How good are NBA GMs at drafting players?
June 30, 2008
Gilbert Arenas. Carlos Boozer. Michael Redd.
What do all those players have in common? Well, in case you had forgotten (which would be hard for a serious hoops fan to do considering how constantly we are reminded), they were all selected in the 2nd round of the NBA Draft and are now bonafide stars. Compare those three with the likes of number one overall selections Kwame Brown and Michael Olowokandi and you might think NBA GMs don’t know what the heck they are doing. We agree that some don’t, but after analyzing the drafts since 1995, we have to say they are doing a pretty good job overall.
We took a look at the 2007-08 PER (Player Efficiency Rating) for each player drafted since 1995 and this is what we saw:
That trend line you see is the logarithmic regression. We won’t get into all the details, but basically that just shows you the overall trend as the draft progresses. As you can see, the top picks have paid great dividends despite the occasional busts that we pointed out earlier. The PER drops off considerably throughout the first 10 picks or so, then it levels off. By the time you get out of the draft lottery (past pick #14 or #13 depending on the season), it’s becomes more of a crapshoot. Players picked 15th will do better on average than players picked 30th, but not by much. Considering this, it makes a lot of sense to package a few lower picks and try to trade for a top pick, if possible. What are the odds your GM is really able to acquire a few solid players late in the draft? Doesn’t look likely.
Drafting is not a science, though, and I’m sure a lot of you have noticed some potential flaws in our chart by now. Let’s point them out:
- Looking at 2007-08 PER only is misleading. It’s true we are only looking at the most recent season’s PER so we are discounting the value of players who have seen better days. This makes our chart also a measure of longevity in some respects, but, to be fair, we did look at a chart for players drafted recently (since 2004) and the same pattern emerged. Also, we didn’t look at 1991-1994 because there are few players from those drafts still in the league.
- PER is not an accurate measure of a player’s value. True, there is no one statistic that can tell us the value of every player in the league, but we think PER is pretty good and while some players will be overrated and some underrated, it will average out in the long run.
There are probably some more flaws, but we don’t think the overall pattern would change much no matter how you look at it.
If you were a GM, what would your drafting strategy be?
Digg This!
Save to del.icio.us
Hype It Up
Comments
One Response to “How good are NBA GMs at drafting players?”
Got something to say?
Loading ...
This is pretty interesting…if I had a non-lottery pick, I’d try to trade down and get some established veterans or at least more than 1 draft pick to increase my odds of getting a good player.