I think I've seen 4 different models and they all have BKN as finishing 3rd or 4th including ESPNs (see link below). The reality is all these win prediction models are based on average margin of victory and may be adjusted for strength of schedule, etc.
http://espn.go.com/nba/story/_/page/BPI-Playoff-Odds/espn-nba-basketball-power-index-playoff-odds.
And from an Insider piece on the Cs today:
Scoring margin is often the most meaningful and predictive column in the standings, and it tells you more about what a team has done than its win-loss record. Early in the season, it's not uncommon for a team's W-L record to tell a very different story from its margin of victory.
Boston has the best scoring margin in the Eastern Conference, while Chicago's is just ninth best.
The Pythagorean method allows us to use points scored and allowed per game to project expected wins. Based on that, eight teams -- more than a quarter of the league -- have won either two more or two fewer games than we'd expect.
As you might expect, the Celtics are one of those teams [who has won two fewer than expected] There is another model that has them winning 3 less games than expected. In the end it tends to even out, but you can get teams like last years Nets who won a lot of close game, but should have won 33 games] outside that group.
There is another model that has the Cs winning 3 less games than expected. In the end it tends to even out, but you can get teams like last years Nets who won a lot of close game, but should have won 33 games. And if you looks at how well average margin of victory predicts how many games you win, it's clear that there's a decent correlation unlike many other factors.
The one problem with the models, however, is that the Nets have no incentive to tank and other teams above them may in the last month. OTOH, there are a lot of teams that look to be chasing the playoffs this year, there are teams out of the tanking business this season (like the T-Wolves), and there are fewer bottom dwellers like the Nets, Lakers, Sixers, and maybe Denver. The other favorable factor is that the Nets have been relatively injury free and have a very thin roster.
As for last night, I can't figure out what's wrong with Harden and the Rockets, but he's not the same guy who deserved the MVP last year for carrying that team. You're also not going to get 22 points out of Iso Joe Johnson many times going forward. His scoring average has fallen for the last 4 season including this year and his total shooting % which includes 2PFG, 3PFG, and FT% is 7% lower than his career average (which is also the league average). If he's not toast, at the age of 34 he's doing an outstanding impression of toast.