It has become a standard doctrine in the “church of basketball stats” that efficiency differential – or offensive efficiency minus defensive efficiency (or points per possession minus points surrendered per possession) – is a pretty good approach to team evaluation. You see this basic argument in the writing of Dean Oliver and John Hollinger (and a few weeks ago Mike Kurylo of Knickerblogger wrote a wonderful essay extolling the virtue of looking at a team’s offensive efficiency as opposed to points scored). So this idea is out there.
What I want to do today is look at a specific team – specifically the Phoenix Suns – and ask the question: Which Phoenix team was the best? For the answer, as noted above, we will turn to efficiency differential.
Phoenix History
When we look at the history of the Suns we see a large number of “very good” teams. Eleven editions of this franchise, since 1973-74 (when all the data to calculate possessions and efficiency was first tracked), Phoenix has posted a differential above 5.0. To put that in perspective, typically five or fewer teams in any given season post a mark that is better than 5.0 (the lone exception was the 1996-97 season when six teams bested this mark). So in eleven different seasons, Phoenix has fielded one of the top five teams in the league. But which of these editions was the best in franchise history?
When we think of top teams we tend to focus on NBA Champions or those teams that at least took the conference title. Two teams in Phoenix history advanced to the finals. But neither the 1975-76 edition nor the 1992-93 team was the best regular season team in Phoenix history. No, as Table One indicates, the best team – if efficiency differential is the criteria – was the Phoenix Suns in 2006-07.
Table One: The Phoenix Suns from 1973-74 to 2006-07
Last season the Suns finished with a differential of 7.4. This bests the mark of the 2004-05 edition (7.2) and the team in 1989-90 (7.1). No other team in franchise history topped the 7.0 mark.
Unfortunately for Phoenix, two other teams in 2006-07 were even better. Table Two – which was posted back in June – notes the efficiency differential for each team this past season. And from this table we see that both the Dallas Mavericks and San Antonio Spurs were even better than the best team Phoenix has ever put on the floor.
Table Two: Efficiency Differential in 2006-07
So “best” in Phoenix history is not good enough to be NBA “best.” Before we get to what that means in 2007-08, let’s discuss how Phoenix got to be this good.
The Impact of Steve Nash
The story begins back in the summer of 2004. That summer the Phoenix Suns acquired Steve Nash. As a result, as seen in Table Three, both Amare Stoudemire and Shawn Marion improved. Marion’s WP48 [Wins Produced per 48 minute] improved from 0.237 to 0.284. Stoudemire’s leap was even bigger, improving from a 0.077 mark in 2003-04 to a 0.225 WP48 in 2004-05.
Table Three: The Phoenix Suns in 2003-04 and 2004-05
Was this improvement due to Nash? In the case of Stoudemire, one can argue that Nash was an important factor. Stoudemire’s improvement is almost entirely tied to his improved shooting efficiency. But was it Nash or just the maturation of a player who entered the league out of high school? It’s not unusual for players to improve early in their career so what we observe for Stoudemire could be just the natural aging process of an NBA player.
As for Marion, we also see improvement with respect to shooting efficiency. But Marion also improved as a rebounder and it’s a bit hard to see what Nash had to do with that dimension of Marion’s game. It’s important to also note that Marion has always been a very productive player. In fact, when I get to my review of the 1999 NBA Draft we will see that Marion was by far the most productive player out of this class. And that was true before he teamed up with Mr. Nash.
What of the other players on the team? Both Joe Johnson and Leandro Barbosa played substantial minutes in 2003-04 and 2004-05. And both improved with respect to shooting efficiency as well. But the improvement for Barbosa was quite small (as seen when we look at the small change in his overall performance). Johnson’s improvement was much bigger, but what is interesting is that his shooting efficiency remained above average even when he went to Atlanta. So apparently Johnson’s improvement was not all about Nash.
We should also note in closing our review of Nash’s impact that the best WP48 Nash ever posted in Dallas was 0.215. In his second stint in Phoenix (he started his career in Phoenix also), Nash has been above 0.300 each season. Consequently, one could also make the argument that Stoudemire and Marion enhance the productivity of Nash.
The Phoenix Trio
Regardless of how we look at who impacted who, it’s quite clear the combination of Marion, Stoudemire, and Nash produce wins. In 2004-05 this trio produced 48.2 of the team’s wins. As noted in Table Four, when Stoudemire missed almost the entire 2005-06 season due to injury, the Suns worsened.
Table Four: The Phoenix Suns in 2004-05 and 2005-06
This past season Stoudemire was once again healthy. And again, as seen in Table Five, the trio of Marion, Stoudemire, and Nash were very productive; combining to produce52.3 victories. Or to put it another way, everyone else on the roster was worth less than ten victories.
Table Five: The Phoenix Suns in 2005-06 and 2006-07
As noted in the discussion of the Pareto Principle (a discussion which includes this table), no team in the NBA had a better leading trio in 2006-07 than Phoenix. Unfortunately for Phoenix, the supporting cast in both San Antonio and Dallas were better. The non-top three in Phoenix only produced 8.5 wins. The non-top three for the Spurs and Mavericks produced close to 20 victories. Consequently, each of these teams ended up being a bit better this past season.
Going Forward
Can the Suns close this gap? This past summer the Suns sent Kurt Thomas away and added Grant Hill. Thomas posted a 0.136 WP48 last season while Hill, who is only a shadow of the player who dominated the NBA in the 1990s, posted a 0.107 mark.
If all we consider is the productivity of each player, it may look like these moves make the team slightly worse. But we need to note that Hill replaces James Jones, a player who posted a 0.001 WP48 last season.
For Thomas, it’s not clear who takes his minutes. But it might be Brian Skinner. Skinner posted a 0.039 mark in Milwaukee last season. So the distance between Hill and Jones is actually slightly bigger than the distance between Thomas and Skinner. Consequently, although Thomas was slightly more productive than Hill last season, the Suns are a bit better with this move.
I should emphasize, though, the words “bit better.” When we look at the projected roster in 2007-08 we see that this team is still going to be all about Marion, Stoudemire, and Nash. Yes, Hill helps a bit. What would help even more is the return of the Boris Diaw we saw in 2005-06.
Two years ago Diaw posted a WP48 of 0.123. This is slightly above the average mark of 0.100. Last year, though, his mark declined to -0.006. If Diaw remains what he was in 06-07, then the Suns will probably continue to place third behind Dallas and San Antonio. However, if Diaw makes a comeback, then the difference between these teams becomes much smaller.
In sum, the Suns have the stars. They just need the supporting cast. If Diaw steps ups (and injuries don’t happen, productivity of everyone stays the same, etc…), then Phoenix in 2007-08 will not only be the best in team history, it might be the best team in the Association.
– DJ
For a discussion of other teams see NBA Team Reviews: 2006-07
Our research on the NBA was summarized HERE.
The equation connecting wins to offensive/defensive efficiency is given HERE
Wins Produced and Win Score are discussed in the following posts
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Wages of Wins Journal 
Tim
October 16, 2007
I’m sorry, but as a Bulls fan I’m still high on the potential of Tyrus Thomas and Joakim Noah, not to mention the further potential of Luol Deng, and even the potential of some of the Bulls’ other young players. And you are to blame! As you noted in “The Return of the Bulls” (https://dberri.wordpress.com/2007/09/26/the-return-of-the-bulls/):
“As a freshman in 2005-06, Thomas posted a 0.452 Win Score per-minute. Last year, as a junior, Noah posted a 0.398 mark. To put these numbers in perspective, Greg Oden had a 0.402 mark last year. Yes, I still expect Oden to be good. And yes, there is a good chance that both Noah and Thomas will develop into above average players.”
I’m afraid you have me convinced that you know something most general managers do not about how to analyze college players. Even Aaron Gray looks pretty good to me now. (https://dberri.wordpress.com/2007/06/26/win-score-and-the-nba-draft/)
And I’m not listening to your cautious qualifiers. (https://dberri.wordpress.com/2007/09/24/searching-for-greatness-in-portland/) because other general managers obviously aren’t using your system, and those qualifiers don’t apply to the Bulls! ;-)
By the way, apparently the Bulls agree with you that Luol Deng is more valuable than Ben Gordon. According to the Chicago Sun-Times, “A source close to the situation said it appears more likely that the team will work out something with Deng than with Gordon, who also is eligible for an extension of up to five years.” http://www.suntimes.com/sports/basketball/bulls/603235,CST-SPT-bull15.article
I swear, I wonder if John Paxson is reading your blog! Add in the contract for Ben Wallace that most analysts not named Dave Berri consider outrageously high, and it really seems that way. The only problem I see is that not everyone on the team can be another Ben Wallace; someone has to score.
Tim
October 16, 2007
Whoops! I went to all the work of finding all these links, and then I post the response to the wrong story! This was supposed to be a response to your next story, “Looking Back at the 1995 Draft or An Antidote for the Potential Drug.” I think I started with your home page and clicked on the wrong “comments.” Oh well. Unless you know how to move the post to the other column, I’ll just hope that you read it anyway. Thanks again from a regular reader.
dberri
October 16, 2007
Tim,
Thanks for the comments. I have no idea how to move these, though. Plus, I like them here. No one had commented on the Suns story before you posted your comments. Which is odd because WordPress says this post was read by a quite a few people.
Ryan Schwan
October 16, 2007
I read it – and thought about commenting. It just seems to me that the Suns are a team on the way down, not up. Selling draft picks and rotation players for cash and replacing them with old, injury prone players isn’t the sign of a team on the rise.
I’ve made this point on my blog a few times, and your charts confirm it – this team is very susceptible to injury. They only have a seven man rotation, and after that, anyone they play is going to hurt them. And those seven guys log a lot of minutes.
Something is going to break some time.
dberri
October 16, 2007
Ryan,
I agree, injuries could be an issue.
By the way, I still remember my assignment. Hopefully on Thursday or Friday.