The Best Offenses, Best Defenses, Best Teams, and NBA Analysis

Posted on December 26, 2006 by



Which team has the best offense in the NBA? If we look at points scored per game the top teams heading in the Christmas break included Phoenix, Denver, Washington, Golden State, and the LA Lakers. As the likes of John Hollinger and Dean Oliver have noted (and as we mention in The Wages of Wins), points scored can be quite misleading. Because teams play at different tempos, measured by possessions per game, a team can accumulate points just by playing faster.

To accurately gauge a team’s offensive ability one needs to consider efficiency, or points scored per possession employed. When we take that step, we see a slightly different top five. Yes Phoenix is still the best offense and Washington still cracks the top five. But the remaining top five consists of Dallas, Detroit, and San Antonio.


When we look at team defense we see a story similar. In terms of points surrendered, the five stingiest teams include Houston, San Antonio, Cleveland, Detroit, and Orlando. When we look at how many points a team surrenders per possession, though, a slightly different story is told. San Antonio, Houston, Cleveland, and Orlando are also in the top five in defensive efficiency. But Detroit only ranks 16th in defensive efficiency while the Chicago Bulls now rank 3rd.

The most significant move in the off-season was Ben Wallace departing Detroit for Chicago. We now see how this move has impacted each franchise.

The Best Teams

Now that we see each team’s efficiency on each end of the floor, we can project how many wins each team will likely finish with 2006-07. As we note in The Wages of Wins, offensive and defensive efficiency determine wins in the NBA. And with a simple regression we can determine how these two efficiency metrics relate to wins. With this model in hand, and with data on each team’s efficiency levels so far this season, we can project the rest of the year in the NBA.

Before we get to these projections one must note that the forecast is based on what each team has done over about the first third of the season. If teams play differently the rest of the way the results will of course be different. And we know if a team changes its roster – which can happen when a player like Yao Ming doesn’t play – how it plays will most definitely change. Still, for most teams rosters will not change too much the rest of the way so we can start painting the picture of how this season will finish.

Okay, enough with the qualifications. Who were the best teams heading into the Christmas break? The answer can be found HERE. As one can see, the best teams are mostly found in the Western Conference. Six teams out West are projected to win at least 50 games. In the East, only two teams have this expectation.

Of the eight teams expected to hit 50 wins, only one is expected to eclipse the 60 win mark. And that team is the San Antonio Spurs. The Spurs have the best defense and the fifth best offense. Given these marks, the wins model projects that the Spurs will win 67 games, or about five more games than their current winning percentage would indicate.

The teams that should contend with the Spurs for the title include Dallas and Phoenix. Since all three reside in the Western Conference, at least one of these title contenders will not even make the conference finals.

Whoever does survive the Western Conference playoffs will face the best in the East in the NBA Finals. Right now the best team appears to be the Chicago Bulls. Chicago started the season with nine losses in its first twelve games. After that start, though, the Bulls have won 13 of 15 and in terms of offensive and defensive efficiency are now projected to win close to 55 games. Although the Bulls are not quite as good as San Antonio, their path to the finals is much easier. Only five Eastern Conference teams are expected to finish with a winning record.

It’s important to note again that injuries and trades will alter outcomes. One key injury has been in Miami, where Shaquille O’Neal has missed most of the season. Despite what the Heat did to the Lakers on Christmas day, Miami has played very poorly with Shaq on the sidelines. If he comes back healthy and begins to resemble the Shaq of old – and not just an old Shaq – Miami can clearly make significant noise in the playoffs.

Analyzing NBA Teams

Looking at teams in terms of offensive and defensive efficiency is hardly original to The Wages of Wins. The innovation we introduced is Wins Produced, which takes what we learn from the analysis of team efficiencies and uses this information to analyze individual players. In essence, we can link the projected wins of each team back to the players the team employs.

Consider the Houston Rockets. Given what the team did after 27 games, we would expect this team to win 53 or 54 games. Which players on the Rockets woud we expect to produce these wins?

If we look at the Rockets after 27 games we see a team whose Wins Produced sum to 17.5, which translates into a mark of 53.3 for the entire season. This is not a coincidence. The Wins Produced metric simply translates what we know about offensive and defensive efficiency to the evaluation of the players. So the summation of Wins Produced will match what the model of offensive and defensive efficiency predicts.

When we look at the individual Rockets we see a team led by Yao Ming. Had he stayed healthy he should have produced 14.6 wins this year. Unfortunately he is now hurt. The Rockets have also lost Tracey McGrady. Without these two stars, this team will have problems. In other words, with these players on the sidelines this team cannot expect to win 50 games.

Of course fans of the Rockets might look down the list and see Dikembe Mutombo. Mutombo’s Wins Produced per 48 minutes (WP48) actually exceeds Ming this season. Mutombo has been an amazingly productive player in his career. Heading into this year his career wins production stood at 214.4. His career WP48 stood at 0.296. Unfortunately his career began in 1962, which means he is currently close to 70 years old.

Okay, he’s not that old. But Mutombo is one of the few NBA players who began his career when the senior Bush was president. And in NBA years that makes him ancient.

Ancient NBA players simply can’t play 40 minutes a night. Right now Mutombo is playing less than 10 minutes per game. The Rockets have 48 minutes each game to allocate at center, and when Mutombo can’t play (which appears to be most of the time), who takes the center position for the Rockets? If it is Juwan Howard the team clearly is going to miss much of the production it has seen from Ming. In sum, as stated earlier, without Ming and McGrady on the floor the Rockets have little hope of hitting 50 wins.

That, though, is not the point of this post. What I wish to emphasize is that we can take the stats these players accumulate and relate these to the outcomes we observe for the team. And because player performance in the NBA is somewhat constant across time– at least relative to the baseball and football – once we understand past productivity we can paint fairly clear pictures of what we will see in the future. Yes, injuries and trades will change the story. But barring roster changes, what we have seen thus far is pretty close to what we will probably witness at the end of the season.

Now that we have accumulated data for this season we can now apply these tools to the analysis of each team. So for the remainder of the regular season look for posts examining each NBA franchise. Please note that I have a few teams I wish to look at first, but I am more than willing to take requests. So if there is a team you wish to know more about, post a comment and let me know. I will try and get to these as quickly as my research and teaching schedule allow.

— DJ