With such high variability and talent drop-off in the NBA draft, it is crucial that teams gather as much information as possible before making a selection.
It’s been an inexact science in that regard, to say the least.
Though Kyrie Irving was considered a solid top overall draft pick in 2011, it’s hard to believe now that scouts referred to him as a pure point guard more than a scorer, and his ballhandling wasn’t considered to be a difference-maker. DraftExpress.com even referred to Irving as, at the very least, a rich man’s Eric Maynor, who played for five teams over five seasons.
When everyone was looking for the next Dirk Nowitzki in 2003, 7-footer Darko Milicic was said to be an ambidextrous post scorer who could shoot from behind the arc with impressive accuracy. ESPN’s Chad Ford called him a one-of-a-kind player who can run the floor, handle the ball, shoot the 3-pointer and play with his back to the basket. Milicic would miss all six of his 3-point attempts and average just 6.0 points for six teams in 10 seasons.
Kwame Brown drew comparisons to some of the greatest power forwards of all time and was said to have the ballhandling of a guard upon being selected No. 1 overall in 2001. Brown didn’t fit either of those descriptions while scoring 6.6 per game for seven teams over his 12 years in the league.
Over the years, teams began turning to data in an effort to get as much information as they can to more accurately evaluate prospects. Kevin Pelton’s wins above replacement player (WARP) projections and FiveThirtyEight’s career-arc regression model estimator with local optimization (CARMELO) ratings have been alternative data-driven tools for prospect evaluation, draft analysis and draft-pick position valuation.
But now, Stats Perform has developed one of the most sophisticated and comprehensive data-driven models to date. Our NBA Draft Model, which was first unveiled ahead of the last year’s draft, analyzes numerous factors to compare this year’s prospects to current and former players with similar profiles at the time they were entering the NBA. It’s important to note that last part.
We’re not going to give away everything that goes into it, but we can reveal that it includes historical data, volume and rate statistics, ratings, popular consensus draft rankings and biographical information. The model takes those data points and calculates a projected win share total for that player over the first four years of his career.
Remember, these are rankings, not a mock draft:
For those wondering, Patrick Williams from Florida State, his former FSU teammate Devin Vassell, ex-New Zealand Breakers guard RJ Hampton, Precious Achiuwa from Memphis and Serbian forward Aleksej Pokusevski come in at 11-15 in the rankings.
With a best-case scenario being that he turns out to be a player like Bradley Beal, it’s easy to see what has attracted teams to Anthony Edwards. At 6-foot-5 and 225 pounds during his only season at Georgia, Edwards was already bigger than Beal (6-3, 207) was during his one year at Florida.
Edwards’ 29.4% shooting from 3-point range at Georgia is less concerning when considering Beal only shot 33.9% from 3 for the Gators in 2011-12. Edwards wasn’t shy about putting them up, as he ranked second in the SEC with 275 attempts.
Much like his comparisons, Edwards, who finished fourth in the SEC at 19.1 points per game, has the confidence to pull up from any distance. The guard is a strong finisher at the rim and has the potential to become a prolific NBA scorer if he can develop a consistent, reliable jumper and maintain a more disciplined shot selection.
Edwards already has an NBA body, so taking contact in the paint should not be a problem. Because of his length, athleticism and playmaking abilities, our model puts him near the top of the draft.
There might be no prospect with player comparisons as compelling as Onyeka Okongwu. That’s because it might be an understatement to call him high ceiling, low floor. In a best-case scenario, he becomes the next Karl Anthony-Towns – the 2015-16 rookie of the year and a two-time All-Star. At the bottom end of the spectrum, the big man from USC turns out to be like Anthony Bennett – a former No. 1 overall pick who is arguably the biggest bust in NBA history.
Okongwu’s per 40-minute stat line in his one season with the Trojans (21.2 points, 61.6 field-goal percentage, 11.3 rebounds, 3.5 blocks) is very similar to KAT’s during his only year at Kentucky (19.5 points, 56.6 field-goal percentage, 12.7 boards, 4.3 blocks), while Bennett shot a lot more 3s and didn’t block as many shots in his only season at UNLV.
Because of his length and abilities as a defender, rebounder and rim protector, Okongwu should at the very least bring plenty of value in that area. The difference might be whether he develops an outside shot as Towns has done. At No. 3 overall, our model obviously points to him turning out more like KAT than Bennett.
It might be alarming that James Wiseman has a legendary bust listed as a comparison. But keep in mind that for the most part, Greg Oden had his career derailed because of chronic knee injuries. Though he only played three games before announcing he was withdrawing from Memphis to prepare for the draft, Wiseman displayed an ability to back defenders down and finish with a soft touch around the rim.
The 240-pound big man has also displayed some Ayton- and Oden-esque explosiveness with two-handed power dunks. While he does not spread the floor like a modern big and only made one 3-point attempt in his three games, Wiseman has shown an ability to shoot from outside the paint on occasion and made 19-for-27 (70.4%) from the free-throw line in college.
If Wiseman can develop a consistent shot, the lefty can become a force in pick-and-pop situations outside the lane. As the fourth-ranked prospect, our model points toward that being the case.
With the ability to score, pass and rebound, Tyrese Haliburton is a tall guard that drew comparisons to versatile initiators who also possess solid all-around games. The sophomore showed the ability to score from almost anywhere on the court before going down for the season with a broken wrist.
In this regard, it makes sense that the model points to a player like Zach LaVine, who averaged 25.5 points and 4.2 assists for the Chicago Bulls in 2019-20, as his best-case projection. Haliburton shot 50.9% from the field, 42.6% from 3-point range and 77.5% from the foul line over 57 college games.
He also compares with other strong defensive players as he led the team with 54 steals before his injury. The 6-5, 175-pound Haliburton is able to use his height to bring down rebounds and see the court clearly – much like Lonzo Ball, and his versatility makes the lengthy guard a clear lottery pick, according to our model.
Jon Chepkevich and Nadeem Patel also contributed.
Enjoy this? Subscribe to The Analyst to receive five stories each Friday from Stats Perform. It’s free.