Objective Draft Grades for the 2015 NBA Draft

The NBA Draft. The Woj-Mas of the offseason. I think we anticipated a more active draft, but it was fun all the same.

I took a shot at assigning draft grades to the selections and other moves of all 30 NBA teams, and true to Nylon Calculus form, I took the most analytic approach I could to this somewhat subjective exercise. Using a consensus draft model ranking of several different draft models, models that included the projections of my own and also those of fellow Nylon-ers Layne Vashro and Andrew Johnson.[1. As well as Kevin Pelton, Neil Paine, Jesse Fischer, Steve Shea, Daniel Myers, Dan Dickey, and the mysterious Masseffectlynk.] I defined the value of each pick of the 2015 NBA draft as the projected output of the player the consensus model ranking would take if left alone to “auto-draft”. I used the consensus projections of all the models included, but scaled them into the target variable range of the variable I use to predict future NBA performance, a RAPM-WS two year peak blend. For reference, the top 6 players by my RAPM-WS blend metric are Chris Paul, Kevin Durant, Dwayne Wade, James Harden, Stephen Curry, and Anthony Davis, destroyer of worlds, with values between 25.5 and 17.5.[2. Includes only players drafted from NCAA or internationally since 2002.]

Why scale the consensus projections into my own target variable? I did this so I could estimate the value of the current players included in the draft day trades, without the target variables of my fellow model-producers. Basically, the projections used reflect the opinion of several different models, but the valuation of current players’ careers is my own. The two are combined as appropriately as practically possible.

To grade each team objectively, I followed a few basic rules. Each draft pick was given the model consensus’ preferred player projection value, and this set the baseline on the amount of value a team was expected to receive back from their assets, whether by drafting a player, or by trade. Selecting a player that the model projected to be of higher value then his draft slot, would result in a net expected profit, a pattern of which would manifest itself into a higher grade. Drafting below the designated value of the draft slot would produce the opposite effect. I did not define the draft picks by their historical results because each draft is unique, and this draft in particular projects to be particularly deep.

Thankfully for this exercise, there was little trade movement, which means there were less draft projections to compare to current players at different stages of their careers, and less future first and second rounder to crudely estimate value for. As a rule, I estimated the value of future first and second rounders to be worth 5 RAPMWS blend and 1 RAPMWS blend respectively, as this was the average amount of projected value NBA GMs picked in the 2015 draft. On average, NBA GMs left roughly about a second round pick of projected value (1 RAPMWS Blend) on the table this year. Of course, this estimate is defined by draft models, and draft models are not the end-all be-all. There are many other factors that go into drafting that draft models do not consider, factors like character, injuries, and mental makeup, and draft models are far from perfect.[3. #WatchTheGames.]

Using these and the mean performance of all GMs, I generated a raw letter grade. This letter grade reflects, in a completely objective manner,[4. Ed. This isn’t to say completely correct manner, rather that Nick didn’t introduce any of his own subjective opinions about players or moves in the initial grading process. His subjective adjustments have been separated out as described.] how much the GMs returned on their assets, relative to expectation as defined by draft models and the profit-returning performance of the rest of the league.

Given that I follow the draft a little bit, and also that I know the flaws behind these numbers, I gave each team’s raw grade a subjective tweak (or not) before deciding on the final grade.

Below is a table detailing the final grades, raw grades, adjusted profit number grade, and raw profit.

Golden State Warriors

Pure Score: 3.220

Pure Grade: A+

Adjusted Grade: A+

People wonder why the champs are the champs. The entire league allowed 7’4” arms, 41.5% 3PT shooting, 2.7 blocks and steals per 40, and a top ten national high school ranking in a 19 year old package fall to 30 overall. Looney may have a hip issue, but he was statistically considered the 11^th best player available. Excellent job by Bob Meyers and staff, even if by default.

Minnesota Timberwolves

Pure Score: 1.040

Pure Grade: B+

Adjusted Grade: A

It’s hard to grade a first overall selection, especially when it’s totally expected. Karl Towns was not my model’s, nor the model consensus’ first overall pick (that honor belonged to D’Angelo Russell), nor was my favorite player subjectively. But he was a really close second in all of those regards. And the consensus “subjective” pick. So if a NBA front office is going to tell me he’s the pick, I’m not going to argue with them at all. Karl Towns projects to be an excellent pro.

In addition to making an appropriate first overall pick, the Timberwolves stole Tyus Jones from the rest of the league by trading up to 24, exceeding the combined value of the 31^st and 36th overall picks by a raw score of 1.5. The model consensus would’ve taken Tyus Jones with the 9^th overall pick, so this is fantastic value.

Great job, Timberwolves. A.

Chicago Bulls

Pure Score: 1.942

Pure Grade: A-

Adjusted Grade: A-

Bobby Portis fell right into the Bulls lap at 22. Portis was expected to go anywhere from 7 to 19, and in my own model projects to be an NBA player on the level of Patrick Patterson. Portis has excellent size and decent mobility for an NBA 4, and put up strong numbers in his two years in the SEC. Exceptional value here.

Denver Nuggets

Pure Score: 1.153

Pure Grade: B+

Adjusted Grade: A-

Emmanuel Mudiay fell right into their laps. Lolz. After fleecing the Bulls on draft night last year, the Nuggets were able to take what may end up being the 4^th best player of the 2015 NBA draft with the 7^th overall pick. Mudiay should be a natural replacement for Lawson’s minutes when the Nuggets and Kings eventually decide they want to swap rosters.

Toronto Raptors

Pure Score: 0.468

Pure Grade: B

Adjusted Grade: A-

My numbers probably overvalue Greivis Vasquez, but don’t get it twisted, the Raptors had a phenomenal draft. The Raptors shed Vasquez’s 6.6 million dollar salary, AND got a future first round pick? When Masai calls, you hang up.

In addition, the Raptors picked up stat model darling Delon Wright from Utah, an excellent value pick at #20 in the draft. It was totally expected that he would be slept on by the league.

Miami Heat

Pure Score: 0.526

Pure Grade: B

Adjusted Grade: A-

Justice Winslow fell right into the Heat’s lap at #10. Excellent value here, and should be a strong NBA player, even if the concerns about “limited upside” and “fake three point shooting” become true. Many individual models considered him a top 3 prospect.

Although Josh Richardson isn’t as good as the expected value of his draft slot as defined by this exercise, he was #40 on the Draft Express Mock Draft, and has a few traits that may lead to success in the NBA, like his senior season steal and block rate of 3.0 per 40 minutes pace adjusted.

Los Angeles Lakers

Pure Score: -2.694

Pure Grade: D

Adjusted Grade: B+

Given that D’Angelo Russell (I like calling him Jello) is a stat model stud, you may have expected this number to be higher. Unfortunately, given the consensus’ expectation of return in this year’s 2^nd overall pick, drafting D’Angelo Russell barely moved their net score up. The pure score you see is a reflection of how objectively poorly the Lakers drafted elsewhere, taking Larry Nance at 27 and Anthony Brown at 32.

Having said that, Jello was not only the projected best player by my model, but by the draft consensus as well, narrowly edging out Karl Towns. Layne Vashro’s model also projects him as having a 70% chance of being a star, by far the highest ranking of any such player in this draft under that model. And the Lakers rightfully took him, as opposed to taking Jahlil Okafor, whom many fans and analysts had long assumed they would take. Intentionally or not, they also gave Sam Hinkie some of his soul-read medicine, and forced Hinkie into drafting his third true center in as many years, breaking Hinkie’s streak of not getting the stat-projected best overall player in the draft.

Given how much more important getting the 2^nd pick right is than getting the 27^th and 32^nd pick right, I’m bumping the Lakers all the way up to B+.

Oklahoma City Thunder

Pure Score: 0.739

Pure Grade: B

Adjusted Grade: B

The Thunder were linked to Cameron Payne since day one, and he is exactly who the models would have taken at pick 14. Although there were higher ranked PG prospects on the board, like Tyus Jones and Delon Wright, there is an intelligent argument to be made that Cameron Payne has more raw upside. I would’ve taken Tyus Jones, but this is fine.

Dakari Johnson is excellent second round value. Like most second rounders, it’s unlikely he’ll ever amount to much, but gambling on seven footers from Kentucky is far from poor form.

Los Angeles Clippers

Pure Score: 0.451

Pure Grade: B

Adjusted Grade: B

The Pelicans had one pick in the 2015 NBA draft, but they sold it to the Clippers, who made an excellent value choice in Branden Dawson at 56. The model consensus would’ve taken him as early as 47.

Houston Rockets

Pure Score: 0.301

Pure Grade: B-

Adjusted Grade: B

I tweeted it out earlier, but kudos to Daryl Morey for possibly being the only GM to be able to fake out the Great Woj. Seconds before announcing the Houston Rockets had taken Sam Dekker, Woj had said that the Rockets were to draft Tyus Jones, who was, at the time, the best available player left in the draft by the consensus. Instead, the Rockets took Sam Dekker, ever so slightly undercutting the expectations of the 18^th pick, but made a great value pick with the 32^nd overall in Montrezl Harrell, to come out on top with a positive return on assets.

Phoenix Suns

Pure Score: 0.230

Pure Grade: B-

Adjusted Grade: B-

The Suns fleeced the Grizzlies for Jon Leuer, but that doesn’t really move the needle much. Devin Booker is the type of player who gets into the lottery because he has a reputation for doing one skill (shooting) well, and physically looks a lot like a player who already succeeds in that role (Klay Thompson). Models would’ve preferred taking him later, at 24, not 13.

Charlotte Hornets

Pure Score: 0.183

Pure Grade: B-

Adjusted Grade: B-

Frank Kaminsky is by no means a bad pick at 9, and the Hornets appear to draft a lot of players like him. But Justise Winslow was just sitting there, and was probably the best player available. We highly advocate taking the best player available here at Nylon. Kaminsky is perhaps a better “fit” for the Hornets, but you can make an educated argument that he was also the best available.

I like the 39^th overall for two second rounder swap too, especially if one of them belongs to Brooklyn. That minor deal could end up being a fleece.

Indiana Pacers

Pure Score: -0.213

Pure Grade: C+

Adjusted Grade: B-

Myles Turner is a good pick for the Pacers at 11. He’s a bit of a beast, and actually has a decent mid-range jumper, which may be helpful to replace the mid-range game of David West, who is rumored to be leaving Indiana for New York.

Philadelphia 76ers

Pure Score: -0.393

Pure Grade: C+

Adjusted Grade: B-

Not sure Sam Hinkie had the best night. Despite having a better than league average return on his assets, Hinkie was forced into taking Jahlil Okafor, the best player available, and adding to his stockpile of skilled seven footers. Who needs smallball when you can tallball?

Immediately after taking Okafor, Hinkie said, true to form, that the 76ers had no intention of trading Okafor, when Hinkie was probably holding a phone designated solely for fielding Okafor/Embiid offers.

Despite technically losing to the Knicks in a draft night trade of Guillermo Hernangomez, Hinkie made a smart deal by effectively turning an early second rounder into two second rounders. Given that one of these picks belonged to the Knicks, it should be worth more than the “average second rounder”, so it’s safe to argue that Hinkie won, despite Hernangomez’s projected value of slightly above the exercise-defined value of two second round picks.

I feel bad for Hinkie. He was on a two year streak of grabbing the projected best available player in the draft despite not having the #1 overall pick. D’Angelo Russell was his guy, and it only served to hurt more that he was the PG the 76ers were looking for.

Washington Wizards

Pure Score: -0.421

Pure Grade: C+

Adjusted Grade: B-

The Wizards paid a relatively fair price to move up and take Kelly Oubre (dem shoes doe), who they took at a fair draft slot value at #15. Kelly Oubre could be really fun on the Wizards, especially if they roll out small ball lineups with Wall, Beal, Oubre, and Porter at 1-4.

New York Knicks

Pure Score: -1.759

Pure Grade: C-

Adjusted Grade: B-

Oh boy the Knicks. Really didn’t know what I was goink to give them here.

The Knicks lose a lot of raw score from drafting Kristaps Porzingis. Stat models don’t hate him, and the great Kevin Pelton’s model sees him as the #2 prospect, but overall the stats had Kristaps as a #11 talent.  Emmanuel Mudiay was right there.

Although it’s a lazy comparison, it’s actually correct. Kristaps is extremely close to Andrea Bargnani statistically. See this table below.

So that comparison has to be concerning. Still, anyone who watches this video of his 1v1 workout in Vegas and doesn’t have their basketball imagination run wild, doesn’t like basketball.  He COULD be the type of player you custom build in 2K or NBA Live. So I hesitate to call this a bad pick. Let’s wait and see. Kudos to Phil for going for it. I wouldn’t have taken him, and if I had somehow decided I especially needed to take him, I wouldn’t have had the stones to take him at #4.

The biggest reason I bumped their letter grade is that the Knicks somehow convinced someone from the league to not only take Tim Hardaway Jr. and his -3.82 RPM off their hands, but also offer a first round pick in exchange. What a steal. Jerian Grant isn’t an amazing pick, and he certainly isn’t my favorite subjectively, be he projects to be almost twice the player Tim Hardaway Jr. has shown to be so far.

The Knicks also traded for Guillermo Hernangomez, a good pick, and Kristaps’s teammate.  They might have given the 76ers too much, but it’s easily worth it if it makes Kristaps more comfortable and allows him to become what he’s capable of.

Detroit Pistons

Pure Score: -1.002

Pure Grade: C

Adjusted Grade: C+

Stanley Johnson is a good value at #8, but I would much sooner have taken Justise Winslow. The Pistons then reached for their second round pick on Darrun Hillard, a good college basketball player, but not a good value in the eyes of the model consensus, and not the type of low risk/reward player I advocate drafting in the second round.

Portland Trail Blazers

Pure Score: -5.299

Pure Grade: F

Adjusted Grade: C+

It’s hard to objectively evaluate the current player part of this deal for the Blazers, because Steve Blake and Mason Plumlee are at two very different points in their careers. The reason the Blazers score so low objectively is because my RAPMWS blend actually judges Steve Blake as the better player (since he has had the better career thus far), but I think we’d all prefer to have the young Mason Plumlee over him, thus the massive bump in letter grade.

The Blazers drafted Pat Connaughton at 41, which is an okay pick subjectively, given his performance at the combine. Most models pegged him as an undrafted talent, however.

The Blazers are preparing for the worst out of the LaMarcus Aldridge situation. Good on them.

Dallas Mavericks

Pure Score: -2.494

Pure Grade: D+

Adjusted Grade: C

Justin Anderson is somewhat of a reach at 21, especially with Tyus Jones, Kevon Looney, Bobby Portis, and others all on the board. Still, Anderson was an absolute monster at the combine, and I won’t hate too long on a bet on athleticism in the NBA. Satnam Singh (the Greg Oden of 2015 NBA Draft faces) is pretty much a projected 0, but he’s exactly the type of player I think teams should spin to win on in the late second round.

Sacramento Kings

Pure Score: -2.531

Pure Grade: D+

Adjusted Grade: C-

Willie Trill Cauley-Stein, first of his name, (probably) was taken 6^th overall by the chaos that is the Sacramento Kings. Subjectively, I am a huge fan of WTCS, and am in on the idea that he has the potential to be an elite defender. Unfortunately, models see him more as a late lottery talent, and Emmanuel Mudiay was just sitting there. Not sure Willie will get the player development he needs with the Kings either. The Kings are also highly unlikely to get appropriate return on DeMarcus Cousins, who, by the way, is one of the best players in the league.

Brooklyn Nets

Pure Score: 1.056

Pure Grade: B+

Adjusted Grade: C-

As I mentioned before, it’s hard to objectively evaluate the Nets deal, as Steve Blake and Mason Plumlee are at very different points in their careers. But I think it’s pretty clear the Nets gave up the best player in the trade.

In return, they were able to draft Rondae Hollis-Jefferson, pegged by the models’ as having solid value at his draft slot. If he ever develops a jump shot, he might find his way into being a better than expected NBA player. In the modern NBA, you need to shoot better than 20.7% from 3.

Nets get dropped here for losing Mason Plumlee, but also for getting fleeced by the Hornets for two second round draft picks. I’m not sure exactly which picks those are yet, but if Brooklyn is as bad as they can be in the next few years, one of those picks may be the equivalent of a late first rounder for the Hornets.

Cleveland Cavaliers

Pure Score: -2.030

Pure Grade: D+

Adjusted Grade D+

In theory, the Cavaliers won their deal with the Minnesota Timberwolves. The expected value of this year’s 31^st and 36^th picks exceeds that of the 24^th overall. However, they used those picks to draft Cedi Osman and Rakeem Christmas, both projected to produce lower than that draft spot. And with players like Tyus Jones (who the original 24th pick was used for), and Kevon Looney on the board, the move becomes even less defensible.

Orlando Magic

Pure Score: -3.358

Pure Grade: D-

Adjusted Grade: D+

Consider me the most down of all on Mario Hezonja. My model really disliked him, and I didn’t get the hype from a subjective level. The model consensus had him as a late lottery pick, and no individual model had him above #12 overall. Emmanuel Mudiay and Justise Winslow were on the board.

New Orleans Pelicans

Pure Score: -1.11

Pure Grade: C

Adjusted Grade: D

Sold their pick for cash (selling picks is bad, mmmkay) in a historically deep draft. D.

Milwaukee Bucks

Pure Score: -1.358

Pure Grade: C

Adjusted Grade: D

The Bucks made a huge reach for Rashad Vaughn at 17. That’s bad. They dumped a future first rounder and the 46^th for the right to pay Greivis Vasquez 6.6 million a year. That might be worse. Not sure what they were thinking here, but Greivis Vasquez and Rashad Vaughn is not the haul you want to receive for your 2015 #17 overall, your 2017 first rounder, and your 2015 #46 overall pick.

Atlanta Hawks

Pure Score: -2.878

Pure Grade: D

Adjusted Grade D

In theory, the Hawks trading the 15 for 19 and two second rounders is a good trade. But what they really ended up trading for is Tim Hardaway Jr. and two second rounders. Maybe Tim Hardaway Jr. is notably different in Atlanta, but this past year he had an RPM of -3.82 and his career so far has been worth about half of the projected value of the 2015 #15 overall pick. Taking no-names with the second round pick didn’t help their score either. Would have rather them had taken Kelly Oubre (or others) and just have been done with it.

San Antonio Spurs

Pure Score: -3.083

Pure Grade: D

Adjusted Grade: D

Who am I to decry the picks of the Spurs? Still, with Kevon Looney and R.J. Hunter on the board, as well as more highly ranked international prospects, I’m not sure Nikola Milutinov was the player to waste a first rounder on. But the Buford knows more.

Utah Jazz

Pure Score: -3.784

Pure Grade: F

Adjusted Grade: F

Trey Lyles is a big time reach at #12, especially with better versions of him in Bobby Portis and Kevon Looney sitting there on the board. Not sure what the Jazz see in him over those two players, and why they couldn’t trade down to take him later. Maybe they tried.

Olivier Hanlan doesn’t project to be much of an NBA player, and worth about half of the expected value of the pick he was taken with. He’s already 22 and looks like the kind of guy who won’t be athletic enough for the association.

Boston Celtics

Pure Score: -3.859

Pure Grade: F

Adjusted Grade: F

A few months ago I told someone to check out Danny Ainge’s basketball-reference page as an example of how to manage. This is part of what makes this draft so perplexing. The Celtics made a huge reach for Terry Rozier at 16, despite having the 28^th overall pick, a place where Terry Rozier was likely to still be available. Given that they had four draft selections, their ability to trade was also higher than that of other teams. R.J. Hunter was a good pick and Jordan Mickey is defensible, but Terry Rozier really weighs everything down, contributing -2.44 in raw score. Marcus Thornton, sworn enemy of the string match, doesn’t project to the best use of the 45h overall pick either.

Memphis Grizzlies

Pure Score: -4.214

Pure Grade: F

Adjusted Grade: F

Even though Jon Leuer is a net negative by most statistical standards, he is still much better than the expected value of the 44^th overall pick, and much much better than the expected value of what became of that pick, Andrew Harrison. The Grizzlies also took Jarell Martin at 25. They had apparently promised to take him there a long time ago. Martin is projected to be a mid-second round talent, and given the names on the board, the Grizzlies left a ton of expected value on the table.