Marc Stein of ESPN.com wrote a column this weekend asking where the combination of Kevin Garnett, Paul Pierce, and Ray Allen rank in the NBA. Specifically, is there a better trio in The Association?
In Stein’s view, here are the top trios in the league.
1. Tim Duncan, Tony Parker and Manu Ginobili
2. Steve Nash, Amare Stoudemire and Shawn Marion
3. Kevin Garnett, Paul Pierce and Ray Allen
4. Dirk Nowitzki, Josh Howard and Jason Terry
5. Carmelo Anthony, Allen Iverson and Marcus Camby
In sum, the Celtics Trio ranks third.
Back in August I wrote a column applying the Pareto Principle – 80% of output is produced by 20% of the population – to the NBA. This column revealed how many wins were produced by the top three players on each team. Table One reports this information again, this time with the trios ranked from best to worst.
Table One: The Top Trios in the NBA in 2006-07
Last season the top five trios were as follows:
1. Shawn Marion, Steve Nash, and Amare Stoudemire [53.3 Wins Produced]
2. Tim Duncan, Manu Ginobili, and Tony Parker [44.4 Wins Produced]
3. Ben Wallace, Luol Deng, and Kirk Hinrich [40.5 Wins Produced]
4. Dirk Nowitzki, Josh Howard, and Jason Terry [39.8 Wins Produced]
5. Jason Kidd, Vince Carter, and Mikki Moore [39.7 Wins Produced]
Where would the Celtics Trio rank on this list? We know what each player did on a per-minute basis last season. We also have some idea now after nine games how many minutes each player is going to play (assuming no injuries) this season. Putting this information together and we see that Garnett, Pierce, and Allen are expected to produce about 45 wins this season. In other words, the Celtics Trio looks to be as good as what the Spurs employ, but not quite as good as what we see in Phoenix.
I would note that after seven games the Celtics Trio were all playing better than what they did last year. I haven’t looked at the last two games, but I am beginning to think that if this trio stays healthy (and that’s a big if), Garnett, Pierce, and Allen might be the best threesome in the NBA before the season is over.
By the way, this is what Bob Ryan of the Boston Globe said last August (in the column that inspired the original Pareto Principle story) after this team acquired Garnett and Allen:
He went on to explain why Boston didn’t have the depth to compete for a title. He didn’t even seem convinced this team could make the playoffs. Is Ryan still whistling this tune? Yes, the Celtics aren’t going 82-0. But as I said before the season started, I fully expect Boston to contend for a title this season. And with my next column I will review why I think this was clear before this team started 8-1.
– DJ
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
Wages of Wins Journal 
Alex Chichilnisky
November 19, 2007
Doesn’t plus/minus explain 94% of wins since plus/minus of all players on the team kind of aggregate to offensive/defensive efficiency?
Pete
November 19, 2007
Alex,
That is correct. Any measure that sums to points scored and points allowed explains 94% of wins in some sense. I don’t even think it has to be “per possession” since that just introduces a scale factor that applies to both offensive and defensive efficiency and seemingly doesn’t influence the regression.
Pete
November 19, 2007
As I understand it (and I could be wrong), the 94% figure is really a Pythagorean expectation figure.
Alex Chichilnisky
November 19, 2007
I liked this post. Maybe the Nets’ top 3 will move up in the rankings this year with Jefferson healthy.
Alex Chichilnisky
November 19, 2007
Up until this year Billy Simmons would say Garnett doesn’t know how to win. simmons uses justifications like that for his picks for NFL games and always finishes below 50%, worse than flipping a coin. For the 2nd straight year his wife has higher accuracy than him.
andrew
November 19, 2007
Dave,
How have the cavs trio fared so far this year. It’s mostly lebron, but Z is playing huge again, and gooden has been no slouch.
Thanks,
Andrew
Tim
November 19, 2007
Gary Payton is apparently retiring, prompting John Hollinger to write a column calling him the fourth best point guard ever, behind Oscar Robertson, Magic Johnson, and John Stockton, in that order, and ahead of Bob Cousy, Jason Kidd, Steve Nash, Isaiah Thomas, Walt Frazier, and Tiny Archibald, in that order. Hollinger’s rankings appeared in a column last year alongside the rankings of several other ESPN experts:
http://sports.espn.go.com/nba/dailydime?page=dailydime-GreatestPointGuards
But rumors of Payton’s retirement cause Hollinger to revisit the issue in a column for which you have to have Insider status:
http://insider.espn.go.com/nba/insider/columns/story?columnist=hollinger_john&page=GP-071116
In the more recent column, Hollinger notes that he ranked Payton much higher than the other experts did, and attributes that to five factors: Payton played in the Pacific Northwest, he played at the same time as Michael Jordan, there is a bias against scoring point guards, Payton had no flash to his game, and Payton was the best defender of any point guard to play the game. Hollinger was particularly “shocked” that the other experts ranked Jason Kidd ahead of Payton, in light of what Hollinger considers Kidd’s poor offense.
I know that you think very highly of Kidd, and I as I recall you don’t think high-scoring point guard Isaiah Thomas was the best player on the championship Pistons, or even the second best player. I wondered what you thought of Payton. Do you think he is the best defending point guard ever, and if so does that show up in the statistics? Do you think he is underrated because he was a scoring point guard?
ESPN’s ranking of point guards was part of a series of rankings for different positions. Here’s greatest power forwards:
http://sports.espn.go.com/nba/news/story?id=2074360
Here’s greatest centers:
http://sports.espn.go.com/nba/dailydime?page=dailydime-GreatestCenters
I feel sure there were rankings for small forward and shooting guard as well, but I’m afraid I have been unable to find them. ESPN’s search function leaves something to be desired.
At any rate, may I suggest that at some point you may want to consider a series of articles about how you would rank the top ten greatest players at the five different positions? If you have done so already, I missed it.
In the meantime, in light of Payton’s imminent retirement, do you care to assess his career?
Oren
November 19, 2007
I wonder if anyone else thinks it would be interesting to see what effect the Boston Trio has had on Perkins and Rajon Raldo. I bet a lot of people would have assumed that their numbers were going to drop this year.
Chirstopher
November 19, 2007
Hi Dave, Hollinger has his Top 5 PER players out now. I’d be curious who the Top 5 WP48 players are?
dberri
November 19, 2007
Andrew,
I will take a look at the Cavs.
Tim,
A post on Gary Payton will come this week.
Chirstopher,
I can’t do all teams at once. Jason Chandler’s Win Score site might be up in the next few weeks. Also other people have proposed a way to have updated WP48 stats. For right now, though, all you get are my team-by-team looks.
Alex Chichilnisky
November 19, 2007
Pete, I think it has to be per possession to control for the pace of the game. So taht adjustment matters.
Pete
November 19, 2007
Whether or not the regression is per possession , does **not** matter. You will still explain 94% of wins. And it is just a restatement of Pythagoras’ expectation formula.