The Chicago Bulls in 1995-96 won 72 games, finishing with a winning percentage of 0.878. This mark currently ranks as the best regular season in history, surpassing the Los Angeles Lakers who finished the 1971-72 season with 69 victories (and a 0.841 winning percentage).
After 60 games the Dallas Mavericks have won 51 times, for a winning percentage of 0.850. With this record the Mavericks are on pace to win 70 games. Given that this record ranks among the all-time marks in NBA history, clearly the Mavericks are the best team in the NBA this season.
Although that may seem obvious, John Hollinger disagrees. His NBA rankings place the Mavericks behind the San Antonio Spurs. The Spurs have played 61 games, but with only 43 victories, are only on pace to win 58 contests. Yes, the Spurs are on pace to finish 12 games behind the Mavericks. So how can the Spurs be ranked number one?
A couple of months ago I discussed the Hollinger’s rankings. Although his formula includes a number of factors, the key issues are offensive and defensive efficiency. Prior to last night’s games, the Spurs were scoring 106.5 points per 100 possessions while only allowing 97.1. Hence their efficiency differential (offensive efficiency – defensive efficiency) stood at 9.3. The Mavericks had a more efficient offense, scoring 109.4 points per 100 possessions. But per 100 possessions, Dallas was allowing 100.6 points. This results in an efficiency differential of 8.9.
Of course, the difference between these marks is rather small. Clearly both Dallas and San Antonio are very good this year. To see how good, in the last fifteen years only five teams managed an efficiency differential as good as the Spurs and Mavericks this season. These five were the
- Chicago Bulls (1995-96): 13.0
- Chicago Bulls (1996-97): 11.6
- Chicago Bulls (1991-92): 10.6
- Utah Jazz (1996-97): 9.4
- Seattle SuperSonics (1993-94): 9.2
So it should be pretty clear, the Spurs and Mavericks are each having very impressive seasons.
Of course there is a problem for the NBA. Each of these teams play in the Western Conference. Therefore, only one of these teams can appear in the NBA Finals. And given that the Phoenix Suns have a differential of 7.9, a mark in excess of any team in the NBA in 2005-06, it’s possible that both the Spurs and Mavericks will fail to make it to the NBA Finals.
If we look at the entire NBA (listed in the table below), the league is comprised of three top teams — Spurs, Mavericks, and Suns – and then everyone else.
Table One: Ranking the NBA in 2006-07
Given the offensive and defensive efficiency of these teams, all three are projected to win at least 60 games. No other NBA team is projected to win more than 53 contests. And since only 11 teams have a positive efficiency differential, the majority of NBA teams are currently projected to finish the season with a losing record. Unfortunately, the NBA’s playoff format requires the participation of 16 teams, which means that a number of average and below average teams will be appearing the in the post-season.
The main attractions in the playoffs will be the NBA’s three elite teams. But since all three reside in the Western Conference, we already know that the NBA Finals in 2007 cannot feature the NBA’s two best teams. In fact, if upsets occur out West – and in a short seven game series that is quite possible – it’s possible the NBA Finals may feature two teams that are less than elite.
Is there are remedy to this problem? Unless the NBA wants to ignore conferences and seed the playoffs in order of the best records, there is not much the league can do. As it stands, the best playoff match-up will probably happen in the Western Conference Finals, with the second best match-up occurring in all likelihood between the Spurs and Suns in the Western Conference Semi-Finals.
Out East, a collection of less than stellar teams will battle to face the best in the West. Currently, in terms of efficiency, the Chicago Bulls rank as the Eastern Conference’s best team. One should note, though, that in the 17 games since the Pistons passed the mid-point, Detroit’s efficiency differential has been 6.6. If this record holds, the Pistons should be favored over the Bulls in the playoffs.
And of course, that ignores what it would mean if Dwyane Wade – bad shoulder and all – could join Shaq on the court in Miami. Yes, the East looks least this year, but it’s possible that the Eastern Conference champion could put up a fight in the finals. Okay, not likely, but certainly possible.
– DJ
Wages of Wins Journal 
Bank
March 7, 2007
Power ranking should be based off win% and point differential as well as strength of schedule.
Bank
March 7, 2007
Or alternatively it could be based off wins produced and win%. To me, in line with Dave’s discussion of San Antonio and Dallas, winning ability at end of games shown by actual record seems like an additional ingredient that should be in the formula beyond the numbers in differential or wins produced.
Bank
March 7, 2007
(or using win score.)
disappointmentzone
March 7, 2007
It would be interesting to see if teams are actually able to “turn it on” in the playoffs, or if it is easier to buckle down and become a solid defensive team during a playoff push or during the actual playoffs or bring out the guns and become an offensive force.
Based on what I know about basketball it would seem to me that if either scenario is likely it is a team improving on the offensive end. Rarely can one person make a huge difference on defense (as far as I can tell) since players are pretty much limited to guarding only one man. Having a dominant big man might allow a single player to take away an actual portion of the court, but with the zone rules as they are there are still limitations in this area.
Conversely, only one person can score at a time, so it’s not hard to imagine a player being able to influence the outcome of a game more on offense than on defense.
Which is not to say that any of this actually happens, only that the Cavs are 22nd in offense and 4th in defense and it would be nice to hear that teams can improve on offense through stellar individual performances (LeBron!) more easily than on defense.
Jack Mott
March 7, 2007
dberri – do you have any thoughts on why the spurs can have the best efficiency differential but a significantly worse record?
Is it a strength of schedule thing? Some other factor at work, like a tendancy to beat bad teams really badly but lose to good teams by a little?
Maybe their coach keeps in top players in blowouts longer than the mavericks?
The Franchise
March 7, 2007
Popovich doesn’t keep his three stars (Duncan, Ginobili, and Parker) in a game for more than 35 minutes, almost regardless of how close the game is. This means that the Spurs are more likely to lose close games than they would be if they played every game with a win-maximizing strategy.
With an elite team that has a higher average age, this strategy seems especially wise to me. True, the chances of finishing with the top seed are diminished, but all a top seed got them last year was a down-to-the-wire series with Dallas that they ended up losing. (It just as easily could have turned out the other way, with how evenly matched the teams were.)
Since it seems like the regular season is more of a qualification for the playoffs than an actual season, and the top teams, like the Spurs, have little chance of missing the playoffs, such a regular-season strategy makes sense. However, for the teams around .500, who are in danger of missing the playoffs, they must play to maximze wins now.
Okapi
March 7, 2007
Jack Pot brings up a good point about a flaw in Pythagorean win expectaitons: “Maybe their coach keeps in top players in blowouts longer than the mavericks?”
(There’s the same problem in baseball where a team down by a lot of runs could have a greater tendency than other teams to use a really bad relief pitcher, allowing the score to run up. )
Assume we have a hypothetical basketball team that plays 3 games and wins 2 of them. The scores are 103-101 (w), 102-100 (l), and 150-100 (l). From that sequence of scores a 2-1 record seems consistent with performance. However, the team scored an average 118 pts and gave up an average 101 pts. Win expectation would generate a win % of over 90%. Wouldn’t it be better to, using the win expectation coefficients, assign a win % per game such that the team gets 0.58 pro forma wins for game 1, 0.42 pro forma wins for game 2, and 1.00 pro forma wins for game 3?
It’s really an empirical matter of course, but I’ll put forth the hypothesis that my suggested methodology would produce better predictions.
–Okapi
dberri
March 7, 2007
John Hollinger has argued that offensive and defensive efficiency is a better predictor of future performance than winning percentage. I have not seen the study, but I will trust that this is true (until I see a study that says it is not).
All that being said, I am not sure that efficiency differential should be thought of as a power ranking. I am really just presenting this ranking to say that, despite differences in the records, the Mavs and Spurs look pretty similar. Are the Mavs winning more close games? Perhaps. Is that a skill they have developed? This I doubt. The differences in winning percentage might just be luck.
Bank
March 7, 2007
It could be either based on small sample this season. right now I’d say, influenced by much better record against top 10 teams, Dallas is the legit #1.
offseason
March 8, 2007
Maybe I’m wrong, but didn’t people say the exact same (wrong) thing about the East having no shot last year?
dberri
March 8, 2007
offseason,
I was one of them. Thought I would give it a shot again. One of these years I will be right.
Harold Almonte
March 8, 2007
I think the real key issue of Hollinger’s ranking is that thing he calls “strenght of squedule of last 25% of season”. A kind of updated power measure.
Kelley
March 8, 2007
The media here in Dallas explain the Mavs’ close games by claiming everyone they play is on their A game, trying to break the Mavs’ current streak or whatever.
I’ve seen a more thoughtful analyst claim the real reason is Avery Johnson is resting his top players an average of 3 to 4 minutes more per game than last year, which will translate as several extra games rest total per starter by the time the playoffs start.
Meanwhile several players (Dirk, Terry) claim in interviews after close games that they DO practice “closing out” (winning close games) and it’s good practice “for the playoffs.”
The Franchise
March 8, 2007
Kelley, that seems like very useful data–and it may be a good choice by the team to practice winning close games, so they set up conditions that will create more games that are somewhat close. It may also be a good decision by the Spurs to not be overly concerned by winning or losing a small number of games, so their strategy may result in more losses than they would otherwise have.
However, both teams trying not to overuse their players is important to maximizing their opportunity for success in the playoffs. The Utah Jazz also do this as well; Deron Williams complained at one point this season that he could easily play another five minutes in each game, but Jerry Sloan responded along the lines of “you can, but your career will be shorter; look at how Stockton lasted an extremely long time as a point guard.”