Star Power and the Washington Wizards

Posted on January 10, 2007 by


Production is defined as the transformation of inputs into output(s) [That’s a really bad lead sentence. I assure you this column gets better].

In basketball, the output we frequently focus on is wins. Inputs can be defined in terms of position — point guard, shooting guard, small forward, power forward, and center. It can also be defined – as I did last summer – in terms scorers and role players.

Today I want to present a third perspective, similar to scorers and role players, which I will call: Stars and Everyone Else.

Each team has a Star, which I will define as the player who leads the team in Wins Produced. And each team has Everyone Else, which I will define as Everyone Else (who is not the Star).

Pretty simple, huh?

Okay, perhaps we are going beyond simple to silly. But let’s see what story this approach can tell.

The following link takes you to a table that reports the productivity of the Stars and Everyone Else for each NBA Team in 2005-06.

Stars and Everyone Else in 2005-06

Consider the first team listed. The Spurs were led by Tim Duncan, who produced 16.9 wins. Everyone Else on the Spurs produced 42.3 wins. Per 48 minutes, Everyone Else on the Spurs produced 0.119 wins. An average player in the NBA produces 0.100 Wins per 48 minutes [WP48]. So Duncan’s teammates were collectively above average, and were actually the most productive teammates in the league.

Next on the list we see Dirk Nowitzki and the Dallas Mavericks, followed by Ben Wallace and the Detroit Pistons. Not coincidently, the top three teams in the regular season last year were the Spurs, Mavericks, and Pistons. Each of these teams combined a productive star with a solid supporting cast.

Fifth on the list (we are skipping the fourth place Grizzlies because they are not the subject of this post) we see Gilbert Arenas and the Washington Wizards (the actual subject of this post). The Wizards last year won 42 games, although their Wins Produced predicted the team should have had close to 46 wins (a mark quite similar to the Wins Produced of the Cavaliers in 2005-06, which tells us why a 42 win team posed such a challenge in the playoffs for a team that actually won 50 games).

When we look at the source of these 46 Wins Produced we see that Arenas produced 11.1. Everyone Else on this team produced 34.9 victories, for a WP48 of 0.102. In other words, the non-stars on the Wizards last year were the fifth most productive combination in the league. But this was not the fifth best team. Why? Because the star – Gilbert Arenas – was one of the least productive stars in the league (a point I made last October).

Now before the many fans of Arenas get upset, it’s important to remember that Arenas was still a very good player last year. In a league with 458 players, Arenas ranked 26th in Wins Produced. If we look at just the point guards, only Jason Kidd, Steve Nash, Chris Paul, and Chaucey Billups were more productive. So Arenas was very productive in 2005-06. But as a leading wins producer on a team, he was not quite as productive as one would like.

If we look at the Wizards in 2006-07 – which you can see HERE – we see a similar problem. After 33 games the leading wins producer is – just barely – Caron Butler. So far Butler has produced 5.6 wins and posted a WP48 of 0.203. Arenas has produced 5.3 wins and has a 0.194 WP48 (a mark that is still 5th among point guards behind Kidd, Nash, Paul, and Billups). So both Butler and Arenas are producing similar amounts.

Both of these players have also improved over last year, which is good news. The bad news is that the top five teams last year in Wins Produced – San Antonio, Detroit, Dallas, Phoenix, and Miami – all had a top player with a WP48 closer to 0.300. In other words, whether Butler or Arenas is the star of this team, neither is as productive as the top stars on the top teams from last year.

As it stands now, if the Wizards are going to win 60 games this year – and the top player (be it Butler or Arenas) maintains his current productivity – the remaining Wizards would have to have a WP48 of 0.133. Currently Everyone Else on the Wizards has a WP48 of 0.088. And as one can see from last year, no team had a collection of Everyone Else that came close to 0.133.

Let me tell this story with some different numbers. If your best player posts a WP48 of 0.300 and plays 80% of the minutes at his position, and if every other player on the team is average (WP48 of 0.100), then the team will win 54 games. If your best player has a WP48 of 0.200 – which is still very good – and the rest of the team is average – wins drop to 47.5. Generally 50 wins are the minimum necessary to be thought of as a serious title contender. And to achieve this total you either need a number of very, very good players – which is hard to assemble – or one really outstanding talent (and a collection of good – but not very, very good – players).

As I noted in reviewing the 2005-06 Wizards, this team has a nice collection of talent. Arenas, Butler, Antawn Jamison, and Brendan Haywood are all above average players. As expected, losing Jared Jeffries and adding DeShawn Stevenson did not help. Still, there are good players on this roster.

What is lacking, though, is that one elite star. And until the Wizards find such a player, this team will probably continue to fall short of the very best in the NBA.

One last bit of good news, though. Last I checked this team still played in the Eastern Conference. And in this conference, being the best is not a prerequisite for contending.

– DJ

Our research on the NBA was summarized HERE.

Wins Produced and Win Score are Discussed in the Following Posts

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say