This is the first application of my draft model to college players expected to enter the 2014 NBA draft. I plan on updating these numbers weekly, but won't do another write-up until we are considerably deeper into the season. If you want to follow the regular updates I will be posting each Tuesday on my twitter (VJL110@VJL_bball) and in the comments here at Canis Hoopus.
For those of you who followed along with my draft posts last season you know where these numbers are coming from and what they mean. For anyone else I can direct you here. In short, these are projections based on pace-adjusted box-score numbers, height, weight, strength of schedule, and age using parameters fit from 30 years of players moving between the NCAA and NBA. I am currently using only the win-share models from last season because they do not rely on any combine measures. This is exactly the same version used just before the 2013 draft. I have some ideas for improving my models, but I won't have time to implement those changes until closer to the draft.
You are probably thinking it is silly to start comparing draft prospects before any of the teams have even settled into their conference schedules. For the most part you are right. Even a complete NCAA season gives us an unfortunately small sample of what players have to offer. However, there is some value in running my model early in the season. First, in spite of the error inherent in samples this small, I am consistently surprised how stable most basketball statistics are over a season. Shooting (especially from three) is the only major exception. Otherwise we get a surprisingly accurate picture of a player's skillset even after just eight or nine games. Additionally, ESPN et al. do a terrible job of identifying the good prospects until right before the draft. This makes it hard for casual college fans to know which players they should attend to outside of guys like Wiggins and Parker. I want this to act as a guide for which prospects to key on throughout the season. Start following the horse-race now while you can still get the excitement of watching it unfold.
If you want some context for interpreting the following numbers, I have historical projections for players dating back to the early 80s in a google document here. You can see those ratings sorted by position by following the tabs on the bottom of the document.
Here are the first 2014 numbers including, expected wins peak (EWP), likelihoods of attaining different tiers of success, and the top-3 statistical comparisons:
Both Jordan Adams and Kyle Anderson looked like top-10 picks last season according to my models, but stayed back due to lack of interest from the league that drafted their vastly inferior co-wing in the lottery. The season is young and UCLA has not faced very challenging competition, but people should really be paying attention to these guys. The box-scores Adams and Anderson have posted across their first nine games are absolutely phenomenal.
Adams is a young sophomore scoring nearly 30 points every 40 minutes at a ridiculously efficient rate, moving the ball to open teammates, and taking it from opponents better than anyone in the NCAA. The extreme levels of his current performance will almost certainly drop off, but the general pattern is consistent with last season so you cannot simply ignore it. Adams has the build of a pro and a complete shooting-guard game in an era lacking stars at that position. To get a sense of his skillset, here is a pretty average game of his against Nevada. Note that he gets his offense hitting spot-up 3s and attacking the rim off the dribble. These are highly translatable offensive skills. The Draft Express scouting report notes some concerns about Adams' effort on defense, which is noteworthy since the boxscore is blind to most aspects of defense. However, I can't imagine his defensive impact is bad enough that he should be considered at the top of the draft. Adams is currently pegged as a mid second-rounder at best by most draft sites. I assume if he keeps playing like this that will change… right?
To put Adam’s current rating in perspective, here are the top EWP scores for shooting-guard prospects historically:
Yup. Obviously I don't think Jordan Adams is going to keep looking like a better version of Ray Allen, but he needs to completely fall on his face the rest of the season to regress towards the non-prospect most scouts think he is.
Kyle Anderson is a weird player. As a freshman he played the role of defensive ace and part time point-forward. This season UCLA has given 6’8" Anderson the primary distributor role and he has responded by doubling his already impressive assist rate while also improving in every other statistical category. It is difficult to think of an appropriate comparison for Anderson’s role at his size. Reggie Theus and Jalen Rose were also tall distributors, but neither player collected rebounds and assists like Anderson. The only other player who jumps to mind is on my "taboo comparison" list… Magic Johnson. I want to reiterate the small sample disclaimer here, but Magic averaged about 19 points, 9 assists, and 8 rebounds every 40 minutes as a sophomore, while Kyle Anderson is averaging 17, 9, and 11. As with Adams, these numbers are very likely to decline. Still… early returns combined with Anderson’s freshman performance are good enough to comfortably label him a contender at the top of this draft class.
If you get nothing else from this write-up, get this: Watch UCLA vs. Duke on Thursday.
The other impressive veterans:
Marcus Smart should be in the NBA. If he had stayed in last season’s draft I would have argued for him at the #1 pick. This season he has come out and removed any remaining doubts. If the NBA decided to hold the draft tomorrow I would also advocate taking Smart #1. I have complete confidence that Marcus Smart will be a badass professional basketball player. Please enter the draft this time!
Willie Cauley-Stein is also proving he can play at the next level. He stepped up after Nerlens Noel went down last season and looked like a top prospect in 2013 and has only improved as a sophomore. Like Noel, Cauley-Stein puts up a very impressive combination of blocks and steals, something that consistently identifies successful pro centers. It helps that Willie also has something close to the ideal center frame to complement his statistical output. On the negative side, Willie’s rebounding has not been as impressive as it could be given his size and he is very limited offensively. These flaws hurt his odds of becoming a star player, but he still looks like a very safe bet to eventually get significant NBA minutes.
As with the previous two players, Jarnell Stokes has a track record of looking like a future NBA player. The fact that the players who did well in my models last season are right back at the top of the list speaks to the surprising stability of basketball statistics I noted in the introduction. In spite of Stokes' top-10 rating in the "expected win production" model he isn't the most exciting prospect, which I think is captured in his player comparisons and "star %". Stokes is another burly power-forward who collects rebounds and scores inside with decent efficiency. His usage and assist rate have increased significantly from his first two seasons and hopefully that keeps up because he will need something beyond good rebounding to become more than a backup in the pros.
Keep your eyes on Warren, Dawson, Harrell, and Dinwiddie as well.
The ballyhooed freshmen:
Six freshman have posted an EWP above six which looks impressive compared to last year’s class in which Nerlens Noel was the only declaring freshman to do so. The "declaring" is important though as one of the biggest problems with last year’s draft was that some really good (though under-appreciated) prospects stayed in school. This class is certainly good, but it is yet to live up to the hyperbole. This is particularly true for Wiggins who has had a good but unspectacular start to his NCAA career. The "best prospect since LeBron" currently looks like a marginal top-10 pick destined to fill the off-ball secondary scorer role in the NBA. Always disappointing when the next Kevin Durant plays like the next Bryant Stith. Wiggins still has plenty of time to get his head around college basketball though, so we shouldn't count him out yet. The value of pre-college hype wears away quickly, but even I am not comfortable completely dismissing it after only eight games. Aaron Gordon is the other top recruit who looks pretty good but needs to do a lot more throughout the rest of the season to warrant top-5 status.
Julius Randle has had a much more interesting start to the season. He dominates the paint, scoring with good volume and efficiency, rebounding nearly everything on both ends, and moving the ball willingly (if at times sloppily). This is all great stuff, but Randle has also waved some serious red flags. He played more than 200 minutes of college ball before collecting his first and only steal. Given the strong relationship between collegiate ball-thieving and NBA success this is a very concerning hole in Randle’s game. He is not compensating with shot-blocking either, adding less than one block every 40 minutes. I worry Randle’s defense might be too bad to allow him to get off the bench in the NBA. Finally, his scoring seems extremely dependent on getting to the line. There is always a bit of the "adult playing against children" feel watching top prospects in college, but rarely is the physical contrast as stark as with Randle. He won’t be as a unique a bully in the pros and I imagine that is going to take a major toll on his scoring efficiency.
Jabari Parker is my favorite of this freshman group. He has scored with ease while posting a really balanced box-score. However, Parker is yet to establish a dominant complimentary skill to his scoring. He can rebound, he can pass, he can collect steals and blocks, but he does not do any of these at a star level and that keeps him from joining the historically elite tier of prospects. This will become especially problematic when his accuracy from three inevitably regresses to the mean.
Another freshman with production that appropriately places him in the "top-pick" discussion is Joel Embiid. While Parker needs to find a secondary skill to become special, Embiid simply needs to stop doing one really problematic thing. Embiid’s rebounding and shot-blocking are both excellent and for his position you could argue the same for his passing, while his scoring is solid. Unfortunately he is fouling at an almost unplayable rate (7.6 per 40 minutes). Fouls are probably the least troubling flaw to see in a young center prospect since many eventually figure it out, but there are plenty of otherwise talented 7’ers who failed to make it in the NBA because they can’t stay on the floor.
Zach LaVine was not nearly as hyped as the other top freshmen in this class (RSCI of 52), but he has fit in nicely as another great wing at UCLA. He isn’t really doing anything exciting though… just hitting shots efficiently and moving the ball around well enough to look like a legitimate 2-guard. I wager his rating is going to drop once his 3s inevitably stop falling at a 50% clip.
Finally, we have Noah Vonleh. With an RSCI of 10 Vonleh was a big recruit, but he was not viewed along with Wiggins/Randle/Parker/Gordon as the future of professional basketball. This shouldn't be a concern since relative to high school scouting, college ball gives us a much better sense of how a player will perform in the NBA. Vonleh is showing more promise than any of his peers so far and is somebody we may want to start getting a little excited about. He is doing all of the good things we see in Julius Randle without any of the glaring defensive concerns. Vonleh has excelled in areas that tend to hold stable over time, so I will be surprised if his production drops off much throughout the season. On paper, he looks remarkably similar to a young Kenneth Faried, but hopefully he can flash some skills that put him a tier above that comparison as we move into the conference schedule.