A few pages back I wrote about how, at least according to Vashro’s analytical modeling, this seems to be by far the strongest draft over the past 10 years in our part of the first round. I’ve also been keeping something of an eye on Looney and have been pleased to see he ranks high in most of the analytic draft models.
So I wanted to test to see if there might be some kind of bias I could discover in Vashro’s modeling that might result in Looney’s high ranking there.
I decided to find all the toolsy 3-4 tweener types I could over the last several years to see if Vashro’s (EWP) model systematically under- or overestimated their draft stock.
It didn’t. It turned out fairly balanced. Out of 34 such players, I found his model overestimated 10 players’ rankings in their class, got 16 players approximately right, and underestimated 8 players.
Maybe even a better sign: out of these 34 players, Vashro’s model order (in the draft the player was in) gave a closer approximation of the success of their career than actual draft order in 21 cases, about the same approximation as draft order for 6, and a worse approximation for just 7. So I have some confidence that while it misses on occasion, Vashro’s model is not particularly biased toward toolsy tweeners, and it also gives a closer approximation of a players’ success than actual draft order does.
With some general confidence in the model now, I may need some cold water to dampen my enthusiasm for what we might get at #12.
Johnson comes in at a 7.6 EWP (essentially an estimation of highest 2-year average of win shares during his career). So does Looney. And Kaminsky (though I didn’t test the model for 4-5 tweeners) comes in at 7.3.
That means if they come in right about this average (yes, very unlikely I know – it’s more likely that they be either somewhat higher or lower) in the prime of their careers, that prime should be about as effective as the past season for players such as Danny Green, Wall, Drummond, Teague, Korver, Nowitski, Lowry, Eric Bledsoe, and Demarre Carroll (or somewhere in the neighborhood of the 30-35th most effective players in the league).
Not bad for a #12 pick.
(Oubre projects to peak out at approximately this year’s season for Giannis, Wesley Matthews, or Rudy Gay – about the 55th most effective player in the league. Turner’s peak projects at this year’s Mirotic, Mozgov, or Faried level, while Dekker comes in at the 2014-15 Kemba Walker/Brandon Knight/Markieff Morris level. All still within the top 100 for win shares.)
What’s more, in looking at all the other toolsy 3-4 tweener types in over the last several years, Johnson’s and Looney’s 7.6 EWP scores are higher than all of those players except one, and quite a healthy bit higher. Unfortunately, that one was one of Vashro’s misses—Michael Beasely, but he probably failed for reasons other than talent.
Here’s the list of the tweeners I looked at, along with their EWP scores.
I’ll start with the players whose whose success relative to their draft class Vashro got approximately right (based on EWP rank order in that draft, not raw EWP score):
5.7 Rudy Gay
5.6 Harrison Barnes
5.2 Jeff Green
5.1 Danny Granger
4.9 Otto Porter
4.9 Gerald Wallace
4.8 Luol Deng
4.5 Trevor Ariza
4.4 Paul Millsap
3.2 Marcus Morris
2.9 Trevor Booker
2.9 Perry Jones
2.6 James Johnson
2.6 Austin Daye
2.4 Earl Clark
2.3 Quincy Miller
Here’s the players whose draft order the model rated too highly in relation to their draft class:
9.1 Michael Beasely
6.2 Terrence Jones
6.2 Marvin Williams
5.4 Derrick Williams
4.8 Jared Jefferies
4.5 Moe Harkless
4.2 Chris Singleton
4.1 Donte Greene
3.5 Anthony Randolph
3.1 Anthony Bennett
And here’s those the model underrated:
4.8 Kawhi Leonard
4.6 Draymond Green
3.7 Thaddeus Young
3.4 Tobias Harris
3.3 Tayshaun Prince
2.7 John Leuer
2.5 Chandler Parsons
2.0 Demarre Carroll