17); UNLESS it is possible to narrow to ONE candidate who obtains a true majority----where participants are asked to specify with ONE of those candidates they favor (at least cursory reasons for one's pick is required). EDIT (9/11/17): In the event the run-off does not yield a majority for one player, we will enter a second run-off between the top 2.
EDIT of 9/10/17: In the event that the ranked choice system does not clearly narrow it to 5 or fewer candidates, the number of 2nd ballot votes received will be used as the means of determining the run-off candidates. Example: there are 11 total votes cast, Player A and Player B each received two 1st ballot votes, and then seven other players received one 1st ballot vote each. Players A & B will automatically be included in the run-off, as they had the most 1st ballot votes. Which 1 [or more] of the other seven candidates is to be included will be determined by which of them has the most 2nd ballot votes.

Summary of changes:
*Although it's highly unlikely at this stage that anyone would every be able to obtain a true majority in the initial vote [before run-off], I thought I better allow for the vague possibility (so that's what the first edit is). If that actually is possible, we will not have a run-off at all.

**Since we're doing run-offs between three (or more) candidates, it occurred to me that it might not be appropriate to award the spot to the guy who has the most votes in the initial run-off, but continue doing run-offs until someone has a majority. It didn't matter in this most recent #41 thread: Gervin (10) had more than Miller (5) and Pierce (3) combined. But what if out of those 18 votes it had been 8 for Gervin, 6 for Miller, and 4 for Pierce? There, Gervin has the most, but he doesn't have a majority (hypothetically, all four of Pierce's supporters would choose Miller if given a choice between Miller and Gervin). So that's what the second edit addresses. We'll run-off until a majority for ONE player is reached.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

In the interest of time, would not just doing an alt in a 3 way not avoid the second day of runoff?

[trex_8063]

In the interest of time, would not just doing an alt in a 3 way not avoid the second day of runoff?

I think if time becomes an issue in this manner, I'll likely just change it so that the first runoff is between the top 2 only.

[ElGee]

It's really interesting see how much value some people place on longevity.

If I am to put my GM hat on, everything in basketball is about winning a title. That's the whole thing. Everything else is somewhat meaningless to most fans and it seems GMs.

So while longevity from a number 2 or 3 guy or a number 1 who transitions is really useful, but guys who build title contenders they are unique and who cares how long they did it?

This is why for me Reed is on another from the likes of Miller or even Pierce (Who I do think is a bit better). I'd need 40 years of those two before they matched Reed's 4 years, or something like that. Now I might be downplaying a few top years for those guys that close the gap (I am to make my point).

So at least for those frustrated by the dismissal of Miller type players, it's because he was just never good enough to move up and I see marginal improvements after a point to be exponentially valuable. A 9 season is worth 3 8's, 12, 7's if you will type growth. Because truly great seasons are hard to come by.

I'd refer you to this study by Elgee (I guess some of the nitty gritty methodology details are shown in the thread that's linked in the the OP there).

The images of lists/spreadsheets have apparently stopped working, but you get some idea of some relevant "longevity giants" within the other posts (check out post #6, for example). The gist is that you'll likely intuit is that 12 years of prime Reggie or Pierce would provide equal or [likely] greater championship odds [in a vacuum] than 4 years of prime Reed.

Well, I get that and I respect that. The issue is it's REALLY hard to win a title with a stockton level guy as your best player, who based on what we see was 29 on his list. You need a guy a tier above to WIN, and that's why I think these types of projections over value good player and under value stars. It is much like why I think WS is a great stat, but horrible for a top 50 list. VORP has some real flaws but at least it is scaled "better" for showing who was great and who was good. It just imo still under values greatness.

The fun of seeing how the 76ers were "rebuilding" is about this whole idea, that you need a franchise player to win, and you should tank until you have that guy to build around.

I don't think Miller or Pierce are guys you stop tanking once you get.

Moving this here because it's a meta thought as I try and follow along with the threads:

First, the linked research is precisely designed to reward stars for their nonlinear improvement over secondary players. What in the data is undervalued?

538, Kevin Pelton and Nylon calculus have independently looked at this and come up with nearly identical results. I actually think my calculator is the most rewarding of superstars. The intuition I believe that trips many people up is they think you only get to have one good player on your team, when in reality you win 95% of the time with multiple stars. So...

In a GOAT list, if one valued all the "No. 1s" first they would list all peaks above some arbitrary cutoff. (Kind of makes it a peak list, no?) That's the thing though -- the cutoff is arbitrary. I get trying to say "at a certain point a team is unlikely to win if a +3 offensive player is its best player." But don't forget the counterbalance: A team is unlikely to win if its best player is a +5 player and it has no other good players! The championships added from a +3 don't usually come from being the best guy on the team, they come from how much a player like that helps the best guy on the team. (And yes, sometimes he can be the best offensive player if that best guy is a defensively oriented big. Mr. Ginobili says hello.)

You can go through all of the top-shelf teams (basically Final Four teams) since the merger and you'll see this trend. And that's the data set used to create the championships added calculator.

[dhsilv2]

I'd refer you to this study by Elgee (I guess some of the nitty gritty methodology details are shown in the thread that's linked in the the OP there).

The images of lists/spreadsheets have apparently stopped working, but you get some idea of some relevant "longevity giants" within the other posts (check out post #6, for example). The gist is that you'll likely intuit is that 12 years of prime Reggie or Pierce would provide equal or [likely] greater championship odds [in a vacuum] than 4 years of prime Reed.

Well, I get that and I respect that. The issue is it's REALLY hard to win a title with a stockton level guy as your best player, who based on what we see was 29 on his list. You need a guy a tier above to WIN, and that's why I think these types of projections over value good player and under value stars. It is much like why I think WS is a great stat, but horrible for a top 50 list. VORP has some real flaws but at least it is scaled "better" for showing who was great and who was good. It just imo still under values greatness.

The fun of seeing how the 76ers were "rebuilding" is about this whole idea, that you need a franchise player to win, and you should tank until you have that guy to build around.

I don't think Miller or Pierce are guys you stop tanking once you get.

Moving this here because it's a meta thought as I try and follow along with the threads:

First, the linked research is precisely designed to reward stars for their nonlinear improvement over secondary players. What in the data is undervalued?

538, Kevin Pelton and Nylon calculus have independently looked at this and come up with nearly identical results. I actually think my calculator is the most rewarding of superstars. The intuition I believe that trips many people up is they think you only get to have one good player on your team, when in reality you win 95% of the time with multiple stars. So...

In a GOAT list, if one valued all the "No. 1s" first they would list all peaks above some arbitrary cutoff. (Kind of makes it a peak list, no?) That's the thing though -- the cutoff is arbitrary. I get trying to say "at a certain point a team is unlikely to win if a +3 offensive player is its best player." But don't forget the counterbalance: A team is unlikely to win if its best player is a +5 player and it has no other good players! The championships added from a +3 don't usually come from being the best guy on the team, they come from how much a player like that helps the best guy on the team. (And yes, sometimes he can be the best offensive player if that best guy is a defensively oriented big. Mr. Ginobili says hello.)

You can go through all of the top-shelf teams (basically Final Four teams) since the merger and you'll see this trend. And that's the data set used to create the championships added calculator.

The hardest part of these lists is always when/where you start doing cutoffs. Some people use the stats, some awards, some rings, some eye test, and then there's always some kind of adjustment.

Thus my view that neither of these guys were true franchise cornerstone players who bring teams titles, though I think pierce was almost there. But I'm not sure how to quantify that or even explain it fully.

But great post and glad you put it here and made it originally.

[iggymcfrack]

Hey, so I've only been on the board a couple weeks but I've been active in the discussions on this list, and I'd really appreciate it if I could get on the voting panel at some point. I feel like I would maintain interest all the way down to the end.

[penbeast0]

Hey, so I've only been on the board a couple weeks but I've been active in the discussions on this list, and I'd really appreciate it if I could get on the voting panel at some point. I feel like I would maintain interest all the way down to the end.

It is Trex's show but come on in and make reasoned arguments and respond to posts. It's more about the discussion than the actual placing on the list anyway.

[trex_8063]

Hey, so I've only been on the board a couple weeks but I've been active in the discussions on this list, and I'd really appreciate it if I could get on the voting panel at some point. I feel like I would maintain interest all the way down to the end.

Hey there. Thank you for your interest, I've appreciated your participation in the last couple threads (and we could certainly use all the interest we can get as we go into the back-half of the project). As penbeast0 said: continue participating, even state who your picks would be [even if you don't have a counted vote] and your reasoning (as this helps us get a handle on both how knowledgeable you appear, as well as your ability to make mature argumentation). If things continue to go well and there are no objections from the other mods, I'll add you to the voter panel, probably somewhere in the next thread or two.

[trex_8063]

Hey, so I've only been on the board a couple weeks but I've been active in the discussions on this list, and I'd really appreciate it if I could get on the voting panel at some point. I feel like I would maintain interest all the way down to the end.

I'm adding you to the voter panel as of #51. If you haven't already done so, read thru the OP and shoot any questions my way (as we've had need to change the voting protocol more than once, so it can be confusing; also one of the visual links stopped working).

[KempwiththeDunk]

We need some people older than 20 to vote. I see a lot on the list that are just players that young people just like watching but that were really not All Time Great Legends.

I'd like to formally request permission to vote.

[drza]

(placeholder)

[TheProfessor]

Why do Walton and Ming never make the top 100, Walton probably has a top 20 peak. Yao is comparable to the likes of mourning and dike interms of impact.

[Owly]

Why do Walton and Ming never make the top 100, Walton probably has a top 20 peak. Yao is comparable to the likes of mourning and dike interms of impact.

You are aware that if you take part (seriously and politely) in the debates for a bit, after a short period your votes would be counted?

Depending on what type of list it is, I'd disagree that Walton "never" makes top 100s and could probably cite a dozen which he has; perhaps theres a specific type of list you mean.

Regarding Walton - accepting, for the purposes of this debate, the premise that he had a top 20 peak (some go quite a lot higher, though he's not there by the boxscore ... as I say accepting that for now) ... where's the longevity. Even assuming we're talking longevity high quality (guys with circa 40000 RS minutes like Clifford Robinson, Buck Williams, Charles Oakley, Otis Thorpe, Andre Miller, Mark Jackson didn't make make the last 100 list done here - so it's not just longevity), even assuming additional value is added by high peaks, and even assuming that most won't take Penbeast's angle of Walton's salary actively damaging your teams title odds ... where does Walton rank in terms of quality production (or impact x longevity, or title probability added or however you might frame it)? If you feel there's a compelling case for top 100 then make it.

With Yao, otoh, again I would think the problem is one of minutes, but of course, if you have a serious case then let it be heard.

[edited to finish incomplete sentence]

[penbeast0]

We need some people older than 20 to vote. I see a lot on the list that are just players that young people just like watching but that were really not All Time Great Legends.

I'd like to formally request permission to vote.

As Owly said, come and debate, participate, bring your point of view and your supporting evidence to the project. We would love to have some good new voices join in and if you are supporting your views and avoiding the sort of baiting attacks that hurt debate, we will take you off probation and you'll be one of the arbiters of greatness until the next time we do this project.

PS, if you didn't know, I'm older than 20 . . .

[janmagn]

We need some people older than 20 to vote. I see a lot on the list that are just players that young people just like watching but that were really not All Time Great Legends.

I'd like to formally request permission to vote.

It's not everything about the age. I'm 17 and I think Walton should definitely be here

Lähetetty minun LG-M250 laitteesta Tapatalkilla