Does Points Differential = Degree of Greatness?

[Solid D]

Doug Collins said during the Spurs broadcast the other night that coaching legend Hubie Brown had told him that the one statistic to watch to determine how good a team really is...is Points Differential.

Here is how the current ranking by Winning Percentage versus Points Differential lines up, through February 5, 2006. Does Points Differential really indicate how good a team is, or is this statistic over-rated?

TEAM..W-L Win%..................................Points Differential
1. Detroit 39 7 0.848.............................1. Detroit +8.65
2. Dallas 37 10 0.787.............................2. San Antonio +6.74
2. San Antonio 37 10 0.787.....................3. Dallas +6.40
4. Phoenix 31 16 0.660...........................4. Phoenix +6.14
5. L.A. Clippers 28 17 0.622.....................5. Miami +4.08
6. Miami 29 19 0.604..............................6. Cleveland +3.30
7. Cleveland 27 19 0.587.........................7. L.A. Clippers +2.86
8. Memphis 26 20 0.565..........................8. Memphis +2.67
9. New Jersey 24 21 0.533.......................9. Indiana +1.93
10. Denver 26 23 0.531..........................10. L.A. Lakers +1.27
11. Milwaukee 24 22 0.522......................11. Washington +0.60
11. NO/Oklahoma City 24 22 0.522............12. Denver +0.49
13. Indiana 23 22 0.511 (.5111)...............13. New Jersey +0.17
14. L.A. Lakers 24 23 0.511 (.5106)..........14. Golden State +0.02
14. Philadelphia 24 23 0.511 (.5106)..........15. Minnesota -0.52
16. Washington 22 23 0.489....................16. Chicago -0.67
17. Utah 23 25 0.479...........................17. Philadelphia -0.80
18. Golden State 21 25 0.457..................18. Sacramento -0.85
18. Minnesota 21 25 0.457....................19. NO/Oklahoma City -0.91
20. Chicago 20 26 0.435.......................20. Houston -1.61
21. Sacramento 20 27 0.426...................21. Milwaukee -1.84
22. Orlando 19 26 0.422.......................22. Orlando -2.22
23. Seattle 19 28 0.404........................23. Boston -2.58
24. Boston 18 30 0.375 ........................24. Utah -2.97
25. Houston 18 29 0.383........................25. Toronto -3.35
26. Portland 17 29 0.370.......................26. Seattle -4.55
27. Toronto 17 31 0.354.......................27. New York -5.19
28. Atlanta 14 32 0.304 .......................28. Charlotte -5.20
28. New York 14 32 0.304.....................29. Atlanta -5.54
30. Charlotte 12 36 0.250.....................30. Portland -6.30

[leemajors]

its acccuracy would probably be better determined at the end of a given year. anyone wanna dig and see if it correlates to nba championships?

[Phenomanul]

Another parameter to consider would be coaching strategies.... For example, does Popovich ever run up the score intentionally? When the bench takes over do they try and run up the score or do they tend to lose leads??

[FromWayDowntown]

I'm not sure that it's all that telling. A team that outscored its opponents by 20 in every win and that was outscored by 1 in every loss would have a really good point differential, even if it was 10 games under .500. That won't realistically happen, but it could.

It does seem to correlate pretty well with winning in the current NBA. I don't know that it will say much about how well a team will do come playoff time, but it would certainly seem that if you're on the positive side of the ledger, you're winning a lot of games.

[Ariel]

Only if all teams were playing all out all games at all times, and even then it would only be meaningful in the context of that particular season.

Since that's not usually the case -and even more so with the Spurs- I wouldn't put too much stock into it.

[JamStone]

Over the last two decades, the championship team, their point differential, and their regular season record:

2005: spurs +7.8 59-23
2004: pistons +5.8 54-28
2003: spurs +5.4 60-22
2002: lakers +7.2 58-24
2001: lakers +3.4 56-26
2000: lakers +8.5 67-15
1999: spurs +8.1 37-13
1998: bulls +7.1 62-20
1997: bulls +10.8 69-13
1996: bulls +12.3 72-10
1995: rockets +2.1 47-35
1994: rockets +4.3 58-24
1993: bulls +6.3 57-25
1992: bulls +10.4 67-15
1991: bulls +9.0 61-21
1990: pistons +6.0 59-23
1989: pistons +5.8 63-19
1988: lakers +5.8 62-20
1987: lakers +9.3 65-17

[Solid D]

Good stats JamStone...

In 2004-05, here were the top 10 teams in Points Differential:

1. San Antonio +7.8
2. Phoenix +7.1
3. Miami +6.5
4. Dallas +5.7
5. Houston +4.1
6. Detroit +3.8
7. Seattle +2.3
7. Memphis +2.3
9. Sacramento +2.1
10. Denver +2.0

Most of the stats ring true, excepting Detroit and Denver who may have had numbers somewhat effected by suspensions and a coaching change respectively.

[Solid D]

In 2004-05, here were the top 10 teams in Points Differential:

1. San Antonio +7.8
2. Phoenix +7.1
3. Miami +6.5
4. Dallas +5.7
5. Houston +4.1
6. Detroit +3.8
7. Seattle +2.3
7. Memphis +2.3
9. Sacramento +2.1
10. Denver +2.0

Most of the stats ring true, excepting Detroit and Denver who may have had numbers somewhat effected by suspensions and a coaching change respectively.

Also, of interest, all 8 Western Conference Playoff teams were in the top 10, but only Detroit and Miami from the Eastern Conference.

Edit: Additional note, points diff. held true on the playoff matchups, with the exception of Detroit and Miami.

SA def. Denver
PHX def. Memphis
Dallas def. Houston
Seattle def. Sacramento
SA def. Seattle
PHX def. Dallas
SA def. PHX

Detroit def. Miami

[PM5K]

1996: bulls +12.3 72-10

Damn....

[Rummpd]

Spurs were over 10 PPG before Duncans injury last year.

[velik_m]

teams that win more will have better point differential on average, it's as simple as that.

[FreshPrince22]

No, the numbers are completely thrown of by the fact that the very good teams play scrubs a lot in garbage time. And I'm sorry, but I can't see how having your scrubs run up (or down) the score make the actual team any more or less "great".

[Phenomanul]

No, the numbers are completely thrown of by the fact that the very good teams play scrubs a lot in garbage time. And I'm sorry, but I can't see how having your scrubs run up (or down) the score make the actual team any more or less "great".


Are you talking to me??? (in a De Niro accent)

No... seriously I guess what I was trying to say was that if obtaining high-pt differntial was a team goal then the coaches strategy would be geared towards highlighting that statistic...

But back in reality all coaches know that they don't hand out awards for this statistic. The goal is still obtaining a championship, and therefore the strategies are geared with this objective in mind.

My point was that based on this phenomenon any correlation would be automatically skewed.

[Peter]

The differential stat I like is assists per game. The Spurs have traditionally been near the top in average assists per game allowed and average assist differential, while their average assists per game have been, well, average, something which can be attributed by their relatively low scoring outputs in games. The assist differential gives you a measure of the the Spurs ability to execute their offensive game plan versus that of their opponents. Granted, that shows up in the points scored differential, but since the assist stats are based solely on made field goals, I think it is a better measure of the Spurs' defensive and offensive team skill.

[JamStone]

The differential stat I like is assists per game. The Spurs have traditionally been near the top in average assists per game allowed and average assist differential, while their average assists per game have been, well, average, something which can be attributed by their relatively low scoring outputs in games. The assist differential gives you a measure of the the Spurs ability to execute their offensive game plan versus that of their opponents. Granted, that shows up in the points scored differential, but since the assist stats are based solely on made field goals, I think it is a better measure of the Spurs' defensive and offensive team skill.


Interesting theory. Here are some of the top NBA teams and their assists numbers and opponents' assists numbers:

Team / apg / opp. apg / differential

PHOENIX 27.1 17.5 +9.6
DETROIT 24.3 18.7 +5.6
SAN ANTONIO 19.6 16.1 +3.5
NEW JERSEY 22.5 20.1 +2.4
DALLAS 17.8 17.1 +0.7
MIAMI 20.8 20.4 +0.4
MEMPHIS 19.2 19.0 +0.2
LA CLIPPERS 21.1 21.1 Even
CLEVELAND 18.9 20.4 -1.5


I don't know how much it tells you. I guess it somewhat gauges a team's ability to execute offense by sharing the basketball instead of playing one-on-one isolation basketball versus a team that relies on that one-on-one stuff in order to score. And, it also gives some insight on how well a team not only shares the ball on offense but denies easy baskets on defense and forces one-on-one moves. Interesting. Phoenix appears to be very good at sharing the basketball, and does not allow their opponents to get very many easy buckets.

[101A]

Last year I remember reading the three stats most likely to predict a champion are:

1. Pt. Differential
2. # of games won by 10 pts or more
3. Opponent Fg%

Last year, Spurs were dominant in all three. Too lazy to look up how they are doing this, but I suspect they are top 5 in each.

[E20]

I thought it was just a slippery slope, but the stats point toward something.

[boutons_]

Stats don't lie. Good teams have comparatively good stats, bad teams, bad.

Note the best teams are at or very near the top in key categories.

http://www.nba.com/statistics/sortable_team_statistics/sortable1.html#top

There are anomalies, eg: neither Spurs nor Pistons are near the top in RB diff.

Also, Utah sucks but have much better +RB than Spur and Pistons.

When the Spurs or Pistons shoot team FG of 38%, or when Brent and Michael shoot individual FG of 25%, you know that really, really sucks compared to the worst shooting team in the NBA averages 42.2%.