Zonkerbl wrote:Nivek wrote:dckingsfan wrote:82Games.com did one a while ago - I used some of their metrics but also weighted them for where they were drafted. EG didn't rate well at all. Like to hear your thoughts and rating system.
When I last did it (probably 2-3 years ago), I used PER as the metric, and looked at the first 4 years for each player -- reasoning that teams can control their picks for the first four seasons of the player's career. After that, teams essentially have to decide to reacquire the player with a new contract.
I used lottery era draft results. I calculated total PER credits, as well as the usual PER number (which is a per minute stat), and came up with the average value for each draft pick. Then I could measure how much each player exceeded or fell short of his expected production (based on the production of other players selected at his spot).
With some hits and some misses, Grunfeld got right around the "expected production" overall (all picks aggregated) based on the picks that he had.
Using PER wasn't ideal, but I hadn't put together PPA when I did the draft study. I want to redo with PPA.
Also, I didn't include stuff like trading the 5th pick for Miller and Foye -- which is something that would be easy to value when I get time to redo this thing better.
Can't wait to see the four year average value of one year's worth of Miller and Foye. Criminy. Or does EG get some credit for the cap space they left behind?
There was no cap space. The Wiz were well over the cap when they left. So, their departures merely meant they didn't have to pay them anymore.
My to-do list of analytic projects is growing. This is on it, but...time. I need more of it.