A few weeks ago I wrote a couple of columns examining the best NBA center across the past two decades.
The Best NBA Center in my Students’ Lifetime
Ewing or Olajuwon: Who was Better?
These posts argued that Olajuwon was better than Ewing, Robinson tops Shaq, and Robinson offered more than Olajuwon. In sum, since the mid-1980s, the Admiral is the most productive center.
The careers of these players over-lapped, so people had an opportunity to see these players play against each other. This is not the same story we see today. Although the game today has a number of very good players – and a few that rank among the all-time best (see Chris Paul) – the NBA in 2007-08 doesn’t have a plethora of great NBA centers. In fact, I think there is only one.
Ranking Howard
Last April I posted the following list of top centers from last season.
Table One: The Top 15 Centers in 2007-08
As Table One indicates, the list of great centers includes Dwight Howard and then… okay, not much after Howard. Camby has had a few good years, but injuries remove him from the list of all-time great centers. The best of remaining centers on the list produced eight fewer wins than Howard last season. In sum, Howard is currently head and shoulders above the current crop of centers. But how does Howard compare to the all-time greats?
Howard just completed his fourth season in the NBA. One way to compare Howard is to compare his performance last season to what other great centers did in their fourth campaign.
Before we get to the comparison, let me note which centers rank among the “all-time greats.” To begin, the NBA only began tracking all of the box score data we use to measure productivity in 1977. Prior to the 1977-78 season, the NBA did not track turnovers for players. And prior to the 1973-74 campaign, there was no data on offensive rebounds, defensive rebounds, steals, and blocked shots.
The lack of data restricts are analysis to players who played their fourth season in the past three decades. So we can’t look at George Mikan, Bill Russell, Wilt Chamberlain, or Kareem Abdul-Jabbar. But we can look at Moses Malone, Hakeem Olajuwon, Patrick Ewing, David Robinson, Dikembe Mutombo, and Shaquille O’Neal. And when we compare what Howard did his fourth season to what these players did in their fourth year, we find a surprising result. As Table Two indicates, of the fourth seasons considered, Howard’s WS48 [Win Score per 48 minutes] of 19.4 led everyone.
Table Two: Top Centers in their Fourth NBA Season
Before Orlando Magic fans think that Howard is going to be better than all these great centers, a few issues to consider:
- This is Howard’s best mark in his career.
- Shaq, though, posted a WS48 [Win Score per 48 minutes] of 20.2 in his second season.
- Turning to the Admiral, he had marks of 21.7 and 21.8 in his second and third campaigns.
- Moses Malone – at the age of 23 – posted a mark of 21.5. If we count Malone’s ABA seasons, this was actually his 5th professional season (Table Two reports his 4th professional season and second NBA campaign).
- Focusing on Howard, he still exhibits a few weaknesses. Like Shaq, Howard has trouble hitting free throws. And like Malone, he thus far is a relatively weak passer.
All that being said, Howard’s performance in his fourth season was outstanding. Howard was only 22 years of age last season and already he is posting many of the numbers we associate with the all-time greats. So not only is he the most productive center in the game today, he is on his way to being one of the best to ever play in the NBA.
Orlando Past, Present, and Future
What does this mean for the Magic? For an answer we turn to Table Three.
Table Three: The Orlando Magic in 2007-08
Table Three reports what the Magic players did in 2007-08 as well as what we could have expected given what the players did in 2006-07. As we can see, Howard, Hedo Turkoglu, and Jameer Nelson all posted much better numbers this past season. As a result, Orlando’s efficiency differential [offensive efficiency minus defensive efficiency] of 5.7 tied with the New Orleans Hornets for the fifth best mark in the league. Only the Boston Celtics, Detroit Pistons, LA Lakers, and Utah Jazz posted a better differential.
In the off-season the Magic drafted Courtney Lee and added free agent Mickael Pietrus. It’s not clear either player will make much difference next season. The team, though, is helped by the retirement of Pat Garrity. Last season Garrity produced -1.4 wins in only 284 minutes. For his career he produced -16.0 (yes, that’s a negative number) and posted a -0.070 WP48 [Wins Produced per 48 minutes]. This may be the worst mark posted by anyone who played at least 10,000 minutes in the NBA (although I am not sure).
The departure of Garrity is probably going to make this team better (hard to believe a player could hurt a team so much in just 284 minutes). The team is also going to be helped by the fact the Celtics and Pistons are yet another year older. For the Magic, time brings more experience to Howard and the possibility of even more output from the top center in the game. Detroit and Boston, though, are led by much older players. Time for these teams is the enemy, as age inevitably must lead to lower levels of productivity.
If I were to guess, I would expect the Magic to close the gap we see between this team and both Detroit and Boston. But I don’t expect Orlando to take the Eastern Conference title in 2009.
Howard, though, is only 22 years old. Furthermore he just re-signed with the Magic. One would expect before Howard has a chance to depart Orlando, he will make at least one appearance in the NBA Finals. And one should also expect, if Howard stays healthy, he will probably go down as one of the best centers to ever play the game.
– DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
AhmedF
September 26, 2008
I have to say though – while a big believer of Howard through the entire NBA season (whilst a fan of the Raptors myself), the Olympics was not very kind to him – to me Chris Bosh completely outplayed him.
I also think Amare’s numbers are skewed with his injury and what not – he seems to have a nastiness about him that can only serve him well in games.
Joe
September 26, 2008
I wonder what the least wins produced in a single season is. I see Adam Morrison managed a paultry -11.6 . Niikoloz Tskitishvilli had a pretty epic season one year as well.
Mountain
September 26, 2008
Howard the most heavily fouled. FG attempts halfway between Ewing and Mutumbo. Will the later change? Will the former?
I’d guess not much for shot frequency this year but likely after Turkoglu probably leaves after this year. Horrible as a jumpshooter (25% FG% on them according to 82games) and nothing special at all even from in close except for the high rate of accurate dunks. He basically dunks 40% than today’s older version of Shaq but is only half the jumpshooter (Shaq 47% FG% on jumpers).
The foul shot rate probably doesn’t decline and maybe even goes up. Last season he was 24th highest in league history among players playing over 2000 minutes in a season. Shaq 3 of top6 and 6 of top 26. Wilt 61-62 highest all-time.
SWaN
September 26, 2008
Seems Mr. Berri is optimistic for Magic.
I can see Howard continuing to perform well, but what level of performance will Turkoglu and Nelson reach this year?
Will return of Battie help? How much of a loss was Ariza?
I expect Magic will have fewer wins this year over last year.
Vince Gagliano
September 26, 2008
By the way, with the exceptions of Howard and Malone, all the other centers listed played at least three years of college.
So Howard’s 4th season would be roughly equivalent to Shaq’s or Olajuwon’s rookie season as far as age is concerned. And for guys like Ewing, who went all four years, his 4th season would be equivalent to Howard’s 8th.
However, in Howard’s defense, he is the youngest player in NBA history to collect 4,000 rebounds; not to mention that he led the league in boards last season.
JB
September 26, 2008
The Magic lost Arroyo, Dooling, and Evans in the offseason. According to this model, that’s 7.6 wins. Even without Pat Garrity and with Anthony Johnson (who falls prey under the same age argument used against Detroit and Boston), you’d have to assume a lot of internal improvement just for the team to keep pace, unless I’m missing something.
Dave
September 26, 2008
I know it’s anecdotal and subjective, which means many here will hate this assertion. But a great center, or even a very good one, shouldn’t be muzzled in the playoffs like Howard was against Detroit … especially since he was frequently matched up against a 6’7 post player in Jason Maxiell. Howard is a wonderful specimen who is still largely clueless against quality opposition – and playing the “only four years out high school” card means less when you consider that few elite players now graduate before coming to the NBA.
Mountain
September 26, 2008
Here is another comparison: Howard vs Biedrins.
Different physiques but similar on some stats= overall FG%, FT%, jumpshooting, rebounding.
Notably different per minute on getting fouled (advantage Howard), taking shots (Biedrins 50% higher per minute), minutes (Howard 50% more), giving fouls (Biedrins 40% higher) and turnovers (Howard twice as high)
http://www.basketball-reference.com/fc/pcm_finder.cgi?request=1&sum=0&p1=biedran01&y1=2008&p2=howardw01&y2=2008
Mountain
September 26, 2008
But in team context, using adjusted +/- Howard got almost a +13 and Biedrins just barely above +2. Do the boxscore stats that separate them directly move Howard that far into the lead- now- or is Howard a more effective teammate or was Howard just in a better situation or were these pure adjusted +/-s stretching the difference between them? Maybe future seasons will help resolve on this last question. In 06-07 Howard and Biedrins were real close on adjusted +/-. Which was true or it the real answer splitting the difference? How much better is Howard than Biedrins?
Mountain
September 26, 2008
Using WP48 the difference is indicated as small. Even looking at multiple metrics (usually helpful) this isn’t easy – or is it?
Mountain
September 26, 2008
The minutes played are a significant difference. Will his foul rate continue to limit his time or could / will he adjust?
Mountain
September 26, 2008
Biedrins, that is.
dberri
September 26, 2008
JB,
I really didn’t look over the roster when I made my guess about the Orlando. So between the two of us, I think I am probably missing something. Still, I think Boston and Detroit will slip as time goes forward so that will help Orlando.
Mountain,
On adjusted +/-… the standard errors for both Howard and Biedrins is quite large. Neither player, according to how this metric is calculated at basketballvalue.com, had a statistically significant impact on outcomes last year. So in that sense, they are equal.
Alex
September 26, 2008
Agree with JB – the Magic lost several good role players in the off season. Anthony Johnson is a productive player (2.4 Wins Produced in 2007-08), but I don’t know how good Pietrus is (he was not productive in 2006-07). If JJ Reddick get more minutes, that’s not helpful.
Mountain
September 26, 2008
I was using the adjusted +/- at countthebasket because I like to see the offensive/defensive splits but I hadn’t noticed Howard’s error was that high. Thanks for noting that Dave.
Jason J
September 26, 2008
DB – I have a small issue with this (and I really like win score and paws – I’m not hating on them). If Shaq scores 6 more points in 2 fewer minutes, get’s the same amount of offensive rebounds, twice as many assists, and does it all in an era of much more competitive centers, I find it hard to believe that 3 less missed field goals and 3 extra defensive rebounds turn the whole thing in Howard’s favor.
I realize that what I believe is irrelevant because the numbers are proven in a sense – but I may need to debate the value of the defensive rebound one more time. It just seems like the kind of stat that can be replaced by teammates. How hurt is Shaq in this comparison because he was able to rely on Horace Grant to share the rebounding burden? What about Hakeem & Thorpe splitting the load or Ewing & Grant? In equating WS, every loose defensive board you allow a teammate to grab hurts your overall ranking. But does it hurt your team to have it’s best low post scorer running down court to post up a while the other big on the team handles the defensive board work and outlet passing?
And even for other players – Did Magic lose his rebounding mojo when AC arrived, or did he run to take the outlet pass and get showtime underway? Did Jordan peak in 1989 and lose his rebounding juice by 1991, or did he start breaking into the wings when Horace developed and Phil made Scottie the primary ball handler?
I understand that these things should perhaps show up in higher shooting percentages or the like, and they might, but each time you fill a lane or run to post up instead of competing with a teammate for a defensive board is the same as missing a shot or having a turnover as far as WS is concerned, but the effect on the team is actually positive – exactly the same as if the player in question had grabbed it.
I don’t know if I’m explaining this very well, and I don’t mean to take anything away from WS, PAWSmin, or Dwight for that matter. I just think the valuation of the defensive rebound seems a little out of proportion in this instance.
dberri
September 26, 2008
Jason J,
In this forum we have discussed the interactions you note before. My argument is that the interactions are not that big of a factor. Player performance tends to be fairly constant over time. That being said, interaction effects are not zero (diminishing returns do exist). I just argue they are not that “big.” The debate really centers around how you define “big.”
Kevin
September 26, 2008
Vince Gagliano – Shaq was 23 in his 4th season. Howard is 22. Howard’s 4th season is equal to Shaq’s 3rd (where Shaq’s WS/48 was 18.7).
Hakeem was 22 as a rookie, Robinson was 24, Dikembe was 25, Ewing was 23.