Every team in the NBA is a work in progress at three weeks into the season, but some need more time than others. The reason I've argued that it is premature to judge the Miami Heat is twofold. In part, it's because of what was covered in last week's study looking at how telling the early part of the season is for any team. However, it also reflects the fact that teams that change as dramatically as the Heat did over the summer presumably need more time to start clicking.
Is it possible to back up the assumption that teams who make major changes play better later in the season? To take a look, I returned to the schedule-adjusted point differential used in last week's study, this time considering a three-year span: 2007-08 through 2009-10.
To start, let's define what it means for a team to notably alter its lineup. Our primary criterion is going to be the percentage of minutes played by returning players. So far this season, here are the teams that have turned over the greatest percentage of their minutes.
New York .410
New Jersey .419
New Orleans .456
L.A. Clippers .481
That's five teams who have newcomers playing at least half their minutes thus far. Where's Miami? Because the team brought back so many role players, 54.0 percent of the Heat's minutes have come from holdovers, which puts Miami behind not only these five teams but also Washington (52.3 percent) and Golden State (53.4 percent).
Fifty percent sounds like a reasonable standard, and it results in the same average number of teams (five per year) over the last three years. The biggest makeover in that span was by the 2008-09 Los Angeles Clippers, who returned just 21.4 percent of the previous year's minutes.
The way the database I'm using classifies players, all their minutes are credited to the team with which they finished the season, which means a few teams don't really belong in this group. The 2007-08 Grizzlies and Heat and last year's Wizards made their big changes at midseason, making it impossible to compare performance early in the year to the full season. Taking those three teams out, how did the others fare? Specifically, let's compare schedule-adjusted point differential through the end of November to that at year's end.
Year Team Ret% Nov Final Diff
2010 Minnesota .462 -12.8 -9.1 3.7
2009 Charlotte .456 -3.8 -1.3 2.5
2009 Memphis .430 -7.0 -5.2 1.8
2008 Minnesota .363 -7.8 -6.3 1.5
2009 L.A. Clippers .214 -9.0 -8.4 0.6
2009 New Jersey .403 -3.0 -2.4 0.6
2010 Toronto .382 -2.0 -1.9 0.1
2009 Miami .488 0.6 0.4 -0.1
2009 Milwaukee .471 -0.3 -0.9 -0.7
2009 Sacramento .428 -7.2 -8.6 -1.3
2008 Boston .496 11.6 9.3 -2.3
2010 Detroit .485 -0.3 -5.0 -4.7
On average, these 12 teams were virtually the same in the month of November (-3.4 points per game) as during the full season (-3.3). For every team like the 2008-09 Bobcats, whose improvement over the course of the year presaged their leap into the playoffs last year, there's an example like last year's Pistons that got off to a strong start before collapsing (in part due to injuries). The quintessential counterexample might be the 2007-08 Celtics, who incorporated two Hall of Famers and multiple new reserves while getting off to an 8-0 start.
Let's go back to the example of the Heat, which reflects the issue with considering only minutes. Looking at returning possession usage gives a slightly different view. Here are the most changed teams in terms of their possessions:
New York .436
New Orleans .502
New Jersey .512
This makes more sense in many ways. While the Hornets may have revamped their bench, their attack is largely the same as it has been for years--centered around Chris Paul and David West. As a result, it makes sense that New Orleans has had fewer adjustments to make than Miami, which is trying to incorporate two high-usage players.
Trying to learn from the past with teams like this is difficult because there are few examples, most of them compromised by midseason trades. Over the last three years, I could not find a single unbiased example of a team that changed two of its three possession leaders (like the Heat) that was not already considered in the teams that turned over the highest percentage of their minutes.
The group of teams that brought in a new leading possession user without a midseason trade is also small. While sample size caveats apply, however, this group does lend some credence to the notion of an adjustment period.
Year Team Nov Final Diff
2010 Memphis -6.4 -1.4 5.0
2008 Charlotte -7.1 -4.5 2.6
2008 Minnesota -7.8 -6.3 1.5
2009 Milwaukee -0.3 -0.9 -0.6
Last year's Grizzlies, who added Zach Randolph, struggled badly in the month of November before playing competitive basketball the rest of the season. The 2007-08 Bobcats (Jason Richardson) and Timberwolves (Al Jefferson) never reached that point, but did make progress after the first month. Only the 2008-09 Bucks (Richard Jefferson) went backwards after November, and that can be attributed at least partially to the loss of Michael Redd to a torn ACL.
Technically, the Heat doesn't belong in this group, since Dwyane Wade remains the team's leader in usage. Still, there is some inconclusive evidence that Miami and other two teams who are incorporating new go-to guys (Minnesota, with Michael Beasley, and New York with Amar'e Stoudemire) might take some time to jell. If you're feeling generous, add the Clippers (Blake Griffin) and Wizards (John Wall) to that group--though, like the Timberwolves, they're going to need major improvement to become competitive.
The 2010-11 Pro Basketball Prospectus is now available in paperback form on Amazon.com. For sample chapters and more information, see our book page.
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.