The Diff: The stats behind drafting Nerlens Noel at No. 1

The Diff

The Diff is your weekly Wednesday WFNY look into the amazing world of sports statistics. For a complete log of articles, click this link. Last week, The Diff covered the historical finishes for MLB teams that started 26-18. This week, it’s NBA Draft talk time.

The Diff

Since the Cavaliers won the lottery last week, Cleveland sports fans have been enamored by the idea of Nerlens Noel. Whether you love him or hate him, the 19-year-old University of Kentucky center already has been the subject of dozens of Cleveland-based analyses and reports. Is he the best prospect in this draft? Will he fit in the NBA? Will he actually make an impact in 2012-13? How good can he be? After perusing through the usual and not-so-usual stats, I’m here today to give you the statistical know-how about why Noel should be the Cavs’ no-doubt pick at No. 1.


Age-based analysis

Let’s start with a little chart. This shows the top 11 of my aggregate NBA big board, a chart I’ve updated here at WFNY several times previously. This top-11 separates itself pretty well from the rest of the group1. Especially, I want you to pay close attention to the ages listed below:



Note that Nerlens Noel is 10 months younger than any of these consensus top-11 prospects2. Then, after just Alex Len and Otto Porter, he’s 13-14 months younger than Ben McLemore and Anthony Bennett. But the point still remains: Since Noel previously re-classified to the high school class of 2012, from his initial standing of 2013, he’s a full year younger than his peers. Now why does this matter?

Let’s pass that along to Kevin Pelton. One of the lead innovators of the basketball statistics movement, Pelton wrote an ESPN Insider article following the discovery of Shabazz Muhammad’s actual age back in late March. Here’s the gist of his point about why age matters:

… history shows that age is key to understanding how college players end up developing in the NBA. In fact, age is one of just two factors that make up my rankings of college players. The other is their performance during the previous season, translated to its NBA equivalent based on the past performance of rookies and adjusted for strength of schedule.

The rookie translations generally do a good job of separating newcomers who are ready to contribute right away from those who are too raw to contribute. But in terms of projecting value beyond the first season, as measured by wins above replacement (WARP) in the players’ first five years in the NBA, age proves almost equally important. Based on regression, age makes up about 40 percent of the WARP projection for rookies, with translated performance accounting for the other 60 percent.

The Cavaliers heavily value proprietary advanced basketball stats, which are certain to be similar to Pelton’s WARP model. Noel projects very favorably not just because of his elite collegiate production — more on that in a moment — but also because he was doing so as an 18-year-old before being injured. That’s why his potential is clearly the highest of any of these players and he has by far the highest WARP projection, too.


Usage-based PER

There have been multiple articles, such as this one over at Hoopsworld, that claim that Nerlens Noel’s 27.3 PER and Otto Porter’s 27.3 PER last season in college are quite similar. Controversially and unconventially, I beg to differ.

Back on February 6 in The Diff, I wrote a very long-winded argument about how usage rates and field goal attempts per 36 minutes positively affect PER. In essence, it’s because PER is only a rate statistic in as far as it can throw; the more shooting attempts a player has, or the more involved he is in the offense, the higher that stat will go. Not exactly fair, and that’s why players such as Bruce Bowen (8.2 career PER) have been criminally underrated by the supposed catch-all metric3.

That’s what brings us to Noel again. During his 765 minutes with Kentucky, he posted a 17.4% usage rate4. That’s absurdly low for a top-end collegiate prospect. In fact, let’s take a look at the usage rates of these top-11 prospects here in 2013 and do a little trick: What have other collegiate players since 2009 within 0.1 of said player’s usage rate then posted in the PER category on average? Let’s look:

RkLastPERFGA/36UsageUsage-Avg. PERRank


Note the fact that I’m using C.J. McCollum’s 2011-12 statistics and his other peculiarities5. Also, again, this only includes player-seasons with 750+ minutes.

Thus, now we can see how proportionally to his usage rate, Noel’s 27.3 PER is far more impressive. On average by multiplying PER per minute, folks with 17.2-17.4 usage rates averaged approximately a 14.6 PER. While looking at Porter’s grouping of players with 24.0-24.2 usage rates, the average PER was much higher at 18.5.

Now, of course, this includes cross-position players. And as Andrew has noted before, some guards have lower efficiency statistics than forwards, especially in the beginning of their development. Overall though, placing these players in the context of their usage helps to gain a larger understanding of the strength of their PER. They’re not all created equal.

For Nerlens Noel, that again means that his numbers are even more impressive than they’d show on the surface. His defensive numbers were outstanding, his shooting efficiency was still above average6 and he did an exceptional job per his overall usage opportunities.


Elite defensive ability

One of the beautiful things about the NBA: the huge number of other amazing websites out there are that talk intelligently about the league. One of those is the Detroit Pistons website Detroit Bad Boys. And in a roundtable before the lottery, one of their contributors wrote in a similar ilk as my thoughts about Nerlens Noel’s elite-beyond-elite defensive potential:

What would Nerlens Noel bring to the Pistons?

Shinons*: How would you like to have the most exciting team in the league? One that leads the league annually in blocks, steals, and rebounds? One that features a potentially historically good defense? Even better than the Bad Boys or Going to Work eras? That’s how highly I think of Noel.

This guy put up a combined 6.5 blocks and steals. He is the best defensive playmaking prospect I’m aware of. Davis last year had 6.1. Before these two, the next best that I can think of was Jarvis Varnado, who put up 5.4 his senior year. This number, 6.5, it’s unbelievable.

This fellow is correct: Nerlens Noel’s 6.5 combined steals and blocks per game last season was elite. As in, once-every-decade elite. As in, All-NBA Defensive Team elite. As in, along with new (old) head coach Mike Brown, could help the Cavaliers buck the trend of ranking 27th, 26th and 29th in defensive efficiency in the last three years, respectively.

In fact, according to, Noel’s 6.5 block-and-steals per game mark was the third-best total dating back to 1999-2000. Among all of the many, many college basketball players since then, only Northeastern’s Shawn James (7.3) in 2005-06 and Alabama A&M’s Mickell Gladness (6.8) in 2006-07 surpassed Noel’s mark. Over this span, the UK center is also one of only 17 players to average at least 2 steals and 2 blocks per game — another very rare feat at any level.

Those statistics above show the historical rarity of a player with Noel’s elite athleticism, defensive awareness and shot-blocking ability. He’s more than just a one-trick pony. On a more efficiency basis, since started tracking this kind of data in 2009, Noel became the first ever collegiate player to play 600 minutes in a season and post a 9.3%+ block percentage (his was 13.2%) and a 3.2%+ steal percentage (his was 3.9%).

In the vein of possible NBA comparisons, let’s take a look at some of the elite defensive big men prospects of the last several decades and what they produced in the steals and blocks categories per 36 minutes7:



Numbers that rival all but practically Olajuwan and Robinson at the college level8? A profile that mirrors a bigger, more block-prone Outlaw? Nearly equal on blocks and far better on steals than last year’s consensus No. 1 pick Davis? Clearly, the majority opinion is that Noel is a good defensive player. These numbers, and the ones above as well, show that he was an elite defensive star at Kentucky with the chance to be one of the best in the NBA.


Overall outlook

Thus far, I’ve shared how the stats back up the narrative that Nerlens Noel has the most potential, had the most impressive usage-based PER and has by far the most elite defensive star power of the top-11 prospects in this draft class. Notably, however, I haven’t touched on two of the main criticisms of Noel: his offense and his recovery from his ACL injury.

Back in March 2012, Kevin Pelton wrote about the historical recoveries of basketball players that suffered ACL tears. It was written just as players such as Derrick Rose and Iman Shumpert went down with the freak injury. Shumpert is back on the court, while Rose is not. But overall, Pelton’s conclusion was that younger NBA players have a much more likely chance of returning to full strength. This doesn’t even include cross-sport examples such as Adrian Peterson. In my mind, ACL tears are so common yet such freak occurrences, that usually won’t be a problem long-term once the initial rehab process is over.

On offense, Noel simply isn’t that skilled right now. For UK, he was 98-166 (59.0%) from the field and 55-104 (52.9%) from the free-throw line. He got better at picking his (limited) spots offensively as the year went on, shooting 66-102 (64.7%) in his final 17 games.

Noel’s offensive game right now consists of dunks, fastbreak opportunities and putbacks. He doesn’t really have much of a post game nor any other kind of go-to move. He is a bit better than your average near-7-footer at putting the ball on the court and making plays (per DraftExpress), but he turns the ball over quite a bit as well.

Over time, similar to the progression of Anderson Varejao and Tristan Thompson, however, I believe Noel’s offensive game can become much more than it is now. Those two Cavs are great examples. Here are shot charts for both players and how they developed over time:

Varejao’04-’08At Rim44573560.5%
Varejao’09-’13At Rim25076665.8%
Thompson’11-’12At Rim13824157.3%
Thompson’12-’13At Rim22436860.9%


Varejao’s time period of offensive evolution was longer than Thompson’s, but the point still remains: Big men can get much better offensively over time. Both of these players, especially Thompson, were ridiculed as they entered the league for their lack of offensive ability. Noel is very much the same9. If he can develop offensively at all in the same progression as these two Cavaliers over the next few years, he’ll be an All-Star soon. His defense is already worthy, but if he develops an offensive game, watch out.

In the end, I’ve long thought that Nerlens Noel is the deserving, no-brainer, don’t-think-twice-about-this No. 1 choice throughout the college basketball season. These numbers above back it up in a variety of ways, showing his elite potential and room for continued improvement. In my mind, he’s not the next coming of Greg Oden just because of one freak injury. He’s more likely to be a transcendent defensive talent, with room to grow offensively, but the ability to mix in seamlessly with Kyrie Irving and Dion Waiters to grow an exciting brand of Cleveland basketball once again.

  1. It’s fun to look back and compare these rankings over the past few months. Of course, former top prospects such as Marcus Smart returned to school, hence forcing others to move up. But Shabazz Muhammad has dropped considerably since the end of his tournament days, while Trey Burke (duh) and C.J. McCollum (next Damian Lillard) keep rising. My January primer also is a fun treat too. []
  2. Another Noel pet peeve: He’s been the consensus No. 1 player on this aggregate big board list through and through. At a couple different points, he was tied with Ben McLemore. But overall, the majority of rankings have had him No. 1 on their big board lists since the beginning of the season. That’s called a consensus, folks. []
  3. Recall: PER is build around the idea that 15 is league average. So no, Bowen wasn’t significantly below league average. He just never had the usage rates of traditional league-average performers. Many NBA stats struggle with enumerating defensive talent; this is just one of many. []
  4. Per Usage percentage is an estimate of the percentage of team plays used by a player while he was on the floor. Hence, in a sense, 20% could be considered average at any one moment in time. []
  5. Due to an injury, the Lehigh star only played 372 minutes this past season. Even still, because of his absurdly high usage rate last season too, I was forced to look at all players with usage rates between 33.1-34.1. His usage was even higher in his limited minutes this past year: 37.2. Again, insanely high. []
  6. Partly because Noel didn’t get many opportunities outside of slam-dunks and fastbreaks, but also kudos to him for knowing his weaknesses and not forcing the issue. []
  7. Obviously, I cherry-picked the list below and the years form which I picked. I tried to use the last two seasons for these players, if at all possible. Clearly, Noel’s sample size is the smallest, which is a major issue. But it’s all we have to go by for his comparable collegiate record. []
  8. For additional reference, here are some other current NBA defensive stars and their collegiate (S+B)/36 — Tim Duncan’s four years: 4.4; Roy Hibbert’s last two years: 3.9; Joakim Noah’s last two years: 4.6; Larry Sanders’ last year: 4.4. This obviously doesn’t include the international or high-school-to-the-pros defensive stars such as Kevin Garnett, Andrew Bynum, Serge Ibaka, Marc Gasol, Dwight Howard and Tyson Chandler. Nothing much I can do there. []
  9. College-level shot charts don’t exist en masse, so I can’t easily share what Noel’s numbers were like in college. I have made the offensive comparison to Thompson before though — one, two and three times, actually. Thompson had some pet moves and over-tried on them when they weren’t very good. Again, at least Noel knows his weakness and doesn’t force the issue with his shot too much. []
  • The_Real_Shamrock

    It was…………………….that damn good!

  • mgbode

    I had Perry Jones behind Zeller but not by too much. I also was extremely high on Jeremy Lamb. Both of them had the same issue as Tyler and Noel have though. None of their bodies were ready for the NBA. They all have to add strength in order to truly be able to compete fully and were projects.

    I wouldn’t give up on PJIII or Lamb yet either. They were put into a situation that they could basically take the year off and get their bodies ready as part of OKC. Makes it difficult for the first year, but, in the long run, it could very well help their careers.

  • Ben Frambaugh

    Yeah, I actually had Zeller behind PJIII. I didn’t want him at #1…but was actually OK with him at 4 (considering the options.) I also knew nothing about Tristan Thompson at the time.

  • The_Real_Shamrock

    Granger is and would be an intriguing player of course alot depends upon his health now. I like his teammate David West even more always have since his days at the X!

  • The_Real_Shamrock

    What what what what?

  • Ezzie Goldish

    Let’s try this going all out and see if anyone thinks it makes sense (please Jacob or someone reply!)…

    Trade #31 and #33 to Dallas for #13 and Shawn Marion.
    Trade #1, Anderson Varejao, and future SAC 1st and MEM 2nd rounders to Minnesota for Kevin Love, #9, #26.
    Trade #9 and #26 and perhaps future Cavs’ second rounders to Charlotte for #4. [Jordan wants more picks.]
    Trade #4, future MIA 1st rounder, and future ORL 2nd rounder to Orlando for #2 pick. [Orlando would still end up with the same player, so you may not even need this much.]
    Draft Otto Porter at #2. Draft a backup guard at #13 (picking randomly let’s use Carter-Williams). Draft a backup center at #19 (again let’s just pick Jeff Withey).

    Here’s your Cavs’ roster:

    PG Irving, Carter-Williams, Livingston
    SG Waiters, Ellington, Miles
    SF Porter, Gee
    PF Thompson, Marion
    C Love, Withey, Zeller

    Even if you have to throw in #19 into one of those trades you’re still in great shape, and you still would have:

    All future Cavs’ 1st round picks.
    Memphis’ future 1st rounder.
    Plenty of cap space.

  • JacobWFNY

    Ew. Xavier. (I’m a UD alumnus.)

    Speaking of Xavier, I can’t stand that Paul guy for the Reds.

  • JacobWFNY

    Agreed with lots of these points. Another underrated, if slightly related skill, that you didn’t explicitly mention: Passing. Noel’s got a bit of a tendency to drive to the basket off the dribble. He’s not that great at it. But when he gets a running start, he’s tough to guard. Then he can just dump it off. His assist rate was quite impressive, even if he still turned it over quite a bit too.

  • mgbode

    I think passing is a skill that he can develop, but does not currently possess. He does “look” to pass, but as you mention his passes often end up in the hands of the other team. It is certainly a learnable/teachable skill though and having the mindset to look for the pass is often the biggest obstacle.

  • dwhit110

    You’re not getting picks back from Minnesota with Kevin Love… I mean you’re not getting Love period, but any reasonable trade scenario includes all of the stuff you have us giving them and only Love coming back in return.

  • mgbode

    I love Xavier Paul. He hits into 1-2-3 DP’s when the bases are loaded and saves his HR’s for when the bases are empty and the game is out of reach.

  • Ezzie Goldish

    That’s what I’d thought, but a piece yesterday suggested that to get the #1 the Wolves would have to give the Cavs that. I didn’t really understand why, but I put the SAC/MEM/Andy pieces in just to make it more even.

  • mgbode

    Ok, I’ll play along:

    “Trade #31 and #33 to Dallas for #13 and Shawn Marion.”
    possible. depends how much the Mavs want to shed salary.

    “Trade #1, Anderson Varejao, and future SAC 1st and MEM 2nd rounders to Minnesota for Kevin Love, #9, #26.”
    not happening. we’d have to give up #1 and likely our 2015 #1 plus more assets (including Andy). maybe they give up #26 in the deal, but no way #9 is coming back to us.

    “Trade #9 and #19 and perhaps future Cavs’ second rounders to Charlotte for #4”
    I don’t think 9+19 would be enough anyway, but we won’t have it. #13+#19 might be able to get us into the 8-10 range.

    ok, so the idea is to get a starting SF, backup guard and backup big man. we have #13, 19 and let’s say 26 going by my “possible” choices above.

    13: Shabazz (wing/SF)
    19: Dieng (big – though he may go higher, I think he drops here with all the centers on the board this draft)
    26: Franklin (guard)

    Here are the issues.

    We have Thompson at center (Love is a ‘4’).
    We are ridiculously young (tough to compete w/o veterans)

  • Ezzie Goldish

    A friend said the Mavs don’t really need to shed Marion, which is fine. Happy to do the trade without him.

    Nice picks at 13/19/26.

    I don’t mind being that young and that talented. I think TT would move to the 5, which is fine.

  • The_Real_Shamrock

    Professor Xavier was a bad man!

  • Ezzie Goldish

    Though I do think Jordan would consider trading down with the right price. Can always try 13, 19, and other trade assets. I don’t think the #4 to #2 price will be a lot since ORL knows they’d still get Burke and screw a division rival.

  • The_Real_Shamrock

    Good golly Miss Molly if Grant did anything remotely like that I’d be floored. I mean I’d look like Mike Brown in the offensive huddle with 2 seconds left in the game and Cleveland down by 1.

  • The_Real_Shamrock

    With Brown out back as HC I doubt Varejao ever leaves.

  • mgbode

    I agree the trade from #4 to #2 would be feasible if Orlando wants Burke (and I think they do as well).

  • Ben Frambaugh

    Dallas had offered the 13 and Marion for the 19, 31 and 33 iirc.
    Kevin Love ain’t leaving.
    That leaves us with the 1 and 13 (plus Marion)
    With the 1, we can draft Noel. With the 13 we can shoot for ShaMu or Karasev (or you can go with another Big instead such as Adams)
    We resign Livingston (or another backup PG) as well as Miles, Ellington and Jones
    Kyrie – Livingston
    Waiters – Ellington – Miles
    Marion – ShaMu/Karsev – Miles
    TT – Jones – Noel
    Varejao – Zeller – Noel
    Kyrie – Livingston
    Waiters – Ellington
    Marion – Miles
    TT – Jones – Noel (til his health improves)
    Varejao – Zeller – Adams
    At the trade deadline, you still have Marion (and his expring) as well as Varejao (and his potential expiring) that you can flip for more future assets. You build around a team that has Kyrie – Waiters and a 3-big combo of Noel, Zeller and Adams. In 2014 you sign one of the SF’s (unless you have flipped Marion/Varejao for him already) and move forward. If you picked up assets, then you combine your 2014 and said assets to move up and get one in next years draft.

  • Ben Frambaugh

    The only problem I have with David West is how to fit him in our current lineup. I’m not sold on moving TT to the bench…and I don’t think a West/TT combo could play together on the court for long stretches of play.
    I agree with Graner’s health being a concern…but I think on a team with Dion and Kyrie he won’t have to carry near as much of the offensive burden…so he won’t have to go quite as hard on that end. At least, that’s my irrational hope.

  • dwhit110

    I saw the Sheridan piece that suggested that. What can I say, that dude’s smoking crack.

  • The_Real_Shamrock

    I was speaking more as a fan of West in general heck I hope the Celtics kick his tires his and Al Jefferson, bring big Al back home.

    But his fit with Cleveland would be tougher. I think Thompson could come off the bench and excel as the new sixth man allowing Varejao to be traded. Granger would be an easier fit though.

  • Jamille Raymond

    Marion was “OFFERED” No need to trade

  • CavsSuperFan23

    Kevin Love is wildly overrated. Stretch 4’s in general are wildly overrated by amateur fans. You don’t want your best or second best rebounder 20 feet from the basket on a regular basis. I’ll refer you to the Pacers, Grizzlies and Spurs. I wouldn’t trade Noel for Kevin Love at all when factoring in contracts.

  • Ben Frambaugh

    Ugh…can’t stand Boston, LOL (Mostly, it’s their irrational fans that I deal with at work though…)

  • Norco_10mg

    Will someone please tell me the relevance of comparing Noel’s blocks/steals per game with those of Shawn James and Mickell Gladness? They faired pretty well in the 2008 NBA Draft, didn’t they, followed by stellar pro careers? Yes, that was sarcasm. So, Jacob, what the heck is your point here? Totally irrelevant and meaningless.

  • Dominic Slater

    Nice job on composting an impressive argument for Noel, but for all of the statistical conjecture the only stat that matters was overlooked. 206LBS All the other compared player stats come with 20+ lbs when they were drafted (with exception of Outlaw 210 who was a 6’8″PF). It is interesting to note that when you break down the LBS adjusted NBA (S+B)/36 average (0.0172 per lb), Noel is set to average 1.1 STL and 2.5 BLK per 36. With those numbers Noel looks more and more like an Outlaw elk player. With this in mind, would you select #1 overall Bo Outlaw? BTW Outlaw was not selected in the 1993 draft.

  • Ben Frambaugh

    206lbs because his doctor asked him to rehab at that weight. He was listed at 212 earlier this week. He was also 4% body fat. This suggests that he’s in great shape for his height as he only has about 8-9lbs of fat stretched out over his 6’10+ frame.
    He was said to weigh 220+ during the season.
    What you are doing as cherry picking a stat (or in this case a measurement) and not giving it any kind of context at all. Not the best way to make a point when you are dealing with an intelligent crowd.

  • Ben Frambaugh

    The point is to find as many comparisons as possible…not just the good ones. This should give you an idea of a ceiling and a floor.
    Also, if you look in the comments below, I pointed out that two of the three guys mentioned had their stats inflated because they played at mid-major programs.

  • Ben Frambaugh

    Not to mention…if we do get Love, what do we do with TT? He can’t play Center for sure. And if injury concerns are a reason for not drafting Noel…then why would you trade for Kevin Love, who has struggled with injuries.

  • Ben Frambaugh

    Plus, I believe that the offer they presented was 19, 31, 33 for Marion and the 13. So it really won’t work for this scenario anyways.

  • mgbode

    we better be careful here. with this many comments someone is going to think it’s about a Browns QB.

  • J.D. McNugent

    Don’t make the same mistake Portland did! Taking a defensive minded, injury plagued center over a longgg crispy scorer! Smells like Oden-Durant all over again! Take Otto Porter!!!

  • Ben Frambaugh

    Except that Oden really wasn’t injury prone. He had the injury to his wrist prior to the season…played his lone college season…and went into the draft. He started having injuries to his knees after he was drafted.

    And Durant was waaaaaaay more skilled than Porter. He was also significantly longer (with his 7’5 wingspan) coming out of college. He averaged 25.8 points and 11.1 rebounds per game, shot over 40% on his 3’s, 47% on FG and 81% on his FT.

    If you could combine two seasons that Otto Porter played, he had 25.9 points per game and 14.3 rebounds. Porter is NOTHING like Durant. Nowhere near Durant’s offensive acument.