NBA Role Probability Model 2017

The NBA Role Probability Model predicts the likelihood that a given college basketball player becomes an NBA All-Star, starter, bench player, or does not make the NBA. The model uses individual college basketball season-long box score statistics, team-level statistics (e.g. strength of schedule), physical measurements, high school scouting ranking, position, and age/experience to predict the probability of each NBA role. For more detail on this model, see here. This model is one of three pieces that we use to evaluate the NBA potential of college players, with the other two being PNSP and Similarity Scores. In the table outlined below, you can find our predicted probabilities of the 2017 NBA Draft prospects landing in each category.


Lakers fans. Lonzo Ball. Wow. With an 83.5% All-Star probability, Lonzo is in rare company of players such as Kevin Durant, Michael Beasley, Ben Simmons, Joel Embiid, and DeMarcus Cousins with an All-Star probability above 70%. As if we needed more evidence, these numbers serve as further confirmation that the NBA Draft Lottery is rigged in favor of big-market teams. Right behind LaVarr Jr. are PNSP favorites Jonathan Isaac and Markelle Fultz, both of which have high probabilities in both the All-Star and Starter category. If you are not on the Jonathan Isaac train yet, just wait till you see his top player comps from our Similarity Scores.

Josh Jackson, a consensus top prospect who falls outside of PNSP’s top 10, shows up with a fairly high All-Star probability and extremely low “bustability” (or-non NBA probability). Jackson’s most likely outcome (highest probability) is becoming a starter, or ultimate role player, which makes sense given his questionable shooting efficiency at Kansas. With that said, the NBA Role Probability Model does see All-Star potential if Jackson can develop his shot. Interestingly, Jayson Tatum has the highest starter probability among 2017 draft prospects, and, like Jackson, he has a very low bustability, but his upside appears somewhat limited, as our model gives him only the 17th highest probability of becoming an All-Star.

Dennis Smith Jr’s high All-Star probability and high bustability fit the bill of a high risk / high reward prospect, which may not be worth a top-five pick in an outstanding draft class. Two other players that fit that mold by the NBA Role probability standards are Zach Collins and OG Anunoby. Collins’s limited minutes in his lone season at Gonzaga creates a cloud of uncertainty; the tools are there, but we just haven’t seen it in a large sample size yet. OG Anunoby’s tantalizing physical profile make him a top defensive prospect, but there are real concerns over whether he will ever develop any semblance of an offensive game.

Also, pretty shocking to see Przemek “the Polish Unicorn” Karnowski and Vitto “Tito’s” Brown among the highest Non-NBA probabilities.

Stay tuned for the release of our 2017 Similarity Scores and other draft coverage leading up to June’s NBA Draft. And as always #TrustTheModel

Lonzo Ball83.5%12.7%0.1%3.6%
Markelle Fultz46.8%33.8%6.3%13.0%
Jonathan Isaac46.0%44.3%2.3%7.4%
Caleb Swanigan41.8%23.0%20.7%14.6%
Dennis Smith37.2%26.4%9.9%26.6%
Josh Jackson34.9%51.9%4.7%8.5%
Jawun Evans30.4%30.5%21.2%17.9%
Zach Collins29.2%36.9%10.2%23.6%
De'Aaron Fox28.9%52.1%6.0%13.1%
OG Anunoby28.7%27.4%21.9%22.0%
Dedric Lawson27.6%22.9%13.7%35.8%
Tony Bradley23.5%29.7%6.7%40.1%
Justin Patton22.3%55.0%9.1%13.5%
Tacko Fall21.7%20.8%27.2%30.3%
Donovan Mitchell20.6%29.3%34.5%15.6%
John Collins18.5%43.0%21.2%17.4%
Jayson Tatum17.6%58.8%14.6%9.0%
Edrice Adebayo16.6%45.3%26.8%11.3%
Nigel Williams-Goss16.2%42.9%15.1%25.8%
Johnathan Motley15.9%28.9%31.0%24.2%
TJ Leaf15.4%52.9%8.9%22.8%
Rawle Alkins15.3%29.3%25.8%29.5%
Harry Giles13.9%53.8%12.8%19.5%
Thomas Bryant13.6%23.2%42.2%20.9%
Isaiah Briscoe13.0%19.2%25.5%42.3%
Aaron Holiday13.0%31.4%40.3%15.3%
Sindarius Thornwell12.5%19.0%26.8%41.7%
Jordan Bell10.7%53.1%18.6%17.6%
Ike Anigbogu10.0%22.1%17.6%50.3%
Kadeem Allen9.5%13.5%41.0%36.0%
Frank Jackson9.3%25.6%36.8%28.4%
Cameron Oliver9.0%17.9%50.1%23.0%
Malik Monk8.9%48.4%34.0%8.7%
Moritz Wagner8.4%51.5%34.9%5.3%
Josh Hart8.2%35.3%35.9%20.6%
P.J. Dozier8.1%26.1%20.8%45.0%
Lauri Markkanen8.0%34.4%41.0%16.6%
Semi Ojeleye8.0%13.2%60.3%18.5%
Tyler Lydon7.9%30.3%41.7%20.1%
Kennedy Meeks7.7%29.7%26.1%36.5%
Ivan Rabb7.2%32.0%44.9%15.9%
Chris Boucher7.1%28.6%35.2%29.2%
Chandler Hutchison7.1%31.4%18.2%43.3%
Kyle Kuzma6.1%29.7%29.1%35.1%
Jarrett Allen6.0%34.5%10.4%49.2%
Justin Jackson5.9%29.5%35.0%29.6%
Luke Kennard5.9%24.8%44.2%25.1%
Wesley Iwundu5.2%22.1%35.2%37.5%
Derrick White4.5%16.9%30.4%48.1%
Jacob Evans4.5%23.6%51.2%20.7%
D.J. Wilson4.5%24.4%51.4%19.7%
Frank Mason4.3%18.7%48.9%28.1%
L.J. Peak4.0%26.1%44.4%25.4%
Drew Eubanks3.9%22.4%43.8%29.9%
Monte Morris3.7%30.6%29.1%36.6%
Devin Robinson3.6%29.5%29.0%37.9%
Nigel Hayes3.5%4.8%29.9%61.9%
Omer Yurtseven3.3%20.9%46.7%29.1%
Dillon Brooks3.2%26.6%45.4%24.8%
Isaac Humphries3.2%17.5%39.4%39.8%
Damyean Dotson2.9%16.4%45.2%35.6%
Melo Trimble2.8%19.6%32.1%45.5%
Alec Peters2.5%14.9%46.1%36.6%
Grayson Allen2.4%19.8%35.1%42.7%
Joel Berry2.4%19.6%40.8%37.2%
Malcolm Hill2.3%11.0%50.9%35.8%
Tyler Dorsey2.3%18.9%44.2%34.6%
Antonius Cleveland2.1%11.4%20.4%66.0%
Thomas Welsh2.1%26.6%28.9%42.4%
Przemek Karnowski2.0%11.3%36.3%50.4%
Isaiah Hicks2.0%14.8%37.4%45.8%
Jake Wiley2.0%19.9%28.5%49.6%
Deng Adel1.9%15.3%43.9%38.9%
Andrew White1.8%10.1%46.3%41.8%
Vitto Brown1.3%8.1%37.9%52.7%
Jaron Blossomgame1.1%13.8%35.5%49.5%
Kobi Simmons1.1%14.6%25.1%59.2%
Sviatoslav Mykhailiuk1.1%17.5%40.0%41.3%
Luke Kornet0.9%5.6%44.5%49.0%
Bronson Koenig0.4%6.2%43.3%50.2%
V.J. Beachem0.3%7.8%41.5%50.4%

*Data retrieved from Basketball ReferenceSports Reference, and DraftExpress.

Written by  Marc Richards and Jack Werner.

2 thoughts on “NBA Role Probability Model 2017

  1. Aw, this was an exceptionally nice post. Spending some time and actual effort to create a superb article… but what can I say… I hesitate a lot and never seem to get nearly anything done.

Leave a Reply

Your email address will not be published.