NBA Draft Update: Joel Embiid Clearly Best Prospect

Denny Medley-USA TODAY Sports

vjl110 has updated his models to include the combine testing information, allowing us to look at his "Bust, Bench, Starter, Star" model. How do the prospects stack up following the combine?

Now that the combine and draft lottery have come and gone, it's time to update our draft board, as it were. Since I know nothing about college prospects, our draft board is vjl's various prediction models, which he once again has made available to everyone here.

Before we get to the models, however, the combine has happened and workouts are going on right now (Milt Newton is in L.A. where a lot of the players are working out under the auspices of various agents), and it changed a lot of things right in the area where the Wolves are slated to draft.  Guys like Zach LaVine (UCLA), James Young (Kentucky) and Nick Stauskas (Michigan) have joined players like Gary Harris (Michigan St.) and Rodney Hood (Duke) as wing players expected to go somewhere from late lottery to pick 20 or so.  The combine brought several guys into play for the Wolves (perhaps not for the better).  So what do the models think?

First, here are the top 40 prospects by EWP, which is expected win peak. This is a model he developed using basic box score stats, along with height, weight, and strength of schedule to predict the prospect's best season based on players from the last 30 years. We've looked at it before:

Name School Pos EWP HUM
Joel Embiid KANSAS 5 14.8 16.1
Kyle Anderson UCLA 3 14.2 9.3
Jordan Adams UCLA 2 12.1 8.4
Marcus Smart OKSTATE 1 11.4 11.7
Aaron Gordon ARIZONA 4 11.4 10.3
Noah Vonleh INDIANA 4.5 11.1 10.9
Jabari Parker DUKE 3.5 11 12.5
Elfrid Payton ULALA 1 10.9 8.2
Tyler Ennis SYRACUSE 1 10.6 9.8
Andrew Wiggins KANSAS 3 8 12.6
Mitch McGary MICHIGAN 5 7.7 7.3
Julius Randle KENTUCKY 4 7.6 9.9
Jarnell Stokes TENNESSEE 4 7.5 5.2
T.J. Warren NCSTATE 3.5 7.5 6
Javon McCrea BUFFALOST 4 7.4 4.5
Gary Harris MICHIGANST 2 7.4 8.1
Khem Birch UNLV 4.5 7 4.7
Sim Bhullar NMSTATE 5 6.6 3.9
Kendrick Perry YOUNGSTOWN 1 6.4 5.6
K.J. McDaniels CLEMSON 3 6.2 5.6
Spencer Dinwiddie COLORADO 2 6.2 5.5
Shabazz Napier UCONN 1 5.5 5.5
Troy Huff NDAKOTA 2 5.5 4.8
Isaiah Austin BAYLOR 5 5.1 2.2
T.J. Bray PRINCETON 3 4.8 4.2
Zach LaVine UCLA 2 4.7 4.9
Glenn Robinson '14 MICHIGAN 3 4.6 3.4
James Young KENTUCKY 2.5 4.6 4.4
Devyn Marble IOWA 2 4.6 3.9
Juvonte Reddic VCU 4 4.6 3.3
Nik Stauskas MICHIGAN 2 4.4 5.1
Taylor Braun NDSUB 3 4.3 3.7
Jerami Grant SYRACUSE 3 4.3 5.6
Scottie Wilbekin FLORIDA 1 4 3
Russ Smith LOUISVILLE 1 3.8 3.8
Okaro White FLORIDAST 3.5 3.8 2.3
Daniel Miller GATECH 5 3.5 2.4
Kendall Williams NMEXICO 1 3.5 2.7
Dwight Powell STANFORD 4 3.3 2.4
Isaiah Sykes CFLORIDA 2 3.2 2.7

The column next to EWP, HUM (short for Humble...or maybe humility. Up to you), uses the EWP results but mixes in a factor for scouting consensus as defined by the Draft Express and Chad Ford mock drafts.  The idea being that the scouts might very well be seeing something not captured in the numbers, and it is worthwhile to take a look at that.

We can quickly see that, for example, some of the biggest differences between the current scouting take and the EWP model are the UCLA guys Jordan Adams and Kyle Anderson (the EWP model likes them a lot better) and Andrew Wiggins (the scouts like him a lot better).

Pretty clearly, the model, whether smoothed out by scouting input or not, sees Joel Embiid as the best prospect in this class. By the humble model, he's 3.5 wins better than the next best guys, Wiggins and Jabari Parker (who EWP doesn't like nearly as much).  If something actually happens with the Cleveland pick and Kevin Love (which seems incredibly unlikely), I would be supremely disappointed if the Wolves didn't take Embiid.  It would require further roster maneuvering certainly with three centers already on the current roster, but at the top, you take the best guy.

Returning to reality, where the Wolves are picking 13th, I have to say that all of a sudden, those wing guys are not so appealing to me. EWP sees Stauskas, Young, and LaVine all late first/early second round talents in this draft, and Gary Harris as perhaps a mid-first rounder.  Meanwhile, the UCLA guys sit there with fantastic numbers but seemingly little or no interest for the Wolves (though Anderson did interview with them at the combine).

That brings us to vjl's "Bust-Bench-Starter-Star" model that assigns percentage likelihoods for each of those outcomes for draft prospects.  These predictions are derived from a regression analysis combining the EWP predictions with physical and athletic traits measured at the combine.*

Here are the top 40 players sorted by "Star" likelihood:

Name School Pos EWP HUM BUST BNCH STRT STAR
Joel Embiid KANSAS 5 14.8 16.1 2% 3% 58% 38%
Marcus Smart OKSTATE 1 11.4 11.7 2% 30% 36% 33%
Aaron Gordon ARIZONA 4 11.4 10.3 2% 27% 40% 31%
Elfrid Payton ULALA 1 10.9 8.2 4% 27% 42% 27%
Andrew Wiggins KANSAS 3 8 12.6 3% 48% 24% 25%
Kyle Anderson UCLA 3 14.2 9.3 6% 1% 68% 25%
Noah Vonleh INDIANA 4.5 11.1 10.9 3% 22% 51% 25%
Tyler Ennis SYRACUSE 1 10.6 9.8 8% 21% 51% 20%
Jabari Parker DUKE 3.5 11 12.5 7% 15% 59% 19%
Jordan Adams UCLA 2 12.1 8.4 11% 6% 66% 18%
Glenn Robinson '14 MICHIGAN 3 4.6 3.4 25% 40% 25% 10%
Jarnell Stokes TENNESSEE 4 7.5 5.2 13% 35% 43% 9%
K.J. McDaniels CLEMSON 3 6.2 5.6 17% 41% 33% 9%
Mitch McGary MICHIGAN 5 7.7 7.3 17% 30% 46% 7%
Khem Birch UNLV 4.5 7 4.7 17% 35% 42% 7%
Julius Randle KENTUCKY 4 7.6 9.9 18% 30% 46% 7%
Gary Harris MICHIGANST 2 7.4 8.1 22% 32% 39% 7%
Zach LaVine UCLA 2 4.7 4.9 31% 39% 25% 6%
T.J. Warren NCSTATE 3.5 7.5 6 25% 26% 45% 5%
Shabazz Napier UCONN 1 5.5 5.5 33% 38% 25% 4%
Markel Brown OKSTATE 2 1.8 1.3 64% 20% 12% 3%
Nick Johnson ARIZONA 2 3 1.9 50% 31% 17% 3%
Jerami Grant SYRACUSE 3 4.3 5.6 35% 35% 28% 2%
Devyn Marble IOWA 2 4.6 3.9 38% 35% 25% 2%
Patric Young FLORIDA 4.5 3.1 2.5 44% 33% 22% 2%
Nik Stauskas MICHIGAN 2 4.4 5.1 41% 35% 23% 2%
James Young KENTUCKY 2.5 4.6 4.4 37% 34% 28% 2%
Russ Smith LOUISVILLE 1 3.8 3.8 47% 34% 17% 1%
Isaiah Austin BAYLOR 5 5.1 2.2 33% 30% 36% 1%
Cleanthony Early WICHITAST 4 1.3 1.9 67% 21% 11% 1%
Jordan Clarkson MISSOURI 1.5 1.7 1.2 63% 24% 12% 1%
Cory Jefferson BAYLOR 4 1.7 1.2 60% 25% 14% 1%
Adreian Payne MICHIGANST 4 2.3 2.9 49% 31% 19% 1%
James McAdoo UNC 4 3.1 2.6 44% 34% 22% 1%
Jahii Carson ARIZONAST 1 1.1 0.8 73% 19% 7% 1%
Doug McDermott CREIGHTON 4 3.1 3.6 48% 34% 18% 1%
Dwight Powell STANFORD 4 3.3 2.4 44% 34% 21% 1%
Semaj Christon XAVIER 1 2.3 1.4 57% 30% 13% 1%
Cameron Bairstow NMEXICO 4.5 2 1.1 55% 29% 15% 1%
Kendall Williams NMEXICO 1 3.5 2.7 50% 33% 16% 1%

*For players who did not appear at the combine, vjl110 used estimates.

You can see that this way of looking at it helps a player like Wiggins, who is the 10th best guy via EWP, but when his fantastic athleticism (particularly his jumping ability) is factored in, he is the 6th most likely "star" guy in the draft.

One guy who measures out fantastically well no matter vjl seems to slice it is Kyle Anderson. This formula puts his chances at being at least a starting level player at 93%.  EWP sees him as the 2nd best guy.  Frankly, after briefly being talked out of him because of potential fit issues (he played a lot with the ball in his hands at UCLA, which he likely won't be able to do in the NBA--Evan Turner?) I am back to thinking he's well worth taking at 13.  Plenty of guys around that area don't work out every draft, few of them available at 13 (which he will be) offer these kinds of expectations based on previously successful models.

If Anderson is out of the mix, another player who looks like he might be available at the Wolves spot with good value is Tyler Ennis, the Syracuse point guard. Hopefully we have gotten the awful taste of Syracuse draftees out of our mouths and can consider him without bias; the system here suggests he has an excellent chance to be at least a starting level point guard at his best, and back up point is a position of need for the Wolves. I admit I'm a little concerned with his mediocre shooting in college, but he young, he got steals, and he didn't turn it over much.

Two players who have occasionally been mentioned with the Wolves that none of vjl's models can understand are Rodney Hood and Adreian Payne (Michigan St.).  Hood fails to make the top 40 in either the EWP model or the star model, and Payne barely makes the bottom of the 40 most likely "stars," and also is not in the top 40 for EWP.  Players of this quality are available in the 2nd round and even as undrafted free agents; spending a late lottery pick on one of them would be a huge waste of resources.

An interesting "battle" exists both in the models and in the actual draft between Aaron Gordon (Arizona) and Noah Vonleh (Indiana), two guys expected to be drafted somewhere in the 5-10 range. Both freshman big men, both young (born within 3 weeks of each other in 1995).  Right now, most mocks see Vonleh going first, likely because of his more refined offensive game. vjl's models sees little to choose between the two: Gordon is just slightly ahead in EWP, and is slightly more likely to become a star; Vonleh with the benefit of scouting input is a bit ahead in the humble model and is slightly more likely to be at least starter quality. Gordon looks like a four who could perhaps play some three, while Vonleh looks like a four who could perhaps spend some time at the five.

I bring these two up because it's not impossible that the Wolves will be in position to take one of them should they make a trade with the Celtics, who hold the #6 pick.  I tend to be a bit risk averse about these things so lean toward Vonleh, but I could easily be convinced I'm wrong.

We'll see who the Wolves bring in for individual workouts over the next few weeks. That might give us some clues as to the direction they are leaning.  I encourage everyone to check out all of vjl110's numbers themselves and draw their own conclusions.  It's a really valuable resource.

Where do you think the draft is heading for the Wolves?

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Canis Hoopus

You must be a member of Canis Hoopus to participate.

We have our own Community Guidelines at Canis Hoopus. You should read them.

Join Canis Hoopus

You must be a member of Canis Hoopus to participate.

We have our own Community Guidelines at Canis Hoopus. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker