August 1, 2014

The Diff: 2013 NBA Mock Draft Database

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’s article covered the makings of the 2014 Cleveland Indians rotation. This week, it’s back to the draft.

The Diff

This is not my first rodeo in talking about the 2013 NBA Draft class. For reference, you can look back at my WFNY articles in January, early March, late March, April and then mid-June. This has been an exhaustive process, but one that can always be handled with a few more stats. That’s the subject of today’s edition of The Diff: I’ve looked at 40 recent mock drafts from around the Internet and I’m here to share the results with what they mean for the Cleveland Cavaliers’ potential plans tomorrow. This activity will help to determine projected draft tiers and availability at different slots in the draft’s first round.

Many of you should still remember the NFL mock draft database that I wrote on the Wednesday before the draft in late April. For that experiment, I looked at recent 60 NFL mock drafts from around the Internet and shared the results. Obviously, the results were inconsistent: only six mock drafts had Kansas City picking Eric Fisher at No. 1, while only three mocks had Barkevious Mingo going No. 6 to the Browns. But the analysis still was helpful: It provided a context for where the media thought teams were leaning, as well as potential availability once moving a few picks into the draft.

So let’s get started in looking at a 40-mock draft sample from the past two weeks and how this can possibly help the Cavaliers prepare for Thursday’s main event.

 

The database

nba mock draft database

I’m presenting a look at my NBA mock draft database in two ways. First, you can take a look at the JPEG image to the right. It’s a pretty large file, so you should be able to see everything clearly with no problems.

Also, here’s the link to a locked Google document so you can view the entire spreadsheet on your own time. The document has information on the dates and names of all 40 mock drafts, so you can come to your own conclusion about their accuracy.

Key:
AVG — The average of the player’s slotting in the mock drafts where they were taken.
#’s — The number of mock drafts, out of 40, where this player was a first-round pick.
No. 2, No. 8, etc., (availability) — The percentage of availability for that player at a given slot.1
CAVS — The number of times, out of 80 mock draft selections by the Cavs at No. 1 and No. 19, that the player is selected by Cleveland.

Obviously, when looking at anything from the Internet, this all deserves a grain of salt. You can hardly trust any of the rumors these days because their all spin in some form or another. By aggregating 40 recent mock drafts, hopefully some of this margin of error is reduced, but then there’s also the very high possibility that most of these mocks are just guys like me sitting behind a computer and guessing randomly. That happens, too.

In the end, this experiment just provides the context for what most mocks are saying right now. And that’s just another reason why I’m such a huge fan of aggregating2.

 

The predictions

All right, enough technical talk. Now, let’s get into what these 40 mock drafts from the past two weeks actually say. And the easiest way to do that is to just go line-by-line.

No. 1 pick: Nerlens Noel (21), Alex Len (13), Otto Porter (4), Ben McLemore (2)

Analysis: So yes, despite all the rumors to the contrary, Nerlens Noel is still the proverbial favorite to be the No. 1 selection for the Cavs on Thursday. Noel is the top pick in 52.5% of these 40 mocks. The number is relatively unchanged (54.2%) among the 24 mock drafts over just the last week.

No. 2 pick: Ben McLemore (19), Victor Oladipo (12), Nerlens Noel (8), Trey Burke (1)

Analysis: Recent rumors indicate Noel is the favorite for Orlando if he slips. But McLemore was still picked more often when available (50.0% to 42.1%). Oladipo is gaining momentum, too.

No. 3 pick: Otto Porter (29), Anthony Bennett (5), Nerlens Noel (3), Alex Len (3)

Analysis: Mock drafts have Porter as the overwhelming favorite for the Wizards, with him being selected 80% of the time when available.

No. 4 pick: Anthony Bennett (14), Alex Len (9), Nerlens Noel (7), Ben McLemore (6), Victor Oladipo (2), Otto Porter (12), Cody Zeller (1)

Analysis: True uncertainty is prevalent with Charlotte at No. 4. Noel was picked 87.5% of the time when available, with the trio of Bennett, Len and McLemore all in the 30-40% range.

No. 5 pick: Victor Oladipo (22), Ben McLemore (11), Anthony Bennett (5), Nerlens Noel (1), Otto Porter (1)

Analysis: Consensus returns with Phoenix’s pick at No. 5. Both Oladipo and McLemore were selected 84.6% of the time when available, so it’s likely they’ll go with one of those two.

No. 6 pick: Trey Burke (19), Alex Len (13), Anthony Bennett (4), Otto Porter (3), Victor Oladipo (1)

Analysis: While Burke was the most common selection for New Orleans at No. 6, the mocks had them selecting Len 86.7% of the time when available.

 

The top draft tier

Sifting through the projections, there was a clear indication of a top draft tier to share with the public. Obviously, many knew of the possible selections for the Cavs at No. 1 — but who are the overall top players picked, per these 40 mocks?

Take a look again at the player averages and, especially, at the results in the No. 8 column. I included availability at that pick for the purpose of explaining the clear top seven that appears in these mocks.

Astoundingly, despite the varied iterations of the early draft order and all the back-and-forth speculation, there was the same seven players selected in the top seven picks in 60% of these 40 mocks. That’s an incredibly high number, especially in comparison to the lack of clear consensus after the first three picks in the NFL Draft database.

Here are those top seven players: Tier #1 — Noel; Tier #2 — Porter, McLemore, Len, Oladipo, Bennett; Tier #3 — Burke.

Although together they’re clearly the top tier, there’s as large of a difference between No. 6 Bennett and No. 7 Burke as there is between No. 2 Porter and Bennett. A rather large difference also exists between No. 1 Noel and Porter. Thus, there are three listed tiers within this top seven.

Availability at No. 13 v. No. 19

The most interesting component of the database JPEG image is the section related to availability at different places in the draft. I shared the No. 2 slot to show the Cavs’ probabilities with the top pick and the No. 8 slot to show the clear top seven tier in the draft.

Now, it’s time to get a bit more complex with the middle of the first round possibilities for GM Chris Grant. Ever since the lottery, there have been rumors related to the Dallas Mavericks possibily dumping the No. 13 pick. Thus, I’m specifically looking at the projected availability of players at those two slots.

The question I’d like to answer: For which players would be it be absolutely necessary for the Cavs to make a deal to move up to No. 13? Obviously, such a deal would require the use of assets (either in future picks, current players or current salary cap space). So when would it be worth it? Let’s take a look at a few more tiers of players.

Would DEFINITELY have to move up to No. 13 (or higher): C.J. McCollum, Cody Zeller, Michael Carter-Williams, Kentavious Caldwell-Pope, Steven Adams, Shabazz Muhammad

Analysis: These players have probabilities of less than 10% of being available at No. 19. Following the top seven players in the draft, this group of six are the players most likely to occupy slots No. 8-13 on Thursday. If the Cavs go down the route of a scoring wing, Caldwell-Pope or Muhammad could be potential targets. For either, No. 13 would likely be the lowest possible trade-up point.

Would MAYBE have to move up to No. 13 (or higher): Shane Larkin, Dennis Schroeder, Kelly Olynyk, Sergey Karasev, Giannis Adetokunbo

Analysis: These players have probabilities of 25-50% of being available at No. 19. This next tier will likely occupy the slots between No. 14-18 in the draft. More likely than not, they won’t be sitting around and available at No. 19. Olynyk is a versatile scoring big, while international players Karasev and Adetokunbo3 are the two most popular secondary selections for Cleveland in these mocks.

Should PROBABLY still be available at No. 19: Gorgui Dieng, Rudy Gobert, Jamaal Franklin, Mason Plumlee, Lucas Nogueira

Analysis: These players have probabilities of 51-75% of being available at No. 19. These folks should still be waiting and would, under most circumstances, not require a trade-up for the Cavaliers. This is an eclectic group of big man, with the do-it-all Franklin being the only wing in this list. Conceivably, these are the back-up options for GM Chris Grant.

Should DEFINITELY still be available at No. 19: Tony Mitchell, Tim Hardaway Jr, Allen Crabbe, Reggie Bullock, Ricky Ledo

Analysis: These players have probabilities of more than 90% of being available at No. 19. The final grouping of likely first-rounders brings us to a few more intriguing Cavs-related names. All five of these players could be potential fits because of their scoring abilities. But, with 90% certainty at least for all, they should easily be available at No. 19 or later.

 

Final thoughts

That’s a lot of math, so far. But in the end, the NFL mock draft database wasn’t that helpful. What makes me think that this NBA database could be any better for fans or for the Cavaliers? There are at least two reasons that come to mind.

First of all, this provides the context for fans in their arguments about the most likely picks at No. 1. Instead of saying “ALL DA MOCKS LIKE LEN”, you can counter with the fact that just over 50% of mocks have said Noel consistently at No. 1 over the past two weeks. He remains the proverbial favorite, although his odds certainly are down from what they were last month.

Second, this then shows the possibilities for the Cavs with that second pick at No. 19. From the get-go, most fans have speculated that the team will not carry all four of its current draft picks into the 2013-14 season. So does that mean they’ll package them up for a better pick again, a la Tyler Zeller last season? Perhaps, and now you know the projected probabilities of dozens of players being available at No. 8 vs. No. 13 vs. No. 19.

In the end, I’m still thinking the pick will be Nerlens Noel. That’s what I said on lottery day. That’s what I was wishing when I wrote The Diff about his enormous potential last month. That’s what I told my 82-year-old grandfather when he called for my prediction earlier this morning. I’m sticking to it; now I also have the numbers to back it up.

___________________________________________________

Footnotes:

  1. For example: Nerlens Noel’s 47.5% availability at No. 2 means he’s the projected pick at No. 1 a total of 52.5% of the time. The same pattern can then be used to understand the rest of the percentages. []
  2. For additional thoughts on my Nate Silver-esque preferences for aggregating data, especially as it relates to sports predictions, click this link to my March Madness post []
  3. There was a very detailed New York Times profile on the Greece small forward today. []
  • JacobWFNY

    Woof. I’m not buying Franklin as a PG one bit, mgbode. Not at all. He’s a SG/wing. His assist rate was atrocious the past two years. He plays out of control.

    I’d wayyyy more rather have Waiters as a back-up PG. Franklin’s a poor man’s Waiters at the point. And that’s saying something.

  • mgbode

    I know I’m higher on him than pretty much anyone and admit as much. Yes, he plays out of control and doesn’t shoot well enough. He’s also a year and a half older than McLemore, which is one of the bigger negatives on him. Fully admit those things.

    However, under-sized? Same height as McLemore (6’5″), but a 3.5″ longer wingspan (and 3″ better standing reach). Much better handle, giving him the ability to play PG too (though, yes, would likely be TO prone).

  • mgbode

    pouting was out of character for Pippen? you remember a different player than I. though, I was definitely biased against Bulls players.

  • JacobWFNY

    Franklin is under-sized as a SF, which is what he played this past and year. Folks want to label him as the next Kawhi Leonard (SDSU connection). They’re very much different players along with the noted size difference.

    He’s a SG. I don’t buy him getting any minutes at PG. Sign me up for a small deal for Shaun Livingston or someone else instead of that experiment.

    That said: As a slasher/scoring wing off the bench, Franklin’s not a bad option. I wrote about him back in March and really do think he’ll carve out a niche in the NBA. Whether the best fit is in CLE, prob not.

    http://www.waitingfornextyear.com/2013/03/2013-nba-draft-second-tier-fits-for-the-cavaliers/

  • mgbode

    i’ve always touted him as a combo PG-SG. that’s his NBA position and it’s hard to say he was a pure SF at SDSU when they had him initiating the offense so much.

    on that thread, I mentioned he could be a Tony Allen type in the NBA.

    I’d rather have KCP because we do need the shooting, but I love Franklin’s motor/drive and athleticism, which is why he’s one of my favorite players in the draft (like MKG last year).

  • mgbode

    think I was too harsh on this thread today. apologies to all. one more day to go.