Kevin Pelton, SUPERSONICS.COM | Nov. 2, 2004
If you're anything like the average NBA fan, you probably had the Utah Jazz and the Milwaukee Bucks winning about 40 games last year. Combined. Instead, both teams topped 40, Utah hitting for 42 wins and Milwaukee 41 and a playoff berth.
The spreadsheet is boggled by Shaquille O'Neal's off-season workout regimen.
Andrew D. Bernstein/NBAE/Getty
|
My predictions a year ago, I'm not ashamed to say, were even worse. Yeah, I had Utah and Milwaukee last in their respective conferences, but so did everybody else. Not everyone else had Chicago and Orlando, the NBA's two worst teams, both in the playoffs and combining for 90 wins. Yikes.
There has to be a better way, right? (Well, you couldn't get much worse.) I'm giving it a shot.
Baseball fans may have heard of a Baseball Prospectus invention named PECOTA, which uses comparable players to project development for baseball players. PECOTA isn't the first baseball projection system, but it has demonstrably outperformed competitors.
It's also proven the most inspirational to basketball analysts. John Hollinger of CNNSI.com used a similar system to predict NBA players' development in his Pro Basketball Forecast 2004-05. Dean Oliver, a statistical consultant for the Sonics, has his own version, and so do I.
What I've done is to take my 2003-04 player ratings and update for this season using the most similar players at the same age by my similarity system (generally, it's the top 25, but less for the really old and really young because there are smaller groups to choose from). If they improved 10%, the player's rating improves 10%. If they stayed the same, the player's rating stays the same.
What I like about this method is that it is more nuanced than simply applying an age-based adjustment. For example, take a pair of former Miami teammates and small forwards, Caron and Rasual Butler. Rasual is only a year older, but his rating is projected to go down while Caron's is projected to go up, because the system senses Rasual was playing over his head last year, when he shot 46.3% on three-pointers.
At the same time, it can't take into account things like Shaquille O'Neal entering training camp in the best shape he's been in the past decade. Obviously, each NBA player is his own person, and we can only make the best educated guesses possible about how they'll improve or decline, not to mention the ratings themselves are up for debate.
For rookies who played NCAA Division I ball last year, I'm able to use my NCAA stats translations. The only players I rate subjectively are those who have not played regularly in the NBA and high-school and foreign rookies.
The next step is projecting playing time for all 29 teams. This is a guessing game, to be sure, but I've scoured rosters, preseason rotations and news stories on all the teams. I limited minutes to 12 players per team in the sake of simplicity and to try to limit the number of players whose ratings I have to totally guess on. I also tried to "project" injuries based on past injury history and current injuries (for example, Jason Kidd projects to play significantly less than 82 games).
Finally, I used an excellent tool sent to me by reader Erich Doerr to plug my ratings for each team into the actual NBA schedule and simulate 500 seasons, ensuring the wins and losses added up and the easier schedule East teams face was taken into account.
Below are the results - please don't laugh. If they're completely wrong, the good news is I can put all the blame on my Excel spreadsheet.
I've intentionally omitted the Sonics from this list for reasons of objectivity. They've been completely ignored in terms of standings and seeding.
Playoffs
These picks are mostly subjective, though I did take into account what the projection system had to say:
West
1. San Antonio over 8. Denver
2. Sacramento over 7. Minnesota
3. Utah over 6. Houston
4. Dallas over 5. Memphis
1. San Antonio over 4. Dallas
3. Utah over 2. Sacramento
1. San Antonio over 3. Utah
East
1. Indiana over 8. New York
2. Washington over 7. Boston
6. Miami over 3. Toronto*
4. Detroit over 5. Milwaukee
1. Indiana over 4. Detroit^
6. Miami over 2. Washington
1. Indiana over 6. Miami
* It's possible that Miami might have home-court advantage in this series. While the new divisional alignment guarantees the three division winners the top three seeds, it does not guarantee them home-court. I suppose this was the case before-hand, but I can't imagine it ever was an issue.
^ SI.com's writers like the East to come down to Indiana and Detroit in the Conference Finals, but this is extremely unlikely because of the new divisional alignment. Assuming either Detroit or Indiana is the top team in the East, the other would have to drop to seed six or seven to be able to play them in the Conference Finals. More likely, one is the first seed and the other the fourth seed, meaning a probable Conference Semifinals showdown.
NBA Finals
San Antonio over Indiana in six
Awards
MVP: Tim Duncan, San Antonio
I've projected most of the other top candidates - Kevin Garnett, Shaquille O'Neal, Kobe Bryant - to play on teams that fall somewhat short of expectations this season, for reasons we'll go into later. That leaves Duncan, who consistently plays around MVP level and should be on one of the NBA's top teams this season. If he could shoot better than last year's 59.9% from the free-throw line, that wouldn't hurt either.
|
Rookie of the Year: Emeka Okafor, Charlotte
To steal the Bobcats marketing slogan, Okafor in '04. The UConn forward enters the league already one of its most advanced defenders and an above-average rebounder, if his college performance (both subjectively and statistically) is to be believed. He should also get plenty of scoring opportunities on an expansion team. Dwight Howard will provide a strong challenge, as will Devin Harris, if he keeps the starting job in Dallas - and don't count out the Sonics own Nick Collison - but this is Okafor's award to lose.
|
Defensive Player of the Year: Bruce Bowen, San Antonio
This is a pick from left field (actually, small forward, to be technical), but recall that Gregg Popovich was furious with the tactics the Indiana Pacers employed to get Ron Artest this award last year (including providing potentially misleading statistics regarding Artest's individual defense), and I expect the Spurs to remember and voters to join them, especially if the Spurs play as well as expected.
|
Most Improved Player: Mike Sweetney, New York
This award is about impossible to pick because of the strange, inconsistent criteria the media uses to select it, but I'll give Sweetney a shot. No, he's not the starter in New York yet, but it's probably only a matter of time until he steals one of the frontcourt positions, and he's easily the Knicks top post-up threat. The numbers confirm Sweetney as a breakout waiting to happen.
|
Coach of the Year: Jerry Sloan, Utah
I'm cursed with regards to this award - my NBA pick last year (Doc Rivers) and my WNBA pick (Richie Adubato) not only didn't win Coach of the Year, they ended up winning the less prestigious "First Coach To Get Fired" award. Somehow, I don't think Sloan will make me three-for-three; his job security in Utah is about as good as it gets. The voters missed chances to reward Sloan each of the last two years, and I don't think there's anyway they can keep him from winning his first Coach of the Year honors if, as the spreadsheet says, the Jazz is the class of the Northwest Division.
|
Wondering why teams ended up rating where they did? Check out SUPERSONICS.COM's team-by-team commentary on the East and the West.
