Friday, May 05, 2017

How Do You Evaluate the All-Time "Best" Basketball Players?

In March and April of 2015 I listed my top 36 NBA players of all-time.

I was chatting with a friend the other day and he said, "You've ranked LeBron #4 overall, but you say he has an argument for #1.  Why don't you move him up?"

I will need to look at updating the list in April 2018, but my friend's comment really brings up the question of how, exactly, do you "rank" players?  How do you rank them across generations?  Are newer players always going to be listed higher due to nutrition, training, conditioning, etc.?  What rank do you assign to someone with a great but short career (example, George Mikan)?  Do you rank them based upon how great they were at their absolute peak (example, Tracy McGrady had a top 15 all-time season -- does he get ranked based on his peak ability, when he shined the most?)

In a prior item, I identified 10 common traits of an all-time great NBA player

1) Was regarded at his time as truly great; evidence of this would be all-NBA selections and top 5 MVP voting;

2) Was regarded as the best player in the game at least once, and, hopefully, on multiple occasions, as would be generally reflected by MVP awards;

3) Was dominant in some aspect of the game - teams needed to adjust their games to stop him and generally did not succeed;

4)  Was physically superior to other players;

5) Was well ahead of his time as a player/there simply were not players that did what he did at the time he did them (and possibly there still aren't today);

6)  Played a sufficient amount of time that his body of work can be considered not just a blip;

7) Was consistently excellent for a long time (this can be determined most easily by advance stats);

8) His playoff performance did not generally decline substantially from his regular season work (same);

9)  Led his team far into the playoffs on multiple occasions;

10) Won multiple titles.

But how, then, do you weight these factors?  Karl Malone was 1-4, 6, 7, 9.  Huge gap for 8 and 10.  So would a player like, for example, Chauncey Billups or Jason Kidd be a "better" player than Karl Malone? 

Similarly, if a player is recognized as many time all-NBA and an MVP candidate (cough, Dwight Howard) and he becomes a somewhat lesser player due to back issues....where does he fall?  There is no question that for a 6 year period, Dwight Howard was unbelievable.  George Mikan, likewise, was great for 5 years.  If we ignore Mikan's multiple titles, should he get REWARDED for playing so few years while Howard gets punished because people have to watch him struggle for 10 years until the gas finally runs out of the car?

I mean, had Johan Santana blown out his arm on his last pitch as a Twin, he is probably a first-ballot Hall of Famer.  By continuing to pitch with the Mets, he is regarded almost certainly as a non-HOF player!

So, what "method" do I use?  What weighting do I give these 10 factors?  Is Robert Horry (7 titles) a better player than Karl Malone (0 titles)?  Of course not.  How close, however, would Horry have to be to Karl Malone as a player to get the nod?  When would the 7 titles be enough?

Here are my thoughts:

A) The primary means of evaluating a player should be his ability to excel in regular season play.  After all, without good regular season play there can be no playoff play. It is also a far larger sample size.  Playing well over 82 games is far harder than playing well over 10-12-20 games.

B)  Next on the list has to be the perception of the player from those who watched him play.  I never saw Bob Pettit or Elgin Baylor play.  I do know, however, that they were deemed to be 10X first-team all-NBA.  George Mikan was named the best player in the world 1900-50.  You cannot ignore that fact.

C) Then I turn to advanced stats, which are an attempt to measure certain levels of play by assigning a number to them.  The greatest players generally have two things in common - a) they have huge single-season number; b) they have great longevity at a high level of performance.

D) It is at this point that you start looking at post-season play.  This is where guys like Chris Paul and Kevin Garnett and Karl Malone and John Stockton start to falter.  You will note, however, that all of those players are in my top 26 all-time, so I do not consider consistent playoff greatness to be a disqualifying factor.  It is, however, a factor and one that can impact how the player is viewed when lines need to be drawn on "is Tim Duncan a better player than Karl Malone?"  Since the goal of playing basketball is to win the title, you need to give Duncan the extra "plus" over Malone.

E) It is at this point that you need to be able to rely somewhat on personal observation (if possible) and sort of a 'feel" for who was better.  (That is why we have all of these lists).  I fear that many lists use this factor E almost exclusively as their criteria.  I mean, ESPN has guys like Iverson and Isiah Thomas and Bill Walton and Kevin McHale very high on their all-time lists.  There is really no objective basis for these rankings - they are almost purely subjective or based upon something that once happened in one game (Isiah scored a million points with a broken ankle; Iverson stepped over Ty Lue; McHale was impossible to stop in the post (he played with Larry Bird yet averaged over 25 ppg once and over 20 only 5X))  Walton's ranking appears almost cult-like.  Yes, he could perform all aspects of the game when healthy.  He was healthy for 1 1/2 years and mostly healthy for 1 Sixth Man of the Year award.  He has 39 career WS.  His career WS performance is similar to Tom Boerwinkle's).

So, I guess my criticism of most ranking systems is that they start with E, then go to D, then maybe include a sprinkling of A.   I'd say you set your base by looking at A, B, C and then use D and E only in cases of extremely tough calls. 

The bias in my method, I will concede is AGAINST the belief that a guy is great simply because he gets a lot of shots and makes some.  And it is biased against guys who have enjoyed great post-season success (Horry, Chauncey, are disqualified from a top 36 consideration before ever reaching category D).

But I think that is the way it ought to be done.

As far as the issue of longevity - I do award "points" for consistently great regular season stats and for all-NBA awards and MVP Award Shares.  All of these are a function of longevity.  Certainly a guy who was great for 15 years has to get a plus over a guy who was great for 6 years.  After you reach 9-10 years of good play, however, there are diminishing returns for playing longer.  But still some credit. 

And on the issue of whether a guy who gets, say, 35 WS one year (has never occurred) and then sucks for 7 years should be ranked highly, I'd say generally no.  The exception would be if the guy was seriously injured. 

No comments: