Who Should the Celtics Take With the Third Pick in the 2016 NBA Draft?

The NBA lottery this year landed every team in their exact reverse order record, without a single team moving up or down for the first time in lottery history, one of the least scintillating “firsts” in televised sports history. For the bottom-dwelling Sixers and Lakers, this was very good news. It was especially good this year given the widespread agreement there are two clear top prospects in this draft[1. Ben Simmons and Brandon Ingram.] with a drop off after that. Of course those expectations could always turn out to be wrong, but for now the Sixers and Lakers are sitting with the chances to land a future star in this draft.[1. The Lakers front office is sitting in a particularly good spot given the disagreement over who to the at the top, just pick the one that’s left and no one will blame you almost no matter the outcome.]

After the top two picks there is little consensus as to who should go next In fact this year looks as though it may have one of the lowest correlations between my draft model projections and the consensus by scouts. Last year, which was a strong draft, there was a very strong relationship. It probably is a bit of a warning sign this year, in that a number of the more highly touted incoming freshmen had disappointing seasons, and some of those with better performances last are still doubted by the scouts.

Point being, there are good prospects in this draft after Simmons and Ingram, but there isn’t clear or obvious next star. And there are also, in my view a couple of potential landmine candidates.

This puts Celtic’s GM Danny Ainge in a bit of a tight spot. The third pick[3. Via Brooklyn in the Pierce/Garnett trade.] is going to have high expectations by the fan base and media, but leaves him with a significant possibility of missing on the pick only to watch a player taken afterwards excel.

To look a little further at the choices at the three spot, or after, I made a few charts based on the top prospects in both the Draft Express website and my workhorse PAWS model, then one that combines information from both the analytic model and scouting ratings. The charts combine standardized measures of the prospects’ performance in five categories, scoring, rebounds, distribution, blocks/steals and age/competition[2. All weighted to mimic the model, so that offensive rebounds count for more than defensive rebounds, for example, and age is the more important factor in the age/competition category] with their score in the model scaled on the right side.

First the Draft Express top ten prospects:

A few points on the chart above, there are basically three tiers, Simmons/Ingram, plausible top five pick, and Do Not Want.

• The PAWS model favors Simmons over Ingram by a significant margin, though Ingram is clearly above everyone else. Simmons is above average in every category tracked, while Ingram is close to average on most except for age/competition in part a reflection of his position as a wing as well as age.
• The next tier are all plausible top ten picks, though not necessarily the favorites as we’ll see in the next charts:
  • Dragan Bender scores well in all categories, relying primarily on his Israeli league minutes. He did struggle in the tougher Euroleague, where he played less. His overall rating is also hurt by his relatively low minutes played this year. On the other hand he is the youngest of the top prospects.
  • Kris Dunn, Jakob Poeltl, Jamal Murray and Henry Ellenson are rated by the PAWS model as good but not spectacular prospects.
• In the Do Not Want group is anyone scoring under a 4 in the PAWS model, which indicates a roll player or less.
  • Jaylen Brown leads the Do Not Want list, his overall play all year indicated less than a starter performance at the college level. In my training model not a single player went from this level of play in their draft year to become a star in the NBA.
  • Buddy Hield falls in that group as well. Hield very obviously excelled as a scorer, a very important basketball talent, no doubt. But he was below average, compared to all other prospects in every other category, including, of course, age. Historically scoring is the least reliable indicator going from college to the NBA. If Hield can hold his own in those other markers and play NBA defense and transfer his scoring, he could be a star, but…
  • Skal Labissiere was great at blocking shots and not a lot else.

Looking at a chart based simply on the top from the PAWS model, there are a few prospects that replace the Do Not Want group.

Again, a couple of notes on the chart, here the story is less model tiers than groupings based on the issues with the prospects perhaps not captured in the data available to the model. Looking only at those prospects not in the Draft Express top ten:

• Athleticism/Defense questions surround a couple of the prospects:
  • Denzel Valentine is slated just outside the top ten in the latest DX ratings, so his inclusion here is not much of a stretch. But his defense and athleticism are widely questioned.
  • Domatas Sabonis is relatively young and put up great rebounding numbers and scoring, but again faces questions over defense, wingspan, and possible “tweener” positionality.
• Point Guards who just pass:
  • Tyler Ulis is relatively highly-rated coming out of the high profile Kentucky program. He rates well in PAWS model, primarily based on his passing but doesn’t excel in any other area. (Though he is only significantly below average as a prospect in rebounding, unsurprising given his stature.)
  • Monte Morris is currently projected as an early second rounder. He rated well in the model based on being on of the best playmakers in the group. In the rest of the of the categories he is merely competent, unlikely to be a star but Morris could be a steal late in the first.[2. Ed. Morris has dropped out of the 2016 draft and is returning to school.]
• Young Big Men
  • Deyonta Davis is rated on the bubble of the top ten by DX and is one of the best rated players by the model. He is young, blocked shots, rebounded and was competent as a scorer and distributor for a big man.
  • Diamond Stone has the biggest discrepancy between the PAWS model and scout ratings. The model rates him highly based on his balance of scoring, youth, shot blocking, and rebounds. Stone’s only disappointing category was passing.
  • Not pictured is Matthias Lesson out of Chalon France[3. The French club of Clint Capela, who was available at the 17th pick, for those scoring at home!] who comes in just next on the list and rebounded like crazy with good scoring numbers plus good blocks and steals.

All of this brings me to my Scouting Informed PAWS model, which in cross validation had the best record of predicting performance in the NBA. In a year where the analytic models and scouting judgement do not agree it is probably especially important to try to weigh both, which is what this model does in a blunt sort of way:

After Simmons and Ingram, there is a general second tier with Bender, Davis, Poeltl, and Stone with virtually identical scores. Even the next tier down is only indicating a small adjustment in likelihood of success in the NBA, again leaving the choice pretty open and vulnerable to being seen in hindsight as a mistake. There is no clear candidate I would take against the field to become a star, but given his youth and impressive FIBA play, not included in my model, I have to lean toward Bender as marginally the best upside candidate.

We should keep the draft situation of the Celtics, who currently hold the third pick in mind. In addition to the third overall pick they currently hold the sixteenth and the twenty-second picks in the first round. Some of the top tier third pick candidates will not last much after that slot, while others like Stone, Davis or Valentine maybe available later if they can package their later picks to move up into the late lottery.