Posts Tagged ‘NBA Power Ranking’

NBA Power Rankings by “Ty Rating”: the rising Heat and the sinking Celts

February 17, 2012

Using the same formula, and the same gambling website (Statfox Sports), that I used to power rank the likely NCAA tournament field, I power ranked the National Basketball Association.

My NBA chart is set up a bit differently because I condensed three steps.  Instead of posting each team’s Win Score and Defensive Win Score, followed by the expected winning percentage and then the winning percentage the rest of the league is posting against the same schedule, and then the “Ty Rating” based upon that, instead I post below the “Comparative Win Score” the “Comparative Defensive Win Score” and the Ty Rating based upon the same.  Let me provide a quick example.

Example using the #20 Milwaukee Bucks

Below on the chart, the 20th ranked team is the Milwaukee Bucks.  Under “WS” the Bucks post a “-1.1”.  That means the Bucks Team Win Score is 1.1 points below the Win Score the rest of the NBA is posting against the same schedule.  Under “DWS” it says “-2.1”.  That means that the Bucks are allowing their Opponents to post Win Scores that are 2.1 points higher than the same teams have been able to post against the rest of the NBA.  (Defensive Win Scores that are indicated as negative mean a below average performance).  If you add the two numbers together, you arrive at “-3.2”.  You then divide that by 10 to arrive at “-0.32”.  This is the Bucks “absolute” Marginal Win Score, from which I can calculate their absolute winning percentage, which is their “Ty Rating”.  Essentially, it is the difference between the winning percentage the team has achieved versus the winning percentage the rest of the NBA has achieved against the same schedule plus 0.500.  So, while the Bucks expected winning percentage is 0.404% (11.7 wins and 17.3 losses — the team is actually 12-17), because the rest of the NBA is only playing 0.455% basketball against the same schedule the Bucks have played, the Bucks “absolute” winning percentage, or their “Ty Rating” is 0.449%, so its a little better.

Here is the chart:

NBA WS DWS Ty Rating
1 Miami 7.6 6.1 0.733
2 Chic 7.1 6.4 0.731
3 OKC 5.5 5.1 0.682
4 LA Lakers 3.4 5.5 0.653
5 Denv 6.8 1.1 0.636
6 LA Clip 4.9 2.7 0.631
7 Orlando 2.6 3.7 0.609
8 Dallas 3.1 2.9 0.606
9 Phila 0.4 4.8 0.591
10 Atl 3.4 1.1 0.579
11 San An 3.1 0.8 0.569
12 Port 0.8 2.6 0.557
13 Memp -1.5 2.8 0.524
14 Hous -1.3 1.7 0.509
15 Ind -1.8 1.9 0.504
16 Minn -1.1 1.1 0.501
17 Bost -4.9 4.8 0.499
18 Utah -0.6 -0.7 0.479
19 NOH  -4.9 1.9 0.451
20 Milw -1.1 -2.1 0.449
21 NY Knicks -4.6 1.1 0.441
22 Phoenix -0.8 -2.8 0.441
23 Clev -1.6 -2.2 0.438
24 Gold St 1.8 -6.1 0.429
25 Sacra -3.5 -5.9 0.343
26 Tor -7.1 -3.1 0.329
27 NJ Nets -5.7 -6.6 0.294
28 Detroit -8.7 -4.2 0.283
29 Wash -6.6 -7.4 0.265
30 Char -9.9 -9.2 0.181

NBA Ty Ratings

Heat and Bulls clearly the NBA elite

Its neck-and-neck between the Miami Heat and the Chicago Bulls for best team in the NBA.  The two teams also rank #1 and #2 in overall offensive efficiency (by which I mean relative Win Score), and they invert that order for #1 and #2 in overall defensive teams in the NBA as well (by which I mean relative Defensive Win Score).

Three teams surprised me with their placement.  The Lakers are a lot higher than I anticipated.  They may have some fight left in the Purple and Gold.  And on the other side, the Boston Celtics placed much lower than I expected at #17.  The Celtics still play top 10 defense, but without Kendrick Perkins, the team is really struggling on the boards, and that is costing them games.  The other team who placed much lower than I anticipated was the New York Knickerbockers.  However, as I discussed two posts ago, the addition of world famous PG Jeremy Lin, the Knicks have shored up a major weakness and may begin to ascend the rankings.

Another surprise was the Minnesota Timberwolves.  I knew they were playing much better this season, but it is actually their defense that is propelling them more so than their offense.  That is surprising.  The aforementioned Bucks seem to have been stuck in the #18-#21 power ranking range throughout the entire Scott Skiles/John Hammond administration.  That is disappointing, to say the least.

Finally, we have the putrid Charlotte Bobcats and almost-as-putrid Washington Wizards.  What is the thread that runs between each organization?  Michael Jeffrey Jordan was in a management position for each.  Bucks fans, we cannot be thankful for much, but we can be thankful for this:  Herb Kohl prevented Michael Jordan from bringing his eye for talent to Milwaukee.  Jordan makes Isiah Thomas look like Branch Rickey.

Finally, has anyone heard from PG John Wall?  I thought he was supposed to be such a game changer for the Wizards when they selected him number one overall last season.  He certainly has not been.  His career is heading toward oblivion, just as many of us predicted when he was drafted.

NBA Power Ranking (by Relative Winning %)

January 20, 2010

I have completed another of my objective NBA Power Rankings using the “Relative Winning Percentage” formula.

RWP compares each NBA team’s pythagorean* winning percentage at home and its pythagorean winning percentage on the road to the rest of the NBA’s pythagorean winning percentage against the same home schedule and the rest of the NBA’s pythagorean winning percentage against the same road schedule.  Each team is then ranked according to the average of the two comparisons plus .500%.  That is what I call its “Relative Winning Percentage”.

Sidebar on what the term “Pythagroean” means as used in RWP (explained by

W Pyth
Pythagorean Wins; the formula is G * (Tm PTS14 / (Tm PTS14 + Opp PTS14)). The formula was obtained by fitting a logistic regression model with log(Tm PTS / Opp PTS) as the explanatory variable. Using this formula for all BAA, NBA, and ABA seasons, the root mean-square error (rmse) is 3.14 wins. Using an exponent of 16.5 (a common choice), the rmse is 3.48 wins.  Why do I use this instead of just using straight winning percentage?  I want to eliminate luck to the greatest extent possible.  I’m trying to measure relative strength, not fortuity.

RWP Example:

Using Pythagorean numbers, NBA home teams beat Team X’s home opponents .673% of the time, whereas Team X beats them .796% of the time.  NBA road teams beat Team X’s road opponents .348% of the time, whereas Team X beats them .262% of the time.  Then Team X’s Relative Winning Percentage would be .518%.  Calculation: (.123 – .086)/2 + .500).

RWP is meant to be a simple and reliable method of judging the objective  strength of each team regardless of opponent or location.  It splits each team’s home and away calculations and weights them evenly to favor teams with more balanced strength.

The Movers this week

This week’s RWP Power Rankings are remarkably consistent with the rankings I did on January4th.  Cleveland maintains the NBA’s highest RWP, with Boston supplanting Atlanta in the number two spot.  The largest rise belongs  to the Utah Jazz, which moved up five spots to number 8.  The Sacramento Kings (18 to 22) and the Dallas Mavericks (8 to 12) tied for the largest drop.

One other note.  When you consider the fact that somewhere around 29% of all games (about 24 games) are decided by luck rather than strength, and that a team ought to win half of those games, then the New Jersey Nets could be one of the worst teams in history.  Its RWP of .126% would translate into 10 wins, less than the amount of wins a team should be able to fall ass backwards into (12).  That’s monumentally bad.

Here are the results.  I listed each team’s previous ranking in parenthesis.

NBA POWER RANKING (by Relative Winning %)

1. Cleveland Cavaliers (1)____________________(.678%)

2. Boston Celtics (3)________________________(.676%)

3. Atlanta Hawks (2)_______________________(.666%)

4. Los Angeles Lakers (4)____________________(.657%)

5. Orlando Magic (5)_______________________(.653%)

6. San Antonio Spurs (6)____________________(.639%)

7. Denver Nuggets (7)______________________(.626%)

8. Utah Jazz (13)___________________________(.599%)

9. Oklahoma City Thunder (10)________________(.597%)

10.  Portland Trailblazers (12)_________________(.590%)

11. Phoenix Suns (9)________________________(.586%)

12. Dallas Mavericks (8)______________________(.572%)

13. Houston Rockets (11)_____________________(.537%)

14. Miami Heat (14)_________________________(.521%)

15. Charlotte Bobcats (16)____________________(.510%)

16. Memphis Grizzlies (15)____________________(.504%)

17. Toronto Raptors (17)_____________________(.492%)

18. Milwaukee Bucks (19)___________________(.462%)

19. New Orleans Hornets (20)__________________(.441%)

20. New York Knicks (21)_____________________(.437%)

21. Los Angeles Clippers (22)__________________(.410%)

22. Sacramento Kings (18)____________________(.406%)

23. Chicago Bulls (27)_______________________(.405%)

24. Philadelphia 76ers (26)___________________(.403%)

25. Washington Wizards (25)_ ________________(.383%)

26. Detroit Pistons (23)_____________________(.373%)

27. Golden State Warriors (24)_______________(.368%)

28. Indiana Pacers (28)_____________________(.332%)

29. Minnesota Timberwolves (29)_____________(.215%)

30. New Jersey Nets (30)____________________(.164%)

NBA Power Ranking (by Relative Winning %)

January 5, 2010

I objectively power ranked the National Basketball Association using a comparative technique I call “Relative Winning Percentage”.

What is “Relative Winning Percentage”?

“Relative Winning Percentage” is a pretty simple calculation and its done like this.  First I split each team’s schedule into home and road games.  Then I compare each team’s pythagorean winning percentage at home to the rest of the NBA’s pythagorean winning percentage at home against the very same opponents.  Then I subract the team average from the NBA average to get the team’s comparative difference.  Next I do the exact same thing for the team’s road schedule.  Then I average the two differences and add .500% to that average to get the team’s “Relative Winning Percentage”.

Basically, Relative Winning Percentage tells you how each team’s performance compares to the performance of the rest of the NBA against the same schedule.  Thus it accounts for location and for strength of opponent.  The new twist is the home/road 50/50 split.  I decided to add that twist because I realized that even if you account for homecourt advantage many NBA teams simply perform at different levels at home than they do on the road.  Realizing that, I wanted to give equal weight to both because I think that doing so gives a truer picture of team strength and, more importantly, a team’s likely playoff performance.

(Calculation Example:  The Bucks Pythagorean Home Winning % is .578%.  Playing the same home schedule, the rest of the NBA has a Pythagorean Home Winning % of .580%.  The Bucks Pythagorean Road Winning % is .319%.  Playing the same road schedule, the rest of the NBA has a Pythagorean Road Winning % of .433%.  Thus the Bucks are (-.002) at home and (-.114) on the road, which averages to (-.058).  Add .500 to that and you get the Bucks “Relative Winning Percentage” of  .442% ).

With that explanation, here is the NBA Power Ranking as of January 4, 2010.  In my next two posts I will rank each team separately according to home and road strength.

NBA POWER RANKING (using Relative Winning Percentage)

1. Cleveland Cavaliers__________(.676%)

2. Atlanta Hawks_______________(.673%)

3. Boston Celtics_______________(.666%)

4. LA Lakers___________________(.663%)

5. Orlando Magic_______________(.653%)

6. San Antonio Spurs____________(.649%)

7. Denver Nuggets______________(.642%)

8. Dallas Mavericks______________(.619%)

9. Phoenix Suns_________________(.613%)

10. Oklahoma City Thunder_______(.584%)

11. Houston Rockets______________(.571%)

12. Portland Trailblazers__________(.567%)

13. Utah Jazz_____________________(.552%)

14. Miami Heat___________________(.513%)

15. Memphis Grizzlies_____________(.491%)

16. Charlotte Bobcats_____________(.483%)

17. Toronto Raptors______________(.450%)

18. Sacramento Kings_____________(.446%)

19. Milwaukee Bucks______________(.442%)

20. New Orleans Hornets__________(.417%)

21. New York Knicks______________(.401%)

22. LA Clippers___________________(.399%)

23. Detroit Pistons________________(.392%)

24. Golden State Warriors__________(.389%)

25. Washington Wizards___________(.383%)

26. Philadelphia Sixers____________(.380%)

27. Chicago Bulls_________________(.359%)

28. Indiana Pacers________________(.346%)

29. Minnesota Timberwolves_______(.230%)

30. New Jersey Nets_______________(.208%)