Pages

A Case for New NBA Playoff Seeding

3/05/2008
This regular season looks to be one of the best in recent memory with the jockeying for position that's going on in the Western Conference. The West Conference playoffs look to be very exciting this year. In contrast, outside of Boston, Detroit and Cleveland, the Eastern Conference playoffs should be a major snooze-fest. Hopefully, we don't end up with the same scenario as last year where the playoffs peak prior to the Finals, only followed by a major letdown.

This problem might be solved if the NBA were to consider seeding the teams 1-16 without regard to conference. The simplest way would be to take overall win-loss record, but this might not be fair since teams play unbalanced schedules. If (hypothetically) Memphis were to go 41-41 in a division with San Antonio, Dallas, Houston and New Orleans, that would be considered more impressive than Washington going 41-41 in a division with Orlando, Atlanta, Charlotte and Miami. This would be a problem with a crude rank ordering.

One easy way around that problem is to use a computer ranking to adjust for the unbalanced schedule. Computer rankings get a bum rap for the most part because they are used poorly in other sports. For college football, it's difficult to rank order teams based on limited sample size and different schedules (how do you compare two 12-1 teams that may not have any overlap in opponents on their schedule). In college basketball, sample size is less of an issue but you still have unbalanced schedules and fuzzy rules around home/road/neutral court that can skew the results. Computer ranking might actually work in college hoops, but the one they choose to use (RPI) is deeply flawed and not very informative.

The NBA doesn't have these limitations because you have 82-games of sample size, each team plays each other team at least twice and even the best teams have at least 10 losses and worst teams have at least 9-10 wins, plus there are no neutral court issues. A computer ranking might actually work well for the NBA.

I've published my Colton Index rankings for college hoops for the last 5-6 years, until now I hadn't even thought about using it for the NBA. It is a pure 'body-of-work' measure, it just assesses whether one team's resume of wins and losses is more impressive than another team's. It only looks at wins and losses and the location of the game (running up the score won't help you). Unlike the RPI, a win will never hurt you (it just might not help you) and a loss will never help you (it just might not hurt you much). Essentially all it is doing is balancing out the unbalanced schedules so every team is on the same footing.

I converted my numerical rating for NBA purposes to the expected number of wins over an 82-game season assuming a balanced, round-robin schedule. Here's the rank order for the season to date:

Rk TEAM W - L JCI
1. BOS 46 - 12 63.81
2. DET 44 - 16 59.98
3. LAL 43 - 18 59.13
4. SAS 42 - 17 57.75
5. NOR 40 - 19 55.24
6. HOU 39 - 20 54.63
7. PHO 40 - 20 54.40
8. DAL 39 - 22 53.87
9. UTA 39 - 22 53.40
10. ORL 39 - 23 52.14
11. GSW 37 - 22 51.35
12. DEN 35 - 24 48.37
13. CLE 34 - 26 46.90
14. TOR 32 - 27 44.12
15. POR 31 - 30 43.68
16. WAS 29 - 30 39.61
17. SAC 27 - 33 37.45
18. PHI 28 - 33 36.38
19. ATL 25 - 34 34.77

In this scenario, the Nuggets (35-24) and Blazers (31-30) would replace the Sixers (28-33) and Hawks (25-34). Obviously, the playoff brackets would look a lot different as well.

Here's what last year would've looked like:

Rk TEAM W - L JCI
1. DAL 67 - 15 67.78
2. PHO 61 - 21 61.91
3. SAS 58 - 24 58.79
4. HOU 52 - 30 53.27
5. DET 53 - 29 52.00
6. UTA 51 - 31 51.87
7. CLE 50 - 32 49.06
8. CHI 49 - 33 48.30
9. TOR 47 - 35 45.90
10. DEN 45 - 37 45.53
11. MIA 44 - 38 43.01
12. LAL 42 - 40 42.93
13. GSW 42 - 40 42.90
14. LAC 40 - 42 41.04
15. NJN 41 - 41 40.22
16. WAS 41 - 41 40.20
17. NOR 39 - 43 39.79
18. ORL 40 - 42 38.90

The Clippers (40-42) would've replace the Magic (40-42) in the playoffs. Again, the brackets would've looked a lot different with the first four seeds coming out of the West.

So, I guess the question is would this be a worthwhile and doable change? Obviously there are trade-offs involved. Any cross-country travel concerns don't really seem to be an issue with charter jets and the playoff schedule being drawn out to begin with. The Eastern Conference would likely lose some playoff teams and obviously those teams are adversely impacted, but the Eastern/Western conference tilt has to be viewed as cyclical and would be expected to even out over the long run. Would the Sixers and Hawks have any legitimate beef over missing the playoffs? And how could that beef be greater than the one that Denver and Portland may have this year? In the long run, isn't it better to have the best 16 teams in the playoffs? The Western side of the playoffs should be great this year, but wouldn't be better for the league to have the best teams play each other later on in the playoffs and have more compelling match-ups throughout the entire bracket? Of course, the chances of this ever happening are slim since it would require the NBA to admit that something was wrong in the first place (something they don't seem too open to doing), but it's at least interesting to think about.

0 comments:

 
Wegoblogger #31 © 2011 | Designed by Bingo Cash, in collaboration with Modern Warfare 3, VPS Hosting and Compare Web Hosting