Friday, June 24, 2016

Basketball Greats

I enjoy lists. I enjoy basketball. I enjoy greatness. So, for a long time, I have been trying to evaluate and rank the greatest NBA players ever. It's difficult because there is no single metric that will give a consistent or accurate result. I appreciate victories and championships, but basketball is a team sport and basing your rankings solely on championships will lead to outlandish conclusions like Horry was better than Jordan or Pippen was equal to Jordan or Karl Malone was worse than Steve Kerr... a lot worse. So, winning can't be the only factor. We can look at cumulative stats, but this weighs far too heavily on health and longevity, rather than tying a career to success or dominance. We can look at standard metrics like per game stats, but that does not factor in a lifetime of playing great basketball and does not bear sufficient resemblance to impact on the Court (especially on defense). So, we often look to advanced metrics, and some are proving valuable, like ESPN's Real Plus Minus (or Player Efficiency Rating) or Value Over Replacement Player. But there are no advanced metrics that are close to giving us a cross-generational ranking of the greatest players of all time. Even in advanced metrics, there are subjective elements (e.g. which data you choose to enter or how much you to choose to weigh that data), which makes it increasingly important to verify conclusions using identifiable markers (or great individual basketball players); in the post-merger era, if the list does not include Jordan, LeBron, Magic, Bird, or Shaq in the top 10, the stat probably requires adjustment. I found a very simple system that works, even though it probably shouldn't. So, I'll give you my rankings with a simple chart, then I'll explain the chart, hypothesize why the list works (generally), describe my impressions of what it means to be this great, and then I'll delve a bit into the predictive power of the list, and who among active players may rise onto the list. So, without further ado:

Players          Finals MVPs     MVPs     Totals     FYI: Titles/Finals

1. Jordan                6                   5            11              (6/6)
2. Kareem**          2                   6             8               (6/10)
3. LeBron               3                  4             7                (3/7)
4. Magic                 3                  3             6                (5/9)
5. Russell**           0+                5             5                (11/12)
6. Duncan               3                  2             5                (5/6)
7. Bird                    2                  3              5               (3/5)
8. Chamberlain**     1                 4              5                (2/6)
9. Shaq                    3                 1              4                (4/6)
10. Moses               1                  3              4                (1/2)
11. Dr. J**              0                3.5*         3.5         (1/3) & (ABA 2/2)
12. Kobe                 2                  1              3                (5/7)
13. Hakeem            2                  1              3                 (2/3)

*- Julius Erving had 2.5 ABA MVPs & 2 ABA Playoffs MVPs
+ Award Named After Bill Russell
** Substantially pre-merger awards

For players having post-merger success adequate to enter the list, here are their career totals:

(1) Michael Jordan- 11
(2) Kareem Abdul Jabbar- 8
(3) LeBron James- 7
(4) Magic Johnson- 6
(5) Tim Duncan- 5
(6) Larry Bird- 5
(7) Shaquille O'Neal- 4
(8) Moses Malone- 4
(9) Kobe Bryant- 3
(10) Hakeem Olajuwon- 3

Let me start the analysis by explaining the basic system. Add Finals MVPs (which are a pretty good marker for greatness by themselves, particularly post-merger, as they account for contributions to ultimate basketball success) to MVPs (which is a good marker for season-long dominance). A player must have a total value of 3 such achievements to qualify for the list. Then give the tie-breaker to the greater number of Finals MVPs. There are three adjustments that I made to the list pertaining to pre-merger issues (the merger of the ABA and the NBA that occurred in 1976). (1) Willis Reed is omitted for a mild overvaluation of his 3 achievements, which leaves him just below the minimum threshold, (2) Julius Erving ("Dr. J") has essentially half of his ABA accomplishments accounted for in addition to his NBA accomplishments, and (3) Bill Russell has an automatic life-long tie breaker for the NBA Finals MVP being named after him.

The pre-merger adjustments make the list more palatable because... (a) the Finals MVP was not awarded until Russell stopped winning trophies. So, access to the award totals was more limited; (b) the NBA Finals MVP award was named after Russell for his career playoff success; (c) the level of competition was lower in part because of: less attention to basketball, less money for basketball players, less sophisticated approaches to the game (often meaning less focus on defense and more focus on pace); (d) there were less teams, sometimes as few as 10, which partially offsets the lower competition (but not fully), but it also means winning a championship often involved winning two rounds of competition, rather than four; (e) for about ten years, much of the league's talent was drained by the new ABA league somewhat affecting Kareem's and Wilt Chamberlain's accomplishments and largely affecting Reed's and Erving's achievements. All of these factors, in my estimation, roughly neutralize each other with respect to Russell who is fully credited for his 5 awards (in which the voting players may have partially accounted for his playoff excellence) provided he gets the tie-breaker for having the Finals MVP award named after him.

The next issue is whether to mildly discount Wilt's and Kareem's awards for largely receiving those awards during the time of the ABA. Wilt is already on the lower end of 5 awards (only 1 Finals MVP), so for the time being, it does not make sense to reduce his achievements... and if there was a Finals MVP Award, Russell may not have won the regular season award so often. Abdul Jabbar is also already on the lower end of 8 (only 2 Finals MVPs), so if we reduced his total to 7.5, it would not impact LeBron's (or most others') ability to pass him. So, while we acknowledge that Kareem probably should have a mild reduction, for now, we can leave him on the list as is. The more complicated questions are what to do with Dr. J's and Willis Reed's awards (who happened to receive a Finals MVP Award quite literally for showing up to Game 7) who were drastically impacted by the existence of two competing leagues. If you devalue Reed's NBA accomplishments just a bit to offset the mildly reduced talent, he ends up somewhere near 3, like 2.8 which is NOT 3. This means Reed unfortunately does not meet the criteria necessary to make this list. Undoubtedly, he was a great player who accomplished a great deal (won 2 Finals MVPs, 1 MVP, and his Knicks won 2/3 of Finals appearances in the NBA during the time of the ABA), which helps his case among the other great players historically. But this list is reserved for the historically exceptional players, identified by a milestone, which is the value of 3 of these awards (again- while Reed technically has 3, the value of those 3 is slightly lower). Dr. J's accomplishments are even more difficult to discern as, in a weaker league, the ABA, he won 3 MVPs (1 of which was a tie, so 2.5 MVPs) and 2 Playoff MVPs during years when his team won the ABA title, but that award is not even referred to as a Finals MVP. So, the most fair method of accounting for his ABA achievements is essentially counting 1 of the 2 metrics. Luckily, it matters little which metric is counted because his 2.5 MVPs are worth approximately 2 Finals MVPs. He also won 1 MVP in the NBA, so he ends up with the value of 3.5 MVPs (or 1 MVP and 2 Finals MVP equivalents with other awards as tie breakers). So, that's how we end up with the 3 modifications to the list.

It's easy to create a terrific list of the greatest players ever. It's much harder to do it based on objective calculations or criteria. So, yes the list is subjective. But, it is tallied entirely objectively post-merger, and otherwise largely objectively. And it happens to be pretty easy to calculate, especially going forward. Each year, the relevant voters make subjective analyses about who is the best player... or who was the best player in the Finals. Not only are these awards subjective, but in my opinion (and probably the opinion of most observers), they are very frequently wrong. Steve Nash likely didn't deserve two MVPs, nor did D. Rose necessarily deserve one. Pierce or Parker may not have deserved their respective Finals MVP. Further, because of the randomness involved (excess award noise), a single award (or even two of them) says almost nothing about your place among the greatest players of all time. Andre Iguodala is a lifelong very good player who played really well in one finals series and may have been one great game 7 away from winning a second Finals MVP. However, at some point, the accumulation of these awards does reflect some level of sustained legendary greatness. Moreover, it's fascinating how the voters mix the inaccuracies of generically over-rewarding talent with voter fatigue to almost neutralize the negative impacts of each. For example, Bird, Magic, and Duncan were each, in all likelihood, the greatest players on their respective teams during the bulk of their careers and worthy of being in the pantheon of greatest players of all time, but each was on a team with incredibly talented players and/or incredibly talented coaches that were occasionally as valuable as they were. And giving these elite players unfettered credit for their many many championships seems unfair in comparison to the import of crediting Jordan or LeBron (who were also on some incredibly talented teams), whose team's fates were exceptionally linked to their individual dominance. The voters seem to have subconsciously or instinctively recognized this valuable distinction, as Jordan and LeBron got all of their team's Finals MVPs (thus far), while Bird got 2/3 of his available Finals MVPs and Duncan and Magic got 3/5 of their available Finals MVPs.

I have heard objections about my adjustments to the list, objections about an individual not making the list, like Jerry West (dynamic playoff performer), Oscar Robertson (per game stats), Karl Malone (cumulative stats), Kevin Garnett (defensive dominance), and even Dirk Nowitzki (offensive efficiency), and objections to a few of the players on the list, or their placement on it... all with varying degrees of persuasiveness. And personally, I might tweak the list such as swap Kobe and Moses Malone, but I'm willing to accept that my own beliefs and thoughts may be less valuable than having and employing a functional definitive roughly accurate system. And this is the only objective calculation that I have ever seen that accounts for our collective nearly-objective criteria (along with my admittedly subjective criteria) for greatness. So, whether the list should or should not work, in my estimation, it does happen to work. And for obvious reasons, the list is delightfully spread almost equally across the eras.

Okay, so generally my understanding of "best players" is based upon the players we would choose, if we wanted to start a team to win the most championships (or put us in a position to win the most championships), using the amount of time played by that player (factoring in injuries, rest, etc.), the player's dominance during their actual career, and the timelessness of that player, (i.e. if the player was drafted with the equipment, training techniques, coaching, player caliber, and rules/officiating of some arbitrarily chosen era, he would still be relatively dominant). The most important hypothetical metrics involved are titles (and wins) over standard or average replacement player (and this is complicated by the uniqueness of position players, regular season value vs. playoff value, historical time frames, etc.). So, a player's individual brilliance over a peak period of time and quality longevity both add significant value to a team. Ideally, you have a player like Kareem who is a unique talent, incredibly dominant for a time, but who also keeps you in the running for decades, thereby adding a great deal of value (and potential wins and titles) over his career. Jordan was better by most estimations (because of Jordan's sheer dominance, which is adequately reflected by the list), but Kareem's extended amazing play makes his career a great mix of the various factors of NBA greatness. So, Kareem has remarkable numbers of awards, championships, cumulative totals, per game averages, and advanced metrics in both the regular season and the playoffs. It's fitting that Kareem spans the pre-merger and post-merger eras and is so high on the list... and that in order to pass him, LeBron will need to extend his greatness further and longer.

The future is unclear. LeBron could win another MVP or more likely another Finals MVP to pass Kareem on the list... or not. Further, while Curry is currently closer to making the list than Durant, I would not argue that Curry has had a better career. But... if Curry earns another MVP or a Finals MVP from winning a second championship, I would certainly argue he outpaced Durant and entered the discussion for the greatest players of all time (having a career similar to Isiah Thomas's, but with greater regular season dominance). And if Durant, who is in his prime, could have a great year and win both awards in one year, he would deservedly vault onto this list. Or maybe, voter fatigue sets in, and Curry wins no more MVPs, and neither Durant nor Curry win any more championships, and they end up merely having really terrific careers. Either way works, as either player might, by the end of their careers, merit landing in the top 20 players of all time, but not necessarily the top 15. I'm at least open to letting their individual/team accomplishments and award-winning achievements make that determination... for now.