Early NBA Draft Prospect Similarity Scores
USA Today Sports
This is year’s NBA Draft is the big one, the kind of draft class that changes lives, adjusts spines, moves grown men to tears and tears to fears. The kind of once a decade accumulation of talent that can singlehandedly change the fortunes of a franchise and alter the course of history (it’s also a nice engine for hyperbole).
All of this assumes, of course, that the players projected at the top of the draft actually forgo the rest of their college eligibility and make for the professional ranks. The excitement also relies on the assumption that collective potential of this group of players has been accurately assessed and will be actually reached to a significant degree. With all the energy, interest and intrigue I thought it would be worth taking an early spin through some NBA Draft Prospect Similarity Scores and seeing who some of this year’s top prospects have been producing like.
If this is your introduction to my NBA Draft Prospect Similarity Scores, here’s the brief explanation:
Draft Prospect Similarity Scores is a project I began three years ago to compare each season’s crop of NBA draft prospects to those from previous seasons. The idea is simple, even if the execution and results aren’t. I wanted to compare draft prospects on purely objective measures, removing noisy elements like skin color, physical appearance and alma mater.
Prospects are compared to those from previous seasons on a 1000-point scale, across 21 different statistical categories. 1000 represents a perfect match of statistical profiles. It’s important to keep in mind that these are comparisons, not projections. Just because two players have similar statistical profiles when they leave college doesn’t mean they will have similar careers; there are too many other variables like fit, coaching, athleticism, health, motor, and just plain luck, that aren’t measure by this system. The comparisons are a snapshot – at this point in time Player A’s statistical production most closely resembles Player B’s. It’s also worth noting that you can often learn as much from the differences between players as you can from the similarities.
Using statistics through last night’s games I ran preliminary Similarity Scores for Andrew Wiggins, Jabari Parker, Joel Embiid, Julius Randle and Aaron Gordon — just enough to whet your appetites. The scores are in the spreadsheet embedded below. Tab over at the bottom to see each player.
Here’s a link to the full spreadsheet if you prefer that to this condensed, embedded view.
The instinct with these scores is always to jump right to the future, but as I mentioned above they’re not projections, just comparisons. But they still can offer some important information and help frame some of the other information we have about the players.
For example, Tobias Harris is a name you’ll see pop up a lot. He’s near the top of the comparable list for Gordon, Wiggins, and Parker. Although he was nowhere near the athletic equal of any of those three, Harris was an offensive-first freshman who played an inside-out game. I think that is at the heart of those comparisons. Another important thing to keep in mind is these comparisons are to players as they were coming into the draft, not who the players they turned out to be. Having Tyrus Thomas come out at the top of Embiid’s list may seem like a scary comparison, but that’s forgetting what an incredible athletic prospect Thomas was at LSU and the huge defensive impact he made in helping lead that team deep in the tournament.
The last thing that’s really striking is how unique most of these players look. Wiggins and Harris came out with a Similarity Score of 930, a very strong number, but the top of everyone else’s list is hovering around 900 or lower. That speaks to how unique the statistical production of these players has been, a fact that could be good or bad depending on the case.
For example, there is a 110 point difference in the Similarity Score between Gordon and Harris but a whopping 33 of those 110 points can be traced to Gordon’s 45% shooting at the free-throw line. If he somehow brought that number up to around 75% by the end of the season his three closest comparables would become Harris, Michael Kidd-Gilchrist and Chris Bosh, all with Similarity Scores above 900.
On the other end of that spectrum we see Marcus Smart separated from a lot of comparable players by his absurd steal totals. His 3.4 steals per 40 minutes creates a Similarity Score difference of more than 20 points from players like Derrick Rose, Kyrie Irving and James Harden.
I’ll be running Similarity Scores for a much wider group of draft prospects at the conclusion of the college season, so please come back and check them out then. If you’re curious about scores for previous groups of prospects you can also check out numbers for the 2011, 2012 and 2013 draft classes.