The Official Criteria Thread (2017-20)---PLEASE CONTRIBUTE if participating in top 100 project! (read OP)

[RSCD3_]

Combination of impact and versatility that questions how much a certain player over his career has provided in helping different teams win. Priority given to elevating good teams to great over mediocre teams to good and horrible teams to mediocre. I definitely care about injuries such as if you miss playoffs in a year because of them that reduces an amount of value. I see coasting as a lesser issue though so long as the team is comfortably in and said player has expected play in the playoffs based on their baseline level of play. Portability is also important to me but if the benefits of a play style are such that most teams would be able to get a lot of production without a decrease in too much of their teammates production I wont have as much of a problem with it compared to guys that need certain players around them or they wont be successful such as a 5 with poor rim protecting abilities that doesnt have any range.

[penbeast0]

Generally speaking, and without pretending to be terribly consistent about it, I use a combination of dominance in era with a discount for strength of era. I tend to rank the current era quite highly, but also the 60s which seem to me to be a much stronger era than the racially segregated 50s or the expansion weakened 70s. I often am a couple of years behind on rerating current players in my head wanting to be sure before annointing an Anthony Davis as the next great thing so some degree of anti-recency bias may creep in there too.

I value stats fairly strongly where available, more than eye test for offense. For players I never saw much, I use the best historical evidence I have read (stats, peers, contemporary analysis, etc.) and try to keep my mind more open as this evidence changes, something I don't do as easily for players I saw a lot of.

I tend to be more skeptical of my own favorite team (the Bullets/Wizards) and it's players rather than overrating them (I think, anyway); maybe because of the 35 years of mediocrity and misery that followed the team's one championship run where they were barely a .500 team that year in the regular season. I tend to rate defensive stars more highly relative to offensive stars than most people too.

[Ambrose]

Who knows if I'll participate or not (if I'm even allowed) but my criteria is pretty simple: Who gives me the best chance to win as many championships as possible over the span of their career, all other factors considered equal.

[Quotatious]

Important question that we have to answer, is how we weigh regular season and playoffs. I mean, there are some players who barely played in postseason (for reasons that they shouldn't be blamed for, basically bad supporting casts - let's say Oscar in Cincinnati, KG in Minnesota, Brand in Chicago/LA Clippers etc.), and even some guys who sustained their RS play in the postseason, individually, but had very little team success, such as CP3 and T-Mac.

Personally I weigh RS more heavily than most people, I value sustained All-NBA caliber play for a long time, and I tend to put a lot of emphasis on RS durability, which is why I'm high on Kareem, Karl Malone, Stockton, Duncan (I'm probably higher on Karl than anybody else except Narigo and Ronnymac - I just find it very impressive that Malone at age 39/40 was still averaging almost 21/8/5/2 stl on decent efficiency, playing 36.2 minutes for 81 games, for a playoff team in a tough conference - especially at power forward position, the competition was brutal - Duncan, Garnett, Nowitzki, Brand, Webber, R.Wallace, Pau Gasol, all played on rival teams in the Western conference that season, that's just an incredible amount of talent, that old Karl had to deal with).

Durability is a reason why a worse player may be more valuable - for instance I would rate Dwight's '09-'11 three-year stretch as better than Shaq's '96-'98 three-year stretch, because even though Shaq was better on a game-by-game basis, he missed 81 games in those three seasons, which is basically an entire extra season, while Dwight missed only 7 games in those three seasons. Shaq was healthy and really good in the playoffs in those years, but he wasn't flawless (had major problems in terms of pick & roll defense, which got exposed by Chicago in '96 and by Utah in '97 and '98), and Dwight was really good in postseason in those years, as well. So, I don't see any small edge that Shaq could have in the playoffs, being nearly enough to make up for Dwight's vastly superior RS durability.

[penbeast0]

Portability is also important to me

Do you only count portability forward or do you feel that players like Kobe Bryant should be significantly discounted because of his reliance on a 3 point shot that didn't exist for a large portion of NBA history?

Who knows if I'll participate or not (if I'm even allowed) but my criteria is pretty simple: Who gives me the best chance to win as many championships as possible over the span of their career, all other factors considered equal.

Does that mean Mikan is a top 10 player for you or do you discount for era?

Important question that we have to answer, is how we weigh regular season and playoffs.

And I count regular season and playoff games reasonably equally with only a slight to moderate bias toward playoffs which means that regular season numbers are more important . . . but I count contributing to playoff series wins a bit more (don't get more much credit for great losing performance than for great regular season performance) and contributing to NBA championships more than that. May be winners bias but winning is what the game is played for.

[feyki]

Individual impact.

Methodology could use some [a lot] of elaborating, fwiw.

First, adjusting every aspect of the game and set the league numbers to average(1.00 for example). Then calculating force of players at every facet(respectively and then collect) of the game based on plus-minus system.

Basically adjusting the league to average and How Players are positive or negative to the average.

For my formula details, I'd rank offence and defence equal. And On Offence, playmaking is the most important aspect; On Defence, rim protection is the most important aspect. Also, defensive rebounding is the least important thing on defence and volume scoring is not strong as people think.

I agree with penbeast on some points here. Before 50 players/legends couldn't rank fairly. Reading about themselves is best way to rank them. Names of like; Holman,Lapchik,Edwards,McDermott,Beckman.. Those legends deserve lots of things, but..

[trex_8063]

Portability is also important to me

Do you only count portability forward or do you feel that players like Kobe Bryant should be significantly discounted because of his reliance on a 3 point shot that didn't exist for a large portion of NBA history?

I could be mistaken, but I got the impression RSCD3_ was talking more about portability in a variety of roster situations within their own era. That's usually what I assume people are talking about when they say "portability" (because that's typically what I mean when I use that word). I otherwise specifically state "era portability" if I mean the other.

[trex_8063]

Individual impact.

Methodology could use some [a lot] of elaborating, fwiw.

First, adjusting every aspect of the game and set the league numbers to average(1.00 for example). Then calculating force of players at every facet(respectively and then collect) of the game based on plus-minus system.

Basically adjusting the league to average and How Players are positive or negative to the average.

For my formula details, I'd rank offence and defence equal. And On Offence, playmaking is the most important aspect; On Defence, rim protection is the most important aspect. Also, defensive rebounding is the least important thing on defence and volume scoring is not strong as people think.

I agree with penbeast on some points here. Before 50 players/legends couldn't rank fairly. Reading about themselves is best way to rank them. Names of like; Holman,Lapchik,Edwards,McDermott,Beckman.. Those legends deserve lots of things, but..

Are all eras on equal footing within your system?

[Ambrose]

Portability is also important to me

Do you only count portability forward or do you feel that players like Kobe Bryant should be significantly discounted because of his reliance on a 3 point shot that didn't exist for a large portion of NBA history?

Who knows if I'll participate or not (if I'm even allowed) but my criteria is pretty simple: Who gives me the best chance to win as many championships as possible over the span of their career, all other factors considered equal.

Does that mean Mikan is a top 10 player for you or do you discount for era?

Important question that we have to answer, is how we weigh regular season and playoffs.

And I count regular season and playoff games reasonably equally with only a slight to moderate bias toward playoffs which means that regular season numbers are more important . . . but I count contributing to playoff series wins a bit more (don't get more much credit for great losing performance than for great regular season performance) and contributing to NBA championships more than that. May be winners bias but winning is what the game is played for.

No, I do not. To keep it brief, he didn't have great longevity (especially compared to top 10 guys) and his production fell dramatically when they expanded the paint and he couldn't just sit there all day and wait for the ball under the hoop (which ended up being a large portion of his career). I don't have to go into his era to expose his dominance, though that does factor in to an extent.

[feyki]

Methodology could use some [a lot] of elaborating, fwiw.

First, adjusting every aspect of the game and set the league numbers to average(1.00 for example). Then calculating force of players at every facet(respectively and then collect) of the game based on plus-minus system.

Basically adjusting the league to average and How Players are positive or negative to the average.

For my formula details, I'd rank offence and defence equal. And On Offence, playmaking is the most important aspect; On Defence, rim protection is the most important aspect. Also, defensive rebounding is the least important thing on defence and volume scoring is not strong as people think.

I agree with penbeast on some points here. Before 50 players/legends couldn't rank fairly. Reading about themselves is best way to rank them. Names of like; Holman,Lapchik,Edwards,McDermott,Beckman.. Those legends deserve lots of things, but..

Are all eras on equal footing within your system?

Exactly.

[PCProductions]

I tend to follow the way ElGee has outlined his championship odds based on SRS impact formula. I then compute the sum of the player's expected championships and rank that way. This might make me a "longevity" guy by definition, but I think it's the fairest way to judge a player's career.

[drza]

Hmm. I've never been terribly formulaic, but I do think my criteria tend to be relatively consistent. Some things I consider, as a complete freestyle...

1) Impact. This is an attempt to estimate how much a given player is lifting his team, and can contain contributions from the eye test, boxscore stats, non-boxscore stats, team-based stats, etc. Of the quantitative approaches, the non-boxscore stats often are useful at pegging the ball-park of a player's impact, with boxscore stats often helping to explain (if needed) how that player made that impact. Context, repeatability, robustness in different circumstances...all are things that help tell the story of a player's impact, over periods of peak, prime and career.

2) Portability. This ties in a very direct way with impact, because a player almost always has X impact given his situation. So I see portability as a way to estimate how many different situations a player can exhibit his max impact, and similarly what that player's impact becomes when the situation isn't ideal. The best players in this category, then, would have a skillset and/or proven ability to generate very high impacts across a wide range of situations.

3) Scalability. This gets into the concepts I've seen mentioned in this thread of "best player on a bad team" vs "can lift a good team to great", and the whole scale thereof. Essentially, some players can have a massive impact on poor teams because they can do a lot of things, but if you put them on a good team where others were already going to be contributing in many areas, that main player's impact decreases. I don't have a formula for relating scalability to the other areas, but I do weigh it in. For instance, the other day there was a thread on whether someone would rather have 10 years of LeBron vs 14 years of Duncan. Despite LeBron having a larger impact than Duncan at their bests, I argued that Duncan was more scaleable to great teams than LeBron, and thus that this helped narrow the gap between how I see them.

4) Era. This is a hard one. I do consider it...for example, I don't rank Mikan based on his era dominance because the playing field wasn't level with other times in history. I also give a side-eye to some of the boxscore-based stats achievements of the 60s/70s due to pace. And I also give a side-eye to some of the dominance exhibited in the 70s because expansion and the ABA/NBA split watered down competition in BOTH leagues (e.g. I've seen it argued that at some point the level of competition of the ABA = level of competition of the NBA in the 70s...but if that's true, that means that BOTH leagues were watered down compared to other eras where all the talent was in one place). I do look at skill sets, athletic ability, etc. and try to give some thought as to whether that would dramatically change what a player could do in other eras. At the same time, I also give credit to innovators...Russell left a defensive template where none existed before him, Oscar and West were offensive geniuses in ways that while yes, maybe players in recent generations lifted the game even higher, they couldn't have done so without having the shoulders to stand on. So when someone argues that Russell couldn't have the same level of defense in an era with a 3-point shot (debatable), in my mind that is countered by the fact that he had the mental genius to take a game where it was considered bad form to jump on defense, and find the best ways to maximize impact in his system. I kind of feel like that same genius would let him maximize his impact, if he were in another era as well. But again, this is more something I consider than something formulaic I could plug in

5) Regular season vs playoffs. It's very interesting. I tend to weigh playoffs more than regular season...but playoffs are also noiser to be able to evaluate, so I require a high standard of evidence before I believe something to be a "postseason fact". Also, I've been increasingly outspoken that the boxscore-based advanced stat default approaches which (over) rely (to a dramatic degree) on individual scoring efficiency may just be the devil. So I need to see a LOT more than just a small change in individual true shooting percentage (which then translates to every boxscore-based advanced stat that basketball-reference has) before I consider something to be a playoff fact. But when everything is brought to the table and considered, especially to the level of detail often reached in these long-term RealGM studies, there's often enough facts laid out to get a good feel for things.

[rebirthoftheM]

individual offensive impact is my #1 criteria.

By offensive impact, I mean the level of impact a solitary player has on his team’s ability to put the ball into the hoop effectively. This can be the result of many factors- dominant scoring, dominant playmaking, gravity etc. I’d focus on primarily the player’s prime here, and take into account both reg season and playoff impact (longevity therefore implicitly effects this criteria). I’d give this particular category a 40% weighting in my GOAT rankings as I believe this the most impactful category a solitary player can have on a game.

Era does matter in the sense that I will be comparing offensive impact against offensive standards during the era the player played in. However rule changes for me should be compared against eras, meaning that if a player’s offensive impact is a bit muted due to rules in a particular era that limit his offensive productivity, I will not hold that against them. Conversely, if a player played in an era perfectly suited to his offensive talents, whilst in another era, he might not have that level impact, I might hold that against a certain player in comparisons with others.

In making the above determination about offensive impact, I’d consider:

• Eye test- see what kind of impact he is having in real time
• Offensive impact stats- If they are available
• Abilit to "carry" your team offensively and take over games offensively
• Team ORTG- This is a big factor, because dominant offensive players tend to have big impact here
• Offensive box score stats (though not as important as impact stats, but used as a “filter” mostly)
• Role on the team- Was the player a #1 or #2 option? How many minutes did he play? Was he a starter or a bench player?
• Offensive talent on the team
• Offensive coaching style- See the Mike D’antoni effect that can warp offensive play.

Criteria #2 for me would be defensive impact. I’d give it a 15% weighting because I don’t see the ceiling of defensive impact touching the ceiling of offensive impact. This particular category is also heavily biased to interior bigs.

Criteria #3 would be longevity. Amongst other things, I’d consider how long they were a superstar for, an all-star etc. I’d give this 10%

Criteria #4 would be peak impact and play. I’d give this 10%

Criteria #5 would be accolades and legacy. Blemishes on your legacy play a part here. I’d give this 10%

Criteria #6 would be skill and talent level, relevant to position on the court and physical endowments. I’d give this 10%. I imagine I weigh this far higher than others do, but I'm with the older timers here on this one.

Criteria #7 would be leadership skills (very subjective- depends on one’s preference of leadership). I’d give this 5%.

There are a number of filters I’d use however that would inform my rankings. I’ve mentioned box score stats, mpg and ones role on a team, as if you didn’t do this, you’d have a dude like Manu Ginobli being far higher than he should be.

I also have the era filter, which we discussed in another thread. This is a rule I apply particularly for players in the 50s and 60s. The filter is- would X player, if he was dropped into a 90s, or 2000s training camp, be a superstar that season? If the answer to this hypothetical question is “probably no” then this hurts their ATG rating. However, if the answer is yes (which for Wilt, is a big yes for me), this majorly helps their ATG rating in my eyes, and would inform my view of them when considering each criteria.


I should note here that some of criteria (some inherently) entail moderate/heavy subjective judgement. I recognise this is increasingly an unpopular approach to these discussions, because everyone and their dog has an opinion. However, there is an equal danger of turning basketball into an empirical science like physics, with only objective litmus tests (even in science, "objectivity" itself can be dangerous, as it can hide certain nefarious povs, but that's another whole discussion). Sports, and particularly here basketball, goes beyond pure data, and you need to supplement your data at every stage with subjective thoughts/thoughts of people around at the time, and then come out with the most reasonable conclusions.

There is no data for determination, heart, courage, will, momentum etc... things old timers used to hang their hat on. If we dismiss them because they cannot be objectively quantified, assuming truth can only be reached through empirical data, then you're missing out on the spirit of basketball, to the detriment of an appreciation of the game. IMO the subjectivity problems with a lot of these "unverifiable" categories of comparisons can be mitigated by combining different subjective thoughts, and seeing which thoughts have the most basis in terms of the eye test and stories we've heard around the league. It might have its problems still, but it ain't like all these metrics don't have issues also. They all have inbuilt assumptions that are self referential.

A perfect case study that ties into the above is LBJ's 2011 performance. Lebron, like all other ATGs has had bad series before. But to me, watching the games (e.g. lebron being ultra passive at crucial times/hanging around the baselines) and reading stories that came out afterwards (e.g. Wade handing the reigns to Lebron), and taking into account LBJ's subsequent play, suggest what occurred in that series with Lebron went far far beyond the Mavericks Strategy or LBJ's poor execution. It had more to do with LBJ and his mental psyche in the finals. This of course cannot be verified like PPG, but to me, the information we have available suggest it is likely so.

[kayess]

My criteria is all the inputs that go into answer the question "who gives me the greatest expected value of championships won over their career?", those inputs being:

1) Impact - A function of how good a player was and how long they sustained it. I use ElGee's SRS/championship odds as a general guide to calculate overall impact/value.
2) Skill set - Allows us to gauge how well this impact translate across different situations. Circumstances can vary greatly across a career, so analyzing a player's skillset allows us to approximate the effect of luck on a player's career - did it help or hurt his impact?

I look at two situations primarily: a) deadeye average cast (i.e., replacement level), and b) average contender's cast (which differs across positions and skill sets, of course, e.g., the average cast around a scoring, playmaking wing is not the same as around a defensive big). I don't look at GOAT level cast situations because at that point, you're going to need tons of luck anyway - at which point the impact of the player in question is far less important than the luck of getting a stacked roster.
3) Did they lead a franchise that had never won before to a title as the man? Even more important than the first two, tbh.

Not included (in most cases)
1) Era - too difficult to account for. Theoretically, athletes are always getting better, but you also can't punish people from older eras because you can only beat who's in front of you.
2) Regular season/playoffs split - I consider the playoffs a terrible way to judge who is best team, and therefore, the best players. The game doesn't fundamentally change , and match-ups, injuries, etc. matter far more than people would like to admit. In an ideal world, the championship is decided by a 2n round-robin tournament, equal home/away games, where every game matters equally. In our imperfect world, the regular season is the most statistically valid test of ability, but the effort is askew, and the playoffs is (nearly) the opposite of this: statistically invalid due to the biased sample, but effort is maximized. A factor only if the comparison is razor thin, and RS/PS performance can help inform us further.
3) Team achievements - should be self explanatory.

[trex_8063]

Spoiler:

So I've been asked about my criteria before, and I last posted it some years ago for someone who asked. I haven't posted it in recent threads because it's generally TL;DR for internet forums, and I'm not going to derail someone's thread with something that has nothing to do with the topic. (Generally, when I post about something to the degree to which I generally think about things, discussion ceases.)

But this thread is specifically asking about criteria, so it's appropriate to post. But first I'm going to preface it with the following:

This is MY criteria based on what I value. As such, it is not open to debate. Particularly, if someone has never taken the time to think about it and develop their own criteria, then I couldn't care less if they think a particular criterion isn't important or shouldn't be a factor. Make your own first. I have made revisions when posters brought up something I thought was a valid issue, and incorporated it into my criteria.

I aspired to replicability. In Sir Arthur Conan Doyle's detective stories, Sherlock Holmes would say, "You know my methods, Watson." Similarly, it should be possible for other people to run the criteria and come up with the same answer, provided they have the relevant information. Another RealGM poster actually has independently run my criteria for their favorite player as part of their campaign process during the last Top 100 project in a post specifically directed to me since I was the only one who had listed a complete set of criteria. But that's how, ideally, it should work. I shouldn't have to be the one who runs it, it should work regardless of who runs it if a good enough job has been done in formulating the criteria.

As I've repeated said during my 11 years as a registered member here and elsewhere (I've been PMed on 5 or 6 different boards about this since I've never published one anywhere), I do not have a GOAT list. I started one during the last Top 100 project, but I lost in during a computer crash, and I didn't care enough about it to re-do it. Additionally, just as I record what happens in basketball in the moment, I couldn't recreate my mindset in the particular moment I was working on it after the fact. I've used it for player vs. player comparisons, but to do a list would require running everyone through it, and order them in relation to each other, which is more time than I care to spend on it.

Any replies that don't foster constructive dialogue should not be made. Since I've explicitly made this clear before the fact, I won't be sympathetic if this is ignored.

1. The ability to integrate oneself and whatever respective abilities one brings to the table with the rest of the players on one’s team in order to enhance the whole for the facilitation of the ultimate objective of winning, and the dedication to employ these abilities for the effectuation of said purpose. An NBA game is not a game of AND1, so who would do what against another player one-on-one—as if they were the only two players on the court in an empty gym—is irrelevant.

The means by which a player helps his team are inconsequential. What is important is the end. The player in question should use whatever skills he brings to the table to help his team win. As different players have different abilities, the means employed will vary. The only thing that matters is results. How well what a player brings to the table translates into victories for his team. No one way of helping one’s team is inherently valued more than another.

My intent behind this criterion was to eliminate style-of-play bias. It's irrelevant whether or not I like the way they play. What matters is that they use whatever it is that they do to help their team win.

There have also been arguments about whether certain styles of play are as impactful as other, more conventionally accepted styles. I only care if it works. "Works" = "helps his team win."

2. The ability to both identify what the team needs at any given moment in order to realize the ultimate object of winning and provide it. Players who can do this will rank higher than those who can’t figure out what needs to be done without being told.

There are occasions where exploiting a favorable matchup is what needs to be done to win, and it might not always be you. Some non-playing spectators penalize a player for taking advantage of the hot hand if that hand isn't their own, because it isn't "alpha" enough.

3. The possession of the rational self-interest to put ego aside in order to do #1 and #2, disregarding the opinions of irrelevant others who are not on the team and thus have no effect on the team’s success.

Opinions of anonymous people on the internet who aren't playing are irrelevant, since they have zero effect on the team's success. Neither do the opinions of members of the media if they're not doing anything to help your team win, and what you're doing is.

4. The ability to block out distractions and anything irrelevant to the maximization of the team’s chances of victory.

A player focusing on anything other than helping his team will receive a lower evaluation. Basketball is a job like any other, and a player’s job is to help bring his team wins, just as a salesman’s job is to make sales for his company. Nothing else matters or is relevant. A basketball player has more impact on the outcome of a basketball game than a baseball player does on the outcome of a baseball game, or an American football player does on the outcome of a game of American football. A baseball player is one of nine players on the field of play, and position players only come to bat 3-4 times during the course of a three-hour game. A pitcher has the most impact on a baseball game, but only pitches once every five games. An American football player is one of 11 players on the field of play, and there are entirely separate teams for offense and defense. American football is the most specialized of the major sports—which limits the impact an individual player can have, and for half of the game an American football player has no impact on the game whatsoever—Chuck Bednarik, who played with the Philadelphia Eagles from 1949 to 1962, was the last American football player to play on both offense and defense. A basketball player however, is one of five players on the field of play, and can be involved in everything that occurs on a basketball court on both offense and defense. Since a basketball player has more impact on the outcome of a basketball game than it is possible for a player to have in any other of the major sports, helping his team win carries more importance for a basketball player. Thus comparing basketball players to players in other sports is flawed and reveals a lack of understanding of the varying natures of the sports.

Basketball players are grown men who make choices. They have the right to make whatever choice they want, but with action comes consequence. That choice they make will be honored and they will be evaluated on the basis of that choice, whether it’s beneficial or detrimental to their team’s chances of winning.

People make the job analogy on occasion, such as when they used it when Durant went to Golden State, saying it's no different from any other employee taking a better job offer. So let's go with it fully and not selectively.

5. Clutchness. The ability to rise to the occasion during big games and crucial moments in order to bring about the ultimate objective of winning, and the mental fortitude to do so.

I may re-name it "rising to the occasion" or something similar.

6. Playoff Translatability. The ability of a player to continue to effectively employ whatever it is he brings to the table to help his team win during the postseason. The sole purpose of the regular season is to determine seeding for the postseason and playoff brackets.

I was inspired by the results of a study ElGee published as far as naming it. As stated in criterion #1, I don't care how a player helps his team win. But, whatever it is, it has to translate to the postseason. If you take a class, the final exam is weighted heavier than the rest of the semester/quarter. If you don't do well on the final exam, it doesn't matter how well you do prior to it. Bomb the final exam and you devastate your grade, even though the rest of the semester/quarter is longer than the final exam.

7. Statistics. Statistics are team-dependent. Doing what is needed in order for the team to win may require sacrificing individual statistics. There will be no penalty levied for doing so, nor will a player’s evaluation be lowered for putting the needs of the team above his own individual statistics. It shows he has the right priority.

Some people don't understand how "individual" statistics may be team-dependent. David Robinson scoring 71 on the last day of the season to win the scoring title was a team achievement. People who don't understand how are precisely the people I'm talking about. When Kevin Garnett and Ray Allen went to the Celtics in 2008, Garnett went from 22.4/12.8/4.1 to 18.8/9.2/3.4, Allen went from 26.4/4.5/4.1 to 17.4/3.7/3.1, and Paul Pierce went from 25.0/5.9/4.1 to 19.6/5.1/4.5. It wasn't because they suddenly got worse. Their statistics decreased due to the team they went to. And they won. In Year One. Context has to be considered.

8. Rings. Rings are only relevant so far as the player’s contribution to his team winning the title that year. Mitch Richmond won an NBA championship as a member of the Los Angeles Lakers in 2002, but played all of four minutes that postseason. Thus, the ring that he won is as irrelevant as he was to the Lakers that year. He gets no boost against a ringless player. Neither does a player who bandwagons his way to a ring.

Obligatory, since some people make dumb statements based on rings.

9. Individual Contribution. The only thing of relevance is how a player helps his team win, which means the player in question’s performance will be evaluated. If that player has a poor performance and another player picks up the slack to help his team win, then that player receives no bonus for his teammate bailing him out. Conversely, just as a doctor can try to the best of his ability to help keep a patient alive but fail, so can a player try to the best of his ability to help his team win but ultimately fail. His individual performance will be assessed, and if he didn’t help his team lose, he will incur no penalty. However, if he was instrumental in his own team’s defeat, he will be penalized accordingly.

One poster here said my criteria seemed to emphasize winning, and asked if a player who played well but lost would be penalized. I replied, "No." This addresses that. As per the previous criterion, rings are only relevant so far as the player's contribution to his team winning. A player's contribution will be assessed and judged accordingly.

10. Awards and Accolades. The object of the game is to help your team win. Awards are extraneous to this objective. A trophy has never once stepped onto a court to help a team win a game. Awards are not needed in order to know how much a player helped his team win.

Awards have nothing to do with how well a player played, as a player’s performance stands independently of whether or not he received an award for it. A player’s performance doesn’t magically get any better for receiving an award for it, nor does it magically get worse for not receiving one. It only matters for people incapable of looking at how a player played and forming an opinion from that.

Another example is All-Star selections. There are only 12 spots available, so not everyone who plays All-Star caliber ball will make it onto the team. Every season there are more players playing at an All-Star level than there are roster spots on the All-Star team to accommodate them. There are snubs and omissions every year. And the starting lineup is literally a popularity contest. The fans vote on the starters, meaning that some players are guaranteed to get in every year on popularity alone whether they deserve it or not, thus reducing the number of spots available for players whose performance actually warrants it. Therefore, what is important as far as that goes is whether a player played at an All-Star level during a given season, not whether he was selected to a team with limited spots from which deserving players will always be excluded. One of the questions on Bill James’ Keltner List to assess whether a player is deserving of induction into the Baseball Hall of Fame is: “How many All-Star-type seasons did he have? How many All-Star games did he play in?” Every player having an All-Star-type season in any given year won’t actually be selected to the team, and it's lauded that James took this into consideration when he formulated the questions.

Additionally, All-Star selections only cover the first half of the season. Presumably—since over half the season has been played at that point, that’s enough of a sample size that one should be able to assume they’ll continue to maintain the same level of play through the season half of the season, but it’s also possible to play poorly the rest of the way. Case in point is 2013-14 Roy Hibbert. “Not since Chuck Knoblauch has a professional player so good deteriorated so quickly,” Chris Mannix wrote on SI.com after Game 6 of the 2014 Eastern Conference first round series between the Indiana Pacers and Atlanta Hawks in which Hibbert had four fouls, zero points and two rebounds. “In the first half of the season, Hibbert averaged 11.8 points and 7.7 rebounds, shooting 46.4 percent from the floor. In the second half, Hibbert’s numbers dipped to 8.9 points and 4.7 rebounds, shooting 39 percent from the field” (Chris Mannix, “Despite Roy Hibbert’s woes, Pacers force Game 7 against Hawks.” SI.com. 2 May 2014. http://www.si.com/nba/2014/05/02/indiana-pacers-win-game-6-atlanta-hawks-nba-playoffs#). The All-Star selection Hibbert received during the first half of the season is not indicative of his play for the entire season, which serves to illustrate that an All-Star selection is not needed in order to determine whether a player played at an All-Star level. The latter is more important than the former. In other words, performance > awards.

I've talked about this criterion before, and ronnymac2 brought up Roy Hibbert, which I liked, so I incorporated it into the criteria.

I don't like how awards have become a shortcut for thinking. The previous criterion concerned individual contribution. That can be assessed without needing awards. I care about how a player played, not whether he won an award for it or not. I also don't like how it's used as a technicality, where people say, "Player X didn't play with an All-Star" in Z year, where they're only looking at whether or not they received an accolade, not whether or not they played like an all-star, which would be the pertinent question. As said in the criterion, all the players who play like an All-Star in a given year will never receive an accolade for it, in any year. It doesn't make sense to me to not acknowledge their play simply because they didn't get anything for it (though that's something that one couldn't just look up on basketball-reference). What awards and accolades a player won are a matter of record, but they don't factor in my rankings. Not every player had the opportunity to win the same awards, so some players would gain an automatic advantage in that regard. I want the same standard across the board.

I was also disappointed by the results of a test question I posed here on RealGM some time ago, which served to strengthen this criteria for me, and illustrate that I was correct to include it.

11. Playoff Advancement. The object of the game is to help your team win. In lieu of actually achieving that objective, helping your team get as close to it as possible. Helping your team get to the semifinals > losing in the opening round; helping your team get to the conference finals > losing in the semifinals; helping your team get to the Finals > losing in the conference finals. Getting closer to the ultimate goal of winning is always a positive. Finishing farther away from it is always a negative. Helping your team get to the Finals but losing is always better than losing in an earlier round.

From what I've read on internet basketball forums, I felt it was necessary to explicitly include this into my criteria. It's irrational to me how an early elimination can be perceived to be better than deep advancement. And I had to actually use my own criteria on the Brady vs. Montana debate that used to go on, which is the entire point. Montana was my all-time favorite quarterback, but the entire point of having criteria is that they objectively apply across the board to everyone. The fact that it put someone I had no sentiment for over someone I did illustrated that it worked. Many people simply change the criteria or make an exception when a player they like is involved.

12. Longevity vs. Peak. The object of the game is to help your team win. Nothing else matters. Thus longevity is only relevant as far as when evaluating a player, the question is: how much did that player help the team(s) he played on during his career win, from draft day to retirement? This encompasses more than just a player’s peak/prime, it encompasses the moment he plays his first NBA game to the moment he announces his retirement, not an arbitrarily selected portion of his career. A player can help his team win before reaching his peak/prime (e.g., Magic Johnson), and can continue to do so after passing it (e.g., Kareem Abdul-Jabbar and Tim Duncan). These years will not be excluded simply because they didn’t fall inside the period labeled as that player’s peak/prime.

A player does not cease to help his team win after passing his peak/prime. He may not be able to make as large a contribution as he formerly did due to age, but continuing to contribute to team wins to the extent one is able is still valuable to the team he plays for and helps the team obtain the ultimate objective. A player’s career consists of more than just his peak. The mere fact that one player “peaked” higher than another at one point in his career does not mean that he helped the team(s) he played for win more from draft day to retirement.

Longevity only has any meaning insofar as the length of time a player can continue to effectively employ whatever skills he brings to the table at whatever degree he is able to at that point in time to remain a positive contributor to team success. Post-prime longevity only matters when adding extra value. That is to say, if a player failed to effectively employ whatever abilities he brings to the table to help his team win during his prime, then simply outlasting the competition long enough to luck into a favorable situation is not adding extra value. Post-prime longevity cannot make up for the failure to meet Criteria #1 and #5 during one’s prime. Only seasons in which a player helped his team win will be considered in the overall evaluation (Criterion #8).

This was necessary for me to codify based on what I've read on internet basketball forums, which is irrational.

Player X only has more titles than Player Y due to his supporting cast.

This is a common argument made on various basketball boards.

Yet, once a player declines from what they used to be, but is still at least as good as those members of the supporting cast which was given as a reason why he had more championships than another player, suddenly he's a "role player" and his contributions don't count because he wasn't "The Man" anymore.

Which is it?

People can't have it both ways. A player gets discredited for what he did because of some players he had around him, which is "the reason his team won," yet when he's no longer himself but still at least as good as some of those supporting players who were "the reason his team won" when he was in his prime, he's discredited again. It's a double standard. So it's something I felt the need to address in some way.

That's what it consists of since my last revision. I haven't seen anything since then that's caused me to think an addition is in order. "Luck" was brought up recently, but I already account for what a player does to help his team. I don't have a problem discussing any of it, as I've specifically incorporated some things posters here have commented on, but, again, if you haven't sat down and considered your own criteria, then I don't particularly care about anyone telling me how wrong my criteria are if they don't even have any themselves.

Items 1-6, as well as 9, all appear to---in some direct or indirect way---deal with the principle of impact [on team's success]. To some degree these criterion appear to break impact down into various aspects or means of being impactful (recognizing what your team needs, avoiding distraction and essentially not being a diva, "rising to the occasion", etc); as well as taking into account portability of impact among different casts/contexts, how well it translates into the playoffs, etc.......but all essentially relating to impact.

Noting "statistics" is its own separate item (#7), my question is wrt how you assess #1-6 and #9. Do you use a primary scouting type approach (i.e. eye-test), or do you employ various impact indicators (RAPM, on/off and WOWY studies, etc; which I assumed were exempt from #7, as your description appeared [to me] to be referring specifically to box and box-based stats)? Or perhaps the former partially informed by the latter?

Also, do you employ any "anecdotal" evidence for these things? I'm referring to things such as reputation, things you've read in books, statements by their professional peers, etc (this obviously applies more to older era players, for whom both more tangible impact indicators as well as game footage [often] is lacking).

[Drylick]

Yeah, I would like to add that I want my GOAT to be the one that I can trust to put the ball in the hoop, or win the game for me, if my life is on the line.

Heart, determination, and intangibles count. Though there's really no way we can quantify it, so 'eye tests'.

[SideshowBob]

I passed on the last two Top 100 projects because I knew I would not be consistently dedicate time once we moved past the top 25-50. Additionally, I cannot make the claim that my level of research has been consistently thorough enough for enough players to feel comfortable filling out a top 100 list. As such, I will not yet commit to participating in this project either.

That said, I will present a baseline by which I do my player evaluation in the case that I change my mind come summer. Full disclaimer: I would not consider my list a GOAT list - it instead is my subjective (but attempt at objective) list of cumulative career value as defined by the following criteria (which is the same I used for the peaks project - except obviously scaled to a full career instead of a single season):

Criteria

My criteria for peak focuses on skillset/ability. What I'm concerned with is evaluating how much I believe player X improves the odds of any random team's title chances (I typically use the SRS/portability scale for this as Elgee has done/provided research on how much players can shift the needle).

Evaluation Period/Health

The in-season time-frame I like to evaluate/weight the most is the late RS and the postseason (though I don't penalize players for missing the playoffs, unless they are unable to play due to health). RS health (missed time) does not concern me too much, but if health has a negative effect on level of play during that crucial time-frame, I do account for that.

Era

Kind of throws me off. I want to stick to worrying about what players did within their era. Trying to consider them transferred across eras seems like a logistical nightmare.

Portability

A good player lifts a team's overall level of play. The better the team plays the better the chance at a title, thus I prefer skillsets that mesh well on already talented/well-built teams, or skills that tend to avoid easy overlap (defense as a whole is by nature is additive, on offense OTOH more particular skills are preferable). I use a 5 point scale, -2, -1, 0, 1, 2, with -2 being the least portable and 2 being the most. Ideally, in the future I'd like to split this up for off/def as even now I think there's room for more nuance in the differences, and I think I could be a bit more precise. There are some situations where I may have preference for a less portable player at the same listed SRS level as well - this is usually because of the limitation of using 0.25 increments.

Spoiler:

Ballot (Top 30)

An * indicates that I have multiple seasons of this player at the listed SIO level, and have gone with a year of preference.

Code: Select all

Rk.   Player   Season   SIO   Health   Port

1.    Shaq     2000    +8.00  100%     1
2.    Jordan   1991*   +8.00  100%    -1
3.    James    2013*   +8.00  95%      1
4.    Bird     1986    +7.25  100%     2
5.    Hakeem   1993*   +7.25  100%     2
6.    Wilt     1967    +7.25  100%     1
7.    Russell  1964*   +7.00  100%     2
8.    Duncan   2002    +7.00  100%     2
9.    Garnett  2004*   +7.00  100%     2
10.   Walton   1977    +7.25  80%      2



Code: Select all

11.   Robinson 1994    +6.75  100%     2
12.   Magic    1987    +6.75  100%    -1
13.   Jabbar   1977    +6.50  100%     1
14.   Erving   1976    +6.75  100%     0
15.   Curry    2015    +6.25  100%     2
16.   Wade     2009    +6.25  95%      0
17.   West     1968    +6.75  60%      1
21.   Dirk     2006*   +5.75  100%     1
19.   Barkley  1990    +5.75  100%     0
20.   Oscar    1964    +5.75  95%      1



Code: Select all

21.   Kobe     2008*   +5.75  100%    -1
22.   Paul     2008    +5.75  100%     0
23.   Durant   2014    +5.50  100%     1
24.   Ewing    1990    +5.50  100%     1
25.   Malone   1998*   +5.50  100%     0
26.   Nash     2007    +5.50  95%     -1
27.   McGrady  2003    +5.50  90%     -1
28.   Moses    1982    +5.25  100%     1
29.   GOAT     2015    +5.50  80%     -1
30.   Mourning 2000    +5.00  95%      2


*Pretty undecided on Dirk, I'm higher on his offense in later years but I don't know that that makes up for the defense earlier.

**Ballot not set in stone by any means. There are a lot of places where I'm iffy, particularly the 5.5-5.75 area, that group is a mess, (Durant could be the cream of the crop here, but I've cooled on him with some separation).

I'll also eventually try to expand out to 50 if I have the time.

[Colbinii]

When the last Top 100 project was being run I didn't feel like I had enough to contribute as my knowledge for the game, specifically pre-2000's was extremely lackluster. I knew the names, but I didn't know the players. Currently I feel like my ability to access greatness has reached a point where I can be a positive contributor in such a project (though not one of the better posters, that is for sure).

I don't have a criteria that is set in stone, and unlike a few posters here, I don't have a "formula" where I input X and out comes a value Y. That being said, I do weigh certain aspects of the game more so than other.

Prime Impact: This is the biggest for me. How much impact did a player have during his prime? This gets difficult because Player X may have had a 10 year prime, but Player Y only had an 8 year prime while at a slightly higher level of impact. For the most part separation is easy to see based on team results and player performance, but for some cases it gets very difficult to find separation, if any separation does even exist.

Portability: How can a players skill set and ability to adapt to different situations raise a teams ceiling? Since the goal is to win championships, I give more credit to players who can raise a teams ceiling than raising a teams floor. I look at what a player actually did first, including winning with multiple franchises/teams/coaches/rosters along with how a player transformed his game to accompany his teammates strengths and cover up his teammates weaknesses. Then, I look at a players skill set in a vacuum and use hypothetical situations of different teams in a similar era and try and predict the outcome of what would happen if Player X and Player Y swapped places (assuming Player X and Player Y were of similar caliber; swapping LeBron James for Tony Allen on this years Memphis team isn't showing me anything).

These are far and away my two main criteria when evaluating players. I don't care much for Era to Era comparisons, as I want to know what a player did in his Era. I don't weigh accolades highly, but I do take into consideration longevity (as long as the player is having a positive impact).

[eminence]

Career Impact - I generally have a pretty high amount of faith in our regression based impact stats (RAPM back to '96, less faith in WOWY before that). It is not however taken as a gospel for ranking players, but rather to inform my own opinions of what skillsets lead to high 'impact'. This can get tricky in that different skillsets may be more or less effective in different eras/team situations, which leads to a portability factor within 'impact'. Generally speaking off-ball offense and defense based skillsets score the highest in terms of portability. Box-score numbers are generally used to inform me what a players skillset is, and less for how good they were. I do not currently try to quantify this 'impact', but split players more into tiers and I suppose could come up with orders within those tiers if pushed. Don't really have a preference for offense vs defense, but would say I have a bit of a threshold where a player can become an 'anchor' on either offense or defense and those players get a boost for that.

Playoff Performance - Playing well in the playoffs (emphasis on playing - sorry injury prone guys) is what wins titles in this league, so playoff performance is quite important to me. Regular season play obviously has it's positives as well - larger sample in pure number of games and in opponents faced (matching up poorly with one opponent shouldn't bury a guy who had a great season - Drob vs Hakeem).

Era Dominance - Being among the best in the world at what you do when you're doing it gets a boost from me. For the very early years this means factoring in segregation, then ABA/NBA talent dilution, and lack of international players until recently. Generally do believe basketball has been consistently improving, but it's a gradual upward curve and I'm not one to put average starters today over stars from the 50's.

And that's really the basics of it.

Edit: Small edit on my playoff injury policy after thinking about this for a bit. I'd kind of built my criteria in the past with a POY/peak approach in mind, so I heavily penalized any and all playoff missed time. But I think that makes less sense in a career context. I'd still say they are more damaging than regular season injuries, but not nearly as damaging in a career context as within the evaluation of a single season.

[Doctor MJ]

For me in a nutshell it's how impressed I am with what they accomplished for their teams.

Now let me break that down a little further.

First, there's on court impact. On first pass this is essentially what APM-style stats measure.

Second, there's durability & longevity, which I think are self-explanatory.

Third, there's the matter that some contexts making raw impact more or less impressive. Having huge impact on an elite team is more valuable than doing it on a weak team.

Fourth, there's the fact that it's not a player's fault that they don't end up in the perfect position to have as much impressive impact as possible. If a player isn't in a position that let's them max out their impressiveness some years, but they have other years where they've proven their capacity, I tend to let the unfortunate contexts slide.

Fifth, off-court impact matters. It's hard to quantify and there's no "right" way to do it, but anything a player does that helps his franchise matters, as does anything he does that hurts his franchise.

Last note on era: I don't treat accomplishment X the same in each era but I'm never utterly dismissive of an era.