2017 NBA Draft Models

Home for all your discussion of basketball statistical analysis.
mrichards
Posts: 10
Joined: Thu Dec 08, 2016 12:26 am
Location: Minneapolis, MN
Contact:

2017 NBA Draft Models

Post by mrichards » Wed May 24, 2017 2:55 am

I know there is already a 2017 Draft Analytics thread (so Crow feel free to bump) but figured I would throw this one out to gear more towards rankings / ratings of players from a Draft Model perspective. I would love to get some comparisons with those sharing their draft model work publicly and/or debate any of the prospects showing up positively or negatively in Draft Models.

Anyway, here are our top 10 ratings / rankings for 2017 Draft Prospects from our Model called Peak NBA Statline Projection (PNSP) which at its base makes use of predictions on individual box score statistics to create a single value metric from 0 to 100:

1. Jonathan Isaac 97.1

2. Markelle Fultz 96.5

3. Caleb Swanigan 93.8

4. Lonzo Ball 91.0

5. Dennis Smith 89.4

6. Jayson Tatum 88.6

7. Zach Collins 83.6

8. De'Aaron Fox 83.2

9. Jordan Bell 83.0

10. T.J. Leaf 82.8

Here is a link to our full rankings / ratings of 2017 prospects http://model284.com/2017-peak-nba-statl ... ion-model/

Also, some of our additional Draft Models:

1. NBA Role / Potential Model - Predicting the probability of becoming an All-Star, Starter, Bench Player, and Non-NBA Player: http://model284.com/nba-role-probability-model-2017/

2. Similarity Scores - Weighted similarity scores: http://model284.com/similarity-score-tool-2017/

We have methodology articles too if you are so inclined to get into how we get to our predicted values or feel free to ask me any questions.

Cheers to Draft Season!

Crow
Posts: 6188
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2017 NBA Draft Models

Post by Crow » Wed May 24, 2017 2:06 pm

No problem with separate threads.

Read about PNSP.

Saw you include high school recruiting rankings. How big a weight? (Don't have to give exact weight I'd you don't want to but would you call it light or moderate?) Any study of indicator value of what college / coach they picked / which & which "level" (college performance or future pro prep) picked them (as an add on to HS recruit rank)?

Hoe do you team stats? Do you use strength of opponent? At game or season level? Any consideration of extra weight on games with / actual counterpart minutes against NBA prospects?

Shooting weighted heavily. Is usage level considered?

Correlations with mainline metrics checked. Closest to PER. Ok with that?

Physical measurables included. How many? Any use of Combine athletic tests?


By position. By college position or projected NBA position? Any strategy for dealing with NBA "playing small"... at PF or C?

Machine learning. Applied to individuals or also "player types"?

Has the model been run before other drafts or is it built / run retrospectively to date?

Plan on offering projections to Draft Express for anticipated draft analytics model? Blending your models? Using what principle? Any use of subjective / scouting for adjustments of player scores or draft ranks? How did your models do on projecting the young TWolves? Does how teams weight all-star vs starter weights depend on what they have, what the intermediate goals are? In general how much should teams shot for stars vs. get realistic value? LaVar Ball with light yard ahead highest all -star potential in draft class? Ever or by year 4? Second highest ever probability in your dataset to Durant?

Basically set with your methodologies or wrestling with additions / changes? What is on your to do or someday wish lists? Any interest in RPM or using what exists / new stuff to yr to project RPM or RPM similars within existing player similars? Any consideration for projection adjustments for college coach impacts on player stats?

Light on defense due to boxscore limitations. Any interest in trying to address this with strategies to guessestimate / incorporate more D?

Have you considered running role probability model customized to selecting team? Who got more / less projected value FOR THEM?

How would you describe your 2 principal components in the similarity model? Is that only model adjusted for SOS or just the only one where that is mentioned? Is there anything you learn about the college to NBA transition in general from your simulated annealing step?

Application of models to international players? Playing overseas... or in the NCAA environment? Is NCAA environment / resulting the same for internationals or do they need a third way translation path?

Significance of your site name?

Seeking NBA positions or direct consulting opportunities? Your website has a team... uses a team approach? Would you consider adding scouting talent or feedback / critique talent or coaching perspective or psychological assessment talent or anything else?

NBA betting model appears to be entirely at team level with no consideration of player or lineup matchups? True or not?

Ideas for types of math or math techniques that could be applied to NBA but not yet seen, to your eyes? Any techniques that could be taken from re-insurance brokerage business, say related to performance "risk"? Does volatility of player performance game to game (or by other filters, yr to yr, opponent type, home / road, tournament, etc.) interest you for projections? Is it actively considered right now?

I've been told by a certain figure in the world of basketball analytics that "question guys" aren't that rare or special. How did I do for 80 minutes of immediate feedback?

mrichards
Posts: 10
Joined: Thu Dec 08, 2016 12:26 am
Location: Minneapolis, MN
Contact:

Re: 2017 NBA Draft Models

Post by mrichards » Thu May 25, 2017 5:49 pm

Crow wrote: Saw you include high school recruiting rankings. How big a weight? (Don't have to give exact weight I'd you don't want to but would you call it light or moderate?) Any study of indicator value of what college / coach they picked / which & which "level" (college performance or future pro prep) picked them (as an add on to HS recruit rank)?
I would say it is a light weight in PNSP but moderate weight in the NBA Role Prob Model. We have looked at college / conference but didn't find anything meaningful.
Hoe do you team stats? Do you use strength of opponent? At game or season level? Any consideration of extra weight on games with / actual counterpart minutes against NBA prospects?
Right now we are simply using strength of schedule. We have explored incorporating Team wins and performance against top level teams but haven't found a solution that provides anymore predictive power.
Shooting weighted heavily. Is usage level considered?
Usage is not used in our models.
Correlations with mainline metrics checked. Closest to PER. Ok with that?
This is interesting. So intuitively it makes sense that our scale and aggregate method of NBA stat lines closely aligns with PER because both are built off basic box score statistics. While PER is not my favorite all encompassing metric, I feel better because our scale and aggregate of predictions on NBA Stat lines predicted "better" than just predicting on any of the all encompassing metrics RPM, SPM, VORP, PER, etc.
Physical measurables included. How many? Any use of Combine athletic tests?
Wingspan, Weight and Max Vert. Obviously, top prospects do not go to the combine these days so we grab Draft Express Measurements (if not too out dated) and impute the remaining players.

By position. By college position or projected NBA position? Any strategy for dealing with NBA "playing small"... at PF or C?
Position is difficult given the position-less NBA. But we do incorporate to account for differences in statistics across positions. We are using their projected NBA position and for the most part use Draft Express's position assignment of players although we have gone back historically and adjusted players to fit into the PG/SG,..,PF/C roles. So we have 9 position groups PG, PG/SG, SG, etc... I will say that shuffling Bigs through PF, PF/C, and C usually gives a similar PNSP score for a given player (within maybe 5-7 points) and similarly for PG, PG/SG, but big position jumps will have massive changes.
Machine learning. Applied to individuals or also "player types"?
Machine learning applied to individuals.
Has the model been run before other drafts or is it built / run retrospectively to date?
We ran it prior to the 2016 draft and those results are posted on our website, but retrospectively if we run them now they are slightly different obviously.
Plan on offering projections to Draft Express for anticipated draft analytics model? Blending your models? Using what principle? Any use of subjective / scouting for adjustments of player scores or draft ranks? How did your models do on projecting the young TWolves? Does how teams weight all-star vs starter weights depend on what they have, what the intermediate goals are? In general how much should teams shot for stars vs. get realistic value? LaVar Ball with light yard ahead highest all -star potential in draft class? Ever or by year 4? Second highest ever probability in your dataset to Durant?

We would be open to offering projections to Draft Express if that is opportunity (although we do not have a model for international players yet). We could incorporate Draft Express rankings into our Model but opt for separate models. I wouldn't be opposed to incorporting subjective rankings with someone that I liked their scouting approach. Models really liked Karl Towns, saw Andrew Wiggins as a solid starter, did not particular like Zach LaVine nor Kris Dunn, but did like Tyus Jones as a solid NBA player. Definitely apart of team construction in terms of NBA Role Probs and Risk / Reward on prospects. The outliers are interesting - here are the top 5 All Star probs in our retrodictions:
1. Kevin Durant 85%
2. Ben Simmons 84%
3. DeMarcus Cousins 80%
4. Michael Beasley 79%
5. Joel Embiid 70%

Basically set with your methodologies or wrestling with additions / changes? What is on your to do or someday wish lists? Any interest in RPM or using what exists / new stuff to yr to project RPM or RPM similars within existing player similars? Any consideration for projection adjustments for college coach impacts on player stats?

We do feel strongly about our methodology but we do know its limitations and have a good sense for some of the "gaps" in the model but we will always wrestle with new ideas, additions and changes. Remember, "All Models are wrong, some are useful." We are working on our version of RAPM that should be primed for next season as well team projections, draft pick values related to our draft model, lineup construction, yr to yr NBA projections and a whole host of other NBA projects on our list.
Light on defense due to boxscore limitations. Any interest in trying to address this with strategies to guessestimate / incorporate more D?
Definitely, searching for ways to encompass a prediction on a defensive NBA measure to incorporate in the calculation of PNSP.
Have you considered running role probability model customized to selecting team? Who got more / less projected value FOR THEM?
Interesting, haven't explored.
How would you describe your 2 principal components in the similarity model? Is that only model adjusted for SOS or just the only one where that is mentioned? Is there anything you learn about the college to NBA transition in general from your simulated annealing step?
So we are using K means clustering to group players prior to running similarity scores. This groups players fairly nicely into Ball Handlers, Wings and Bigs (obviously not perfect). We just use the first two PC to visualize the data. More interestingly, the simulated annealing finds different weights for statistics by clustered groups. Most heavily weighted by position measured by correlation to NBA similarity:
Ball Handlers - Points, Steals, Assists
Wings- 2P Percent, Rebounds, Blocks, Wingspan
Bigs - 3P Percent, FT Percent, Assists
Application of models to international players? Playing overseas... or in the NCAA environment? Is NCAA environment / resulting the same for internationals or do they need a third way translation path?
In the works - currently work on something similar to what Layne Vashro put together a couple of years ago comparing leagues.
Significance of your site name?
Comes from a room at school where my buddy Sam and I first worked on sports analytic projects many years ago (oh, and where he ran the March Madness Model that successfully predicted the Shabazz Napier led 7 seeded UCONN Huskies to win it all).
Seeking NBA positions or direct consulting opportunities? Your website has a team... uses a team approach? Would you consider adding scouting talent or feedback / critique talent or coaching perspective or psychological assessment talent or anything else?
Most of the work we do in pairs or groups. I cannot speak for all of us but for now we will continue to publicly post our research while trying to add some humor for fun. Definitely, open to adding scouting, critiques, and definitely psychological assessment talent. We will get a podcast running in the near future but that will cover multiple sports.
NBA betting model appears to be entirely at team level with no consideration of player or lineup matchups? True or not?
True - plenty of room to explore in this area.
Ideas for types of math or math techniques that could be applied to NBA but not yet seen, to your eyes? Any techniques that could be taken from re-insurance brokerage business, say related to performance "risk"? Does volatility of player performance game to game (or by other filters, yr to yr, opponent type, home / road, tournament, etc.) interest you for projections? Is it actively considered right now?
Lots and lots of opportunities. Volatility is probably the most common term used in both avenues of work for us (especially in the betting world). We do already incorporate differing levels of volatility measures.
I've been told by a certain figure in the world of basketball analytics that "question guys" aren't that rare or special. How did I do for 80 minutes of immediate feedback?
Keeping me on my toes, love it.

Last caveat I will add, is that while predicting actual NBA box score statistics is difficult, we have found when scaling predictions by position they have been predictive of where a player lies in comparison to other players at a similar position in the NBA. Breaking down the models and identify those areas of under/over performance in a given statistic is very useful in prospect evaluation, IMO. We plan to breakdown some of the top prospects and pull out interesting parts of the Models for each prospect over the next couple of weeks.

Crow
Posts: 6188
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2017 NBA Draft Models

Post by Crow » Thu May 25, 2017 5:55 pm

Thanks for the replies.

Will check again for the 2016 draft info / analysis.

Look forward to more in future.

Crow
Posts: 6188
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2017 NBA Draft Models

Post by Crow » Thu May 25, 2017 6:11 pm

I see some 2016 draft articles, posted after the draft. I'll read them.

These are just random guys I had interest in, not a full review.

First name checked with similarity tool: D. Sabonis, top similarity score with Markieff Morris. Maybe in time. First season, Morris way better. Pretty much everyone on the top similarity list was way better in year 1, except Thomas Robinson.


Jaylen Brown... top sim Wiggins. Not wildly different but different in ways.

Brogdon / Richard Hamilton... Brogdon way better in first season.

Ben Simmons... tbd... but similarities all to big men. J Randle #1 similar but rest with very little assist-making even for bigs.

Jamal Murray? A guy I never heard of for #1, Galloway #2.


PSNP: Dunn projected weak, check. Baldwin projected one of the best? Not yet. D Murray? Not yet. Weak class.

Rd11490
Posts: 111
Joined: Mon Sep 29, 2014 4:54 am

Re: 2017 NBA Draft Models

Post by Rd11490 » Thu May 25, 2017 7:13 pm

Just for my own curiosity why simulated annealing as opposed to other optimization algorithms like MIMIC or HCRR.

mrichards
Posts: 10
Joined: Thu Dec 08, 2016 12:26 am
Location: Minneapolis, MN
Contact:

Re: 2017 NBA Draft Models

Post by mrichards » Sat May 27, 2017 2:40 am

Rd11490 wrote:Just for my own curiosity why simulated annealing as opposed to other optimization algorithms like MIMIC or HCRR.
Honestly, no particular reason. Could definitely opt for a different optimization algorithm but simulated annealing worked well in our exploration and had easy implementation so we stuck with it.

mrichards
Posts: 10
Joined: Thu Dec 08, 2016 12:26 am
Location: Minneapolis, MN
Contact:

Re: 2017 NBA Draft Models

Post by mrichards » Sat May 27, 2017 2:44 am

Crow wrote:Ben Simmons... tbd... but similarities all to big men. J Randle #1 similar but rest with very little assist-making even for bigs.
Simmons has few similar players which makes sense given his rare combination of size, rebounding, scoring, and playmaking. Randle happens to be the most similar but sim scores are relatively low compared to other players.

bchaikin
Posts: 286
Joined: Thu May 12, 2011 2:09 am

Re: 2017 NBA Draft Models

Post by bchaikin » Sat May 27, 2017 3:09 am

Simmons has few similar players which makes sense given his rare combination of size, rebounding, scoring, and playmaking.

how about evan turner, draymond green, and/or royce white?...

mrichards
Posts: 10
Joined: Thu Dec 08, 2016 12:26 am
Location: Minneapolis, MN
Contact:

Re: 2017 NBA Draft Models

Post by mrichards » Sat May 27, 2017 3:41 am

bchaikin wrote:Simmons has few similar players which makes sense given his rare combination of size, rebounding, scoring, and playmaking.

how about evan turner, draymond green, and/or royce white?...
Those are players with the aforementioned skillet. Why are they not similar to Simmons (high level perspective) by this method?

Royce White - low FT percent, low athleticism, classified as a Big technically -> Simmons classified as a Wing barely (not perfect), scored less.

Draymond Green - shot threes, considerably lower 2P percent (strong weight for wings), low athleticism (i.e. Max vertical) relative to Simmons.

Evan Turner - physical measurements -> much smaller and less athletic than Simmons and did shoot a few threes.

bchaikin
Posts: 286
Joined: Thu May 12, 2011 2:09 am

Re: 2017 NBA Draft Models

Post by bchaikin » Sat May 27, 2017 4:43 am

fwiw what set simmons apart from the pack in college (majors, top 6-7 conferences) was he was a frontcourt player that not only scored and rebounded well - but that also passed for assists at a high rate (a very high rate for a frontcourt player). he scored at 22 pts/40min and rebounded at 13.5 reb/40min, and passed for 5.5 ast/40min...

if you watched LSU play you would know simmons often brought the ball up the floor, especially when tim quarterman was not on the floor. he lead the team in assists (158)...

combine the best college seasons of turner (09-10), green (11-12), and white (11-12), and you get 20 pts/40min, 5.8 ast/40min, and 11.7 reb/40min - and each of these players also played in the frontcourt yet lead their team in assists (or were a close 2nd) and often brought the ball up the floor...

jgoldstein34
Posts: 240
Joined: Fri Sep 30, 2016 6:38 pm

Re: 2017 NBA Draft Models

Post by jgoldstein34 » Sat May 27, 2017 5:03 am

I know we're a bit off topic here but man I'm so excited to watch Ben Simmons next year. His passing during summer league last year was enough for me to buy in.

Dr Positivity
Posts: 298
Joined: Thu Sep 20, 2012 6:44 pm

Re: 2017 NBA Draft Models

Post by Dr Positivity » Sat May 27, 2017 7:43 am

bchaikin wrote:fwiw what set simmons apart from the pack in college (majors, top 6-7 conferences) was he was a frontcourt player that not only scored and rebounded well - but that also passed for assists at a high rate (a very high rate for a frontcourt player). he scored at 22 pts/40min and rebounded at 13.5 reb/40min, and passed for 5.5 ast/40min...

if you watched LSU play you would know simmons often brought the ball up the floor, especially when tim quarterman was not on the floor. he lead the team in assists (158)...

combine the best college seasons of turner (09-10), green (11-12), and white (11-12), and you get 20 pts/40min, 5.8 ast/40min, and 11.7 reb/40min - and each of these players also played in the frontcourt yet lead their team in assists (or were a close 2nd) and often brought the ball up the floor...
This doesn't mention what I think was 2nd most impressive thing about Simmons stats after the assists, his ability to get to the FT line at 10.3 attempts per 40 minutes on only 13.4 FGA. Not even Harden had as good a college season drawing fouls. I think he's as close to being a stone cold lock as you can get in two areas for a PF, passing and FTA, and his upside in both those areas is all time level for a PF

bchaikin
Posts: 286
Joined: Thu May 12, 2011 2:09 am

Re: 2017 NBA Draft Models

Post by bchaikin » Sat May 27, 2017 5:05 pm

This doesn't mention what I think was 2nd most impressive thing about Simmons stats after the assists, his ability to get to the FT line at 10.3 attempts per 40 minutes on only 13.4 FGA.

carl landry in 06-07 (purdue) had similar numbers - the difference being landry got fouled at almost twice the rate per touch than did simmons...

rarely does a college frontcourt player who got very high touches/min on offense come into the nba and maintain those high touches. someone like grant hill is one of the few exceptions - yes he played in the frontcourt and was the pistons' primary ball-handler on offense but they also rarely employed a traditional PG (high assists), rather playing the likes of joe dumars, lindsey hunter, allan houston, jerry stackhouse, etc. in their backcourt..

the sixers have not 1 but 2 traditional PGs (relatively high assists) in mcconnell and rodriguez, meaning simmons would not be the primary ball handler on offense. so unless they start simmons in the backcourt (at 6-10 unlikely), or in the frontcourt but start guards like stauskas and henderson (non-PGs) in the backcourt, simmons won't be getting the high touches he got in college...

and rarely do players go from college to the nba and significantly increase their rate of getting fouled per touch, which is what simmons would have to do to continue getting to the FT line alot should his touches on offense drop...

high touches/min college frontcourt players like luke walton, danny ferry, kyle anderson, et al all saw their touches drop dramatically once in the nba...

so unless simmons is that rare exception like grant hill (which is quite possible) the odds are that as his touches on offense go down so will his FTA/min...

Dr Positivity
Posts: 298
Joined: Thu Sep 20, 2012 6:44 pm

Re: 2017 NBA Draft Models

Post by Dr Positivity » Sat May 27, 2017 7:53 pm

bchaikin wrote:This doesn't mention what I think was 2nd most impressive thing about Simmons stats after the assists, his ability to get to the FT line at 10.3 attempts per 40 minutes on only 13.4 FGA.

carl landry in 06-07 (purdue) had similar numbers - the difference being landry got fouled at almost twice the rate per touch than did simmons...

rarely does a college frontcourt player who got very high touches/min on offense come into the nba and maintain those high touches. someone like grant hill is one of the few exceptions - yes he played in the frontcourt and was the pistons' primary ball-handler on offense but they also rarely employed a traditional PG (high assists), rather playing the likes of joe dumars, lindsey hunter, allan houston, jerry stackhouse, etc. in their backcourt..

the sixers have not 1 but 2 traditional PGs (relatively high assists) in mcconnell and rodriguez, meaning simmons would not be the primary ball handler on offense. so unless they start simmons in the backcourt (at 6-10 unlikely), or in the frontcourt but start guards like stauskas and henderson (non-PGs) in the backcourt, simmons won't be getting the high touches he got in college...

and rarely do players go from college to the nba and significantly increase their rate of getting fouled per touch, which is what simmons would have to do to continue getting to the FT line alot should his touches on offense drop...

high touches/min college frontcourt players like luke walton, danny ferry, kyle anderson, et al all saw their touches drop dramatically once in the nba...

so unless simmons is that rare exception like grant hill (which is quite possible) the odds are that as his touches on offense go down so will his FTA/min...
For being undersized PF with average athleticism, Landry's free throw rate in the pros was pretty decent. I see Simmons FTA rate as exciting not just cause of the college track record but also because his talent level backs it up with a great combo of ball handling, athleticism and size

Post Reply