NBA win predictions 2018-2019

Adapting my NBA season model for 2018-19

Posted by RDJ on October 16, 2018

Image is Arena by bartb_pt used under license CC BY-NC-ND

Here we go! I realize that I’m a day late, but I’m not cheating. I made sure to get the tables on github last night, so you can, like, totally verify that. It’s almost like the season starts earlier every year.

Gimme them predictions

Here we go. The method is exactly the same as last year. In fact, I only dug into this code again yesterday. I sure am glad it worked…

Eastern Conference

Team Projected Wins 2017-18 Wins Improvement Playoff Probability
Toronto Raptors 57.6 59 -1.4 99%
Philadelphia 76ers 49.4 52 -2.6 88%
Boston Celtics 48.7 55 -6.3 86%
Washington Wizards 42.4 43 -0.6 65%
Indiana Pacers 42.2 48 -5.8 65%
Milwaukee Bucks 41.6 44 -2.4 63%
Detroit Pistons 40.8 39 1.8 59%
Miami Heat 40.6 44 -3.4 58%
Cleveland Cavaliers 39.4 50 -10.6 54%
Charlotte Hornets 38.7 36 2.7 50%
Brooklyn Nets 37.0 28 9.0 42%
Orlando Magic 31.7 25 6.7 25%
New York Knicks 28.8 29 -0.2 15%
Atlanta Hawks 28.5 24 4.5 14%
Chicago Bulls 23.9 27 -3.1 4%

Western Conference

Team Projected Wins 2017-18 Wins Improvement Playoff Probability
Golden State Warriors 62.6 58 4.6 98%
Houston Rockets 55.0 65 -10.0 90%
Oklahoma City Thunder 51.4 48 3.4 85%
Utah Jazz 48.6 48 0.6 77%
Minnesota Timberwolves 48.0 47 1.0 73%
Denver Nuggets 47.6 46 1.6 73%
San Antonio Spurs 44.1 47 -2.9 60%
Los Angeles Lakers 43.8 35 8.8 58%
Portland Trail Blazers 41.2 49 -7.8 47%
New Orleans Pelicans 40.0 48 -8.0 43%
Dallas Mavericks 37.4 24 13.4 32%
Los Angeles Clippers 36.3 42 -5.7 28%
Memphis Grizzlies 31.5 22 9.5 14%
Sacramento Kings 26.2 27 -0.8 5%
Phoenix Suns 24.9 21 3.9 5%

The Warriors are very good

Yeah. Isn’t that wonderful.

Minnesota might still be solid

The tables here assume Jimmy Butler stays in Minnesota. I also ran the numbers without him (but not placing him on any other team). Minnesota drops 3 wins and 10% on their playoff probability. Even without Butler, they’re a good team. They might improve by subtraction, since Jamal Crawford put up poor BPM numbers last year in substantial minutes.

Except, at Zach Lowe said, Minnesota needs “an alternate reality where everything isn’t always ruined there”. Who knows what will happen here.

Bad players are as damaging as good players are beneficial

A lot of these teams are really dragged down by having a few weak players, especially if they’re expected to play significant minutes. The few teams that lack bad players really gain in the rankings. That’s partially how Toronto has such a great ranking. They don’t have any bad players. Danny Green is an underrated pick-up, who had a weak year last year while playing through injury. Having a line-up full of good players makes a team resilient to injuries. Since the simulation takes into account minute variation, such as if injuries occur, this doesn’t hurt Toronto as much as other teams because weak players can’t gain those minutes.

I was very surprised to see San Antonio this high up. They lack top talent now. But they also lack BPM bottom dwellers. DeMar DeRozen and Jakob Poeltl might not be Kawhi Leonard, but they’re solid players. The Spurs made the playoffs last year without Kawhi Leonard. Maybe they can do it again. Although, I’m skeptical. I think the model underrates New Orleans, because of how weak they are behind Anthony Davis and Jrue Holiday.

Log jam in the middle

Especially in the East, half the teams in the conference are only separated by a few predicted wins. This could lead to some close playoff races again. Even Cleveland has a chance. Maybe.