Posted by Neil Paine on December 16, 2009
Whenever Hall of Fame arguments come up, especially in baseball, I have a tendency to tune out from the sheer tediousness of the typical debate. On one side, there's always an arrogant guy who saw many of Player X's games and "knows" he's a Hall of Famer, so he cites other, lesser players who are already in the Hall (as though that were somehow evidence Player X should be in), brings up a couple of memorable career moments, and generally fudges on borderline issues to make the player seem better than he actually was. On the opposing side, another equally narcissistic guy splits hairs about the "magic numbers" Player X failed to reach, denigrates his career because A) if he won titles, he didn't have enough individual honors; or B) if he had a lot of individual honors, he didn't win enough titles. Throw in a few unsubstantiated jabs at Player X's character and/or manhood, and then start the whole process over again -- how fun.
I don't know about you, but I'm sick of these ceaselessly inane arguments. Call me a purist, but when it comes to the subject of the Hall of Fame -- and this is the case with every sport (except maybe the Pro Football HoF) -- I think it's vastly overcrowded with players whose inclusion does nothing but lower the standards of induction for future generations and cheapen the accomplishments of the players enshrined before them. For instance, in our hypothetical little talk-radio example above, Player X is almost certainly not deserving of enshrinement, because if you have to stretch the numbers to make a case, it's not a good one. Similarly, if you have to rely on subjective anecdotes and opinions to build your case, your guy doesn't deserve to be in. In other words, I believe "borderline" candidates shouldn't be borderline at all -- the answer is simply "no".
But how do you enforce such tough standards after years of leniency? I mean, all of the major Halls of Fame already opened up Pandora's Box many decades ago by allowing lesser players into their ranks, forever creating problems for future voting generations by setting the minimum standard for induction ridiculously low. That's why I think it's important to set aside a certain subset of Hall of Famers as "Inner-Circle" guys, players who have established themselves as the absolute elite of the elite and therefore deserve to be rewarded with an exclusive place in the HoF, one distinct and more prestigious than the one inhabited by "ordinary" Hall members.
Obviously, in order to do this you have to set a cutoff, a point of quality below which no player will be allowed to enter the inner circle. The question is always, where is that cutoff? And why is it there? Today I'm going to go through the method by which I think players' Inner-Circle worthiness should be judged, and hopefully provide some rationale into what level of exclusivity is right for such a high honor.
Above all else, the recognition has to be made that this is a resetting the old standards, an effort to push them sky-high. One basic, non-quantifiable premise is this: if you have to think about a player's candidacy, even for a second, he shouldn't be there -- the Inner Circle is a place for no-doubters only. Of course, your conception of a "no-doubter" might be different from mine -- I'm sure you can find people who don't have any reservations about Latrell Sprewell's candidacy if you look hard enough, but that doesn't mean the majority of the basketball world would agree. So you do have to establish a consensus in this process, and apply some kind of objective set of standards to players' qualifications, lest we end up with somebody like Nick Van Exel as an Inner-Circle Hall of Famer.
I think the main standard we can probably agree on is this: Was the player considered one of (if not the) best players in pro basketball? To me at least, it's obvious -- if, in your prime, there was a question as to whether you were one of the best players in the league, you don't deserve to be in the Inner Circle. Also, a player had to have maintained that dominant level for an extended period of time: high-peak-but-short-career guys or fluky one-year wonders need not apply, because I think it's safe to say that the Inner Circle is reserved only for ultra-high-peak guys who stayed that way for a good decade. I feel like the Hall of Fame sometimes tries to balance short, brilliant careers with long, consistent ones; well, the Inner Circle should only be composed of players who had long, consistently brilliant careers.
To quantify said brilliance, it's useful at this point to establish how I think greatness should be defined for Inner Circle purposes. Basically, in the broadest sense I think any assessment like this has to strike a balance between considering how a player was regarded in his day and what the numbers say about him after the fact. To achieve this goal, then, one must consider players in two areas: awards from the media/coaches, and statistical accomplishments. A ranking that combines both of these elements is calculated as follows:
- "Media" Points: Three awards are used to calculate this total: MVP voting, All-NBA teams, and All-Star selections. For the MVP, the top 5-12 players are ranked according to their MVP votes; the remaining players receive an average rank based on the # of players who did not receive MVP consideration. For All-NBA Teams, players are tiered according to whether they received 1st-, 2nd-, or 3rd-Team honors. Each honoree receives a ranking based on the average rank of the player in their tier -- for instance, a 1st-Team All-NBA player in a year with 5 1st-Team selections would receive a rank of "3", since (1 + 2 + 3 + 4 + 5) / 5 = 3. And 2nd-team selection in the same year would receive a rank of "8", as (6 + 7 + ... + 10) / 5 = 8, and so forth. Again, the remaining players receive an average rank based on the # of players who did not receive All-NBA consideration. Finally, All-Star selections receive a ranking based on the average rank of an All-Star, in the same manner outlined above; the remaining players receive an average rank based on the # of players who did not receive All-Star consideration. Note that a player did not have to actually appear in the All-Star Game itself to receive points, but only needed to be voted to the roster. Media points are then determined by summing the player's rank in the 3 categories detailed above (MVP, All-NBA, All-Star), and ranking inversely by this sum. Points are given for each rank in relation to the number of players in the league; for example, a league with 445 players would award 445 points to the #1-ranked player, 444 to the 2nd-ranked player, etc., all the way to 1 point for the 445th-ranked player.
- "Stats" Points:Three well-known advanced statistics are used in this area: Win Shares, Player Efficiency Rating, and, in lieu of Adjusted Plus/Minus, Statistical Plus/Minus (derived from regressing box score stats on Adjusted +/-). To account for the value of players who log many minutes, PER was converted to Estimated Wins Added (EWA) by the following formula: EWA = ((PER - 10.82) * MP) / 2010. Also, SPM was converted to a "value" metric by the following: VAL = ((SPM + 5) / 48) * MP. Since each metric has known flaws/biases, the median ranking among the 3 statistical categories was taken for each player in each season, and was then ranked inversely, with Stats points awarded for each rank in relation to the number of players in the league; for example, a league with 445 players would award 445 points to the #1-ranked player, 444 to the 2nd-ranked player, etc., all the way to 1 point for the 445th-ranked player.
- Composite Ranking: To encourage high marks in both the Media and Statistical categories, a composite point total was achieved by taking the geometric mean of each player's Media and Stats points. The resulting product is the composite total, which is then divided by the total number of possible points for a percentage score.
Still with me? If so, let's take last year as an example... This was the final MVP voting in 2008-09:
Those are already conveniently ranked for us, but here are the All-NBA Teams, which we have to "rank" ourselves using the method outlined above:
Finally, these were the All-Stars, which we also have to give a "ranking" to:
Math Alert! If you're curious, to determine the average ranking of a group of n players tied for rank r, you simply take: ((n * (2r + n - 1)) / 2) / n.
Adding the three rankings together and ranking the total inversely, you get this result for "Media Points":
Similarly, here are the rankings for "Stats Points":
|Player||Tm||SPM Rk||PER Rk||WS Rk||StatsPts|
Combine the two, and you get this list:
Every season, players are given a percentage score like that, based on how well they performed statistically and how they were regarded around the league. For the purposes of the Inner Circle, I used those numbers to form a career rankings list; to qualify for the Inner Circle list, a player must: A) have played 10 years professionally after 1951 in either the ABA or NBA; and B) have won at least 1 NBA Championship in their career. The second qualification is controversial -- and a difficult conclusion to come to, because I love John Stockton, Charles Barkley, Karl Malone, etc. -- but if this is truly the Inner Circle, you can't in good conscience include players with a glaring hole in their career resume, whether deservedly so or not. Qualified players were then ranked by an average of their 10 best seasons, and the Top 4 players whose careers came primarily in each decade were named to the Inner Circle. Friday, we'll begin with the Inner Circle members from the 1950s and 1960s, and continue going through the decades through New Year's (which, fittingly, marks the end of a decade itself). So stay tuned as we honor the absolute greatest pro basketball players of the NBA era, I'm sure you'll have a lot to say about the selections.