1952-2014 statistical rankings of 708 NBA/ABA careers

[Mike G]

We were only discussing a very small sample size in the previous post, and LeBronORL had more impressive individual metrics than JordanPHX. I don't see why using defensive rating there should apply really. It shouldn't matter that LeBron wasn't in the Finals that year. The same sort of theory applies to the 2014 Heat, especially when Portland Jordan happened.

Over 1000/2000/3000+ minutes I can see the usefulness of Faux win shares. But not when trying to determine who had the most successful handful of games in the playoffs. D-rating is even worse like that. The previous post was about Finals average, which is just another way of saying losing to the 2007 Wizards improves LeBron's career average in the Finals. My whole point was, it is a nice feat regardless that James got there, given that Jordan was in his rookie season at that same stage in his career.

Well, that's why Averages were invented. You could argue that we had a cold winter, and so why believe all the evidence around the world that Earth is warming.

I didn't know you were actually arguing that a select handful of games would prove that LeBron is better than Jordan. Never occurred to me that this could be useful.

Yeah, getting to the Finals 5 times is very good. Dragging below-avg teammates to get there is also very good. Last time I brought up the 5 Finals visits, you didn't seem to like it.

Meanwhile, I rank ORtg/DRtg of 119/112 as better than 118/113 -- both with Usg of 39%. How could it not be?

[huevonkiller]

Well, that's why Averages were invented. You could argue that we had a cold winter, and so why believe all the evidence around the world that Earth is warming.

Actually, the cold winter was indeed blamed for the contraction in GDP, in the first quarter. ;]

I've never thought of this debate in terms of: "In this one series Lebron was great", or "In that one series Jordan was not great". Those are ancillary matters really. I've always thought in terms of multiple post-season runs. Faux win shares could be used more easily there. But if we are going to start a discussion about 2009 again, then I'll point out the obvious. LeBronORL has better individual metrics than JordanPHX.


Let's not get into politics. ;]

I didn't know you were actually arguing that a select handful of games would prove that LeBron is better than Jordan. Never occurred to me that this could be useful.

I think there's a difference between handful, and 100+ games. D-rating seems better suited for the latter.

Yeah, getting to the Finals 5 times is very good. Dragging below-avg teammates to get there is also very good. Last time I brought up the 5 Finals visits, you didn't seem to like it.

I agree with you and I like it. The discussion is more robust than this though.

Meanwhile, I rank ORtg/DRtg of 119/112 as better than 118/113 -- both with Usg of 39%. How could it not be?

It evidently is not, Orlando was a better opponent.

Anyway, if I asked you, "who was the best defender in that game four of the 2012 Eastern Conference Semi-Finals?" , would you use defensive rating? Not really. It is one thing to compare multiple title runs, that is much simpler.

[Mike G]

The 2009 Magic had SRS of 6.49
The 1993 Suns had SRS of 6.27
Upon this 0.22 ppg difference, you claim 118/113 > 119/112 ?

[huevonkiller]

The 2009 Magic had SRS of 6.49
The 1993 Suns had SRS of 6.27
Upon this 0.22 ppg difference, you claim 118/113 > 119/112 ?

No that was not the issue either. I didn't use the 113 or 112, instead I made an individual analysis of their offense.

[Mike G]

, Orlando was a better opponent.

I'll point out the obvious. LeBronORL has better individual metrics than JordanPHX.

The 2009 Magic had SRS of 6.49
The 1993 Suns had SRS of 6.27

One of these is true.

[huevonkiller]

No the following is what I am referring to:

"and the Suns had a 110.2 defensive rating.... or in other words worse than the 2014 Philleh Sixers.
The 2009 Magic had a defensive rating of 105.3..."


JamesORL has better individual metrics than JordanPHX. Both players had "high" d-ratings and I disregarded both. The reason was explained in my previous two iterations. ORL's defense is better, so they were the better opponent.

[joveLeffud]

Give this post a 10 great analysis and read

The critic in me wants to say would this have turned out this way without the popularity vote of giant city fans placing their player in the ASG
Assist seem slightly undervalued.

Really good work though

[Mike G]

.. would this have turned out this way without the popularity vote of giant city fans placing their player in the ASG...

Can you elaborate on this?

[schtevie]

Though I find the consumption of inter-era lists to be empty calories, if one is to play the game of above-averageness in any or some given years, why is +/- data not introduced when comparing Jordan and James?

Perhaps Jeremias will one day find time and interest in calibrating his post-1990/pre-2001 quarter-based numbers with the post-2000 PbP numbers, but what currently exists is rather suggestive of the fact that James is the more above-average player.

To briefly summarize, in the overlapping ages (1991-93 and 2012-14) James' xRAPM is 4 points higher. Now, I don't know if these data have the desired playoff focus, and it seems apparent that quarterly xRAPM biases the best players downwards, but the bias doesn't seem to be close to 4 points. Perhaps 2? And that leaves a clear margin.

Then from another perspective, if Jordan is particularly ill-served by such a measure, which of his teammates are correspondingly, undeservedly puffed up?

[Mike G]

You seem to be saying the Jordan era and the LeBron era are comparable. And Jordan played against Magic, who played with Kareem, who vied with Wilt ...
The alternative to all-one-era-faith is a choppy mess of stuff like, "Was Erving '70s or '80s ?"

The next phase may be to find a player's "plus-minus/productivity" ratio: Some guys have puffed up stats (even after standardization), and others are not individually productive yet valuable.

Sometimes these differences do take care of themselves -- DeMarcus Cousins doesn't get into the playoffs; Dominique Wilkins never did well in the playoffs.
Bruce Bowen and Joe Dumars probably had good plus-minus, and their productivity stats do not sufficiently convey it. But they got more minutes, and they had playoff success, thanks to their uncounted contributions.

Since the latest greatest RAPM seems to need many years to rate a player accurately, the player is no longer what he was when the data were accruing. It may be mostly good for whole-career ranking, in fact.
But if his "pm/p ratio" is fairly constant (or we get an aging curve for that), we can then rank prior and current seasons: If his numbers are up this year, while his PM is down, his PM is probably registering too low.

[schtevie]

You seem to be saying the Jordan era and the LeBron era are comparable. And Jordan played against Magic, who played with Kareem, who vied with Wilt ...
The alternative to all-one-era-faith is a choppy mess of stuff like, "Was Erving '70s or '80s ?"

The next phase may be to find a player's "plus-minus/productivity" ratio: Some guys have puffed up stats (even after standardization), and others are not individually productive yet valuable.

Sometimes these differences do take care of themselves -- DeMarcus Cousins doesn't get into the playoffs; Dominique Wilkins never did well in the playoffs.
Bruce Bowen and Joe Dumars probably had good plus-minus, and their productivity stats do not sufficiently convey it. But they got more minutes, and they had playoff success, thanks to their uncounted contributions.

Since the latest greatest RAPM seem to need many years to rate a player accurately, the player is no longer what he was when he was at his best. It may be mostly good for career ranking, in fact.
But if his "pm/p ratio" is fairly constant (or we get an aging curve for that), we can then rank prior and current seasons: If his numbers are up this year, while his PM is down, his PM is probably registering too low.

Any time you create a "greatest" list, you are comparing players across years and eras, and you are forced to make some very strong assumptions in attempting to create a common basis. I am certainly not saying (nor have I ever said) that the Jordan and James era are directly comparable. Very much to the contrary.

However, if one is going to engage in such a speculative venture, at least you can narrow the specification to come up with some conditionally firmer conclusions. Specifically: is Player X in era A as above-average compared to Player Y in era B. And I find the latest and greatest +/- stats to be preferred for determining "above-averageness", what in the instance suggests a pretty clear answer to the Jordan vs. James comparison.

As for xRAPM requiring many years to rate a player accurately, well... kinda, but no. But in the instance we do, in fact, have these many years. We have James entire career and the entire backside of Jordan's, what includes all of his championships. I don't know what the prior was for the initial 1990-91 numbers, but this was roughly at the peak of Jordan's career, and the subsequent two years showed no change. And the comparison noted was for James same age.

So, the issue then is how significantly the 90s leaders of quarter-based xRAPM estimates are biased downward compared to the hypothetical PbB alternative. As a simplistic first cut, we note that Jordan was never better than the 4th best player in the NBA by this measure. If one then looks at the post-2000 average for 4th best NBA player (implicitly assuming an constant distribution of estimates across years/eras) this increment above Jordan's base of approximately 5.0 only gets him halfway or so to James' 9+. And 2 points of +/- is a pretty big deal.

[kjb]

I don't think we need to get much into the relative quality of different eras when comparing players across eras. The key issue in my view is relative dominance -- how much did Player X dominate HIS contemporaries compared to how Player Y dominated his contemporaries. That's a question we can answer with more certainty.

[wilq]

NBA and ABA titles are considered [...]
No awards are considered [...]

Isn't this an inconsistent approach? Aren't titles basically awards for the teams?

Code: Select all

rank   per36 rates   - years -      G   Min  Eff%    Sco   Reb   Ast   Stl   TO   Blk   3's    T    po/rs   pog
1   Michael Jordan   1985  2003   1251   39  .559   31.2   6.5   5.1   2.3   2.7   .8   729   44.4  1.06   .154 [...]
7   LeBron James     2004  2014   1000   40  .571   29.4   7.5   6.5   1.6   3.1   .7  1359   43.9   .98   .168

Out of curiosity... based on your numbers how many good years would LeBron need to overtake Jordan as #1?

[Mike G]

If you play 23 postseason games and lose at the Finals, is that just as good as playing 23 games, winning it all, and having nobody else to play?
At one time there were just 2 rounds -- the Russell Celtics would beat one opponent to get into the Finals.
If there had been more rounds, they'd have gotten into more playoff games. Since the league has expanded, later players can get into a lot more games.

The bonus of 10 equivalent playoff games assumes that another round would go on avg 5.5 games; a champion would have a 50/50 chance of yet another 5.5 games; and so on. These possible series add up to 11. The number 10 just rounds it down a bit, since we never really have an infinite tournament.

It's a team award just as any other playoff minutes are a team award. A player may be lucky or unlucky in a given year; but the longer he plays, the more likely his 'team attitude' is reflected in his playoff appearances -- fraction of minutes in playoff games, which in the list is shown as pog.

Robert Horry got lots of playoff minutes and titles. Clifford Robinson did not. Their po/rs ratios are at the opposite ends of the individual success spectrum, and this definitely affected the fortunes of their teams.

Awards, meanwhile, supposedly go to the best players. If you're among the best at something, it's usually contained in your stats. A major motivation to make a list that's 100% statistically determined is that I didn't want to perpetuate the hype -- or conversely, continue to overlook a player's excellence. Kobe's rookie all-star game, and his selection this past season, do nothing to elevate his stature.

... how many good years would LeBron need to overtake Jordan as #1?

Good questions, btw. LeBron currently has been in 1000 games (with playoffs) and 40,000 minutes. Ranked #7 at present. If he continues at 40 mpg, career production rates, same pog and po/rs, he needs:
- 28 games to catch Karl Malone
- 73 games to catch Shaq at #5
- 93 to pass Duncan -- but that's if Timmy stopped playing
- 142 G to pass Wilt at #3
- 168 to pass Kareem to the #2 spot

These numbers assume he hasn't won another title in the meantime. To pass Jordan to the top spot, he can continue his career rates for :
- 346 games with no title.
- 330 games with a 3rd title
- 314 G with 2 more titles.

[Mike G]

I don't think we need to get much into the relative quality of different eras when comparing players across eras. The key issue in my view is relative dominance -- how much did Player X dominate HIS contemporaries compared to how Player Y dominated his contemporaries. That's a question we can answer with more certainty.

I agree, mostly. If you go back in time far enough, you have a league that is white and regional. Nobody from outside the northeastern quarter of the US; and until Russell/Baylor, just a smattering of black players.

In that era, player stats are hampered by shorter schedules, shorter playoffs, and generally shorter careers. So it may be that they're "punished" just about right.

In any case, since we can't determine relative strength of era, we can just accept the numbers as being relative to the competition, and adjust for pace, etc.