From 1991 to 2006, there were 1,098 players who made their debut in the NBA (according to data at Basketball-Reference.com). By my count, 327 of these players were never drafted. This works out to an average of about 20 such players each year.
Just last year, 27 players (again by my count) debuted in the NBA without first being drafted. For many of these players, the stay in the NBA is quite brief. Specifically from 1991-92 to 2005-06, 300 players embarked on a career in the Association. For 128 – or 43% — this career lasted only one year. And 79 of these players played less than 100 minutes (which includes 23 who played less than 10 minutes).
Many undrafted free agents simply make a brief appearance, get their name listed at Basketball-Reference.com, and then go on to some other career. For some undrafted players, though, a lengthy career does result from the initial try-out. And it’s these lucky few we wish to talk about today.
From 1991-92 to 2005-06, there were 29 players undrafted players who played 100 games and averaged 20 minutes per contest. In other words, there are about two players each year who are undrafted and yet go on to become regular members of a team’s rotation.
When we examine the career performance of these players – reported in Table One – we see that Ben Wallace is easily the most productive undrafted player from the past 16 seasons.
Table One: The Top Un-Drafted Players (1991-92 to 2005-06)
Ben Wallace has now played 11 NBA seasons. In these eleven seasons he has produced 176.6 wins and posted a WP48 (Wins Produced per 48 minutes) of 0.348. Obviously the strength of Big Ben’s game is not scoring. He only averages 6.6 points per game. No, Wallace’s game is about rebounds and blocked shots. His ability to secure missed shots, thus either ending the opponent’s possession or allowing his team’s possessions to continue, is the key contribution he makes to his team’s success.
Wallace is not the only productive undrafted player. Next on the list are Brad Miller, Bo Outlaw, and Darrell Armstrong. Jose Calderon – whose career just began in Toronto – is already 5th on this list. In all, eight of the 29 players listed have a career WP48 above the average mark of 0.100.
What stands out when you look at this list is not just who is productive and who is not, but who the un-drafted tend to be. The scorer vs. role player story has been told frequently in this forum (with this column being the best explanation of the distinction). An average NBA player will score 0.400 points per minute. Scorers can be thought of as players who exceed this average, role players are those who do not.
When we look at the productive undrafted players we see that everyone, except Brad Miller, is clearly a role player. And Miller just makes it over the 0.400 bar.
In The Wages of Wins we defined scorers as players who average 0.500 points per minute. When we adopt this stricter standard we see that none of the undrafted players listed in Table One fits the WoW definition of a scorer. In fact, if you look at the 300+ undrafted free agents since 1991, only Trevor Ruffin (point guard for Phoenix and Philadelphia in the mid-1990s) managed to eclipse the 0.500 mark in scoring per-minute. Yes, only one undrafted player in 16 years managed to score one point every two minutes played.
When we look at the scoring rates of the undrafted a clear lesson is learned. Players who do not hear their name called on draft night should abandon their dreams of being the next Michael Jordan or Allen Iverson. To earn the NBA paycheck these players are going to have to contribute to team success is some way other than scoring.
Of course just because these players are not the primary scoring option on their teams it doesn’t mean they don’t contribute. When we look at all undrafted free agents, we see the same average productivity that we see in second round picks. Both post an average WP48 of 0.066. The implications of that equality should make for a good story on another day.
– DJ
Tom
July 26, 2007
I am curious about your evaluation of
role players such as Bruce Bowen and
Raja Bell. Both are considered valuable to
their teams as defensive specialists,
yet both have below average WP48 numbers.
Does this mean that they are overrated by their
coaches and the media, or does the WP48
number not accurately reflect their value
to a team as defensive specialists?
dberri
July 26, 2007
Tom,
In calculating WP48 players are evaluating in terms of the defense played by their team. So for Bowen to be undervalued by WP48, he would have to be much better on defense than the average defensive player on his team.
I did play around with incorporating 82games.com defensive data and posted the rudimentary results (I think the post is called incorporating defense). When the 82games.com data was incorporated the results were virtually the same.
Mr. Parker
July 27, 2007
Bruce Bowen is great at not screwing up.
That is his role on the spurs. Duncan, TP,
and Ginoboli take about 75% of the teams shots
when they are on the floor. That leaves nothing
for the other two guys. Whoever plays these
4th and 5th positions on the floor is going
to have a low wp48 unless they are a great
rebounder. As a sf that is pretty much impossible
in the NBA.
Brian
July 27, 2007
I know this isn’t basketball reltaed, but are there any stats that adress what happens when a college or pro sports team moves stadiums? (This is in light of the Miami Hurricanes fans’ outcry over potentially leaving the Orange Bowl for Dolphins Stadium.)
Steve
July 27, 2007
Comon, Bruce Bowen is overrated both by the media and coaches. He’s there just to annoy the best player on the other team with cheap tricks and dirty play. That’s the secret to the Spurs success. Decent offense, lots of dirty defense, and flopping.
Tim G
July 27, 2007
Am I missing something here? I see 6 other players with points per minute averages over .400
Tim G
July 27, 2007
I must be missing something here. I see 6 other players with point per minute averages over .400.
dberri
July 27, 2007
Tim G,
This statement was a bit unclear:
“When we look at the productive undrafted players we see that everyone, except Brad Miller, is clearly a role player. And Miller just makes it over the 0.400 bar.”
By “productive” I mean those with a WP48 above 0.100. Sorry for the confusion.
Jason
July 27, 2007
The perimeter defender who does not score much will be ‘undervalued’ by WP48. If someone causes a missed shot, they don’t get individual defensive credit if they don’t grab the rebound and only get 1/5th of the credit for the missed shot based on opposition’s stats and 1/5th of the credit for an opponent (nonsteal) turnover. This may not be fair, but I don’t really know a way to correct for it that seems reasonable either.
It seems _possible_ that on a team where scoring points is not a problem and rebounds are taken care of by the bigs, a designated pest who excels at forcing turnovers (e.g. makes people throw the ball errantly out of bounds) and who excels in making people miss shots might be valuable and his value won’t show up in the box score (even the degree to which things improve would be divided up 5 ways). But how common are these players? I suspect that situations where a guard/SF who cannot score is an asset are uncommon enough to say that the model still works, even if there are exceptions who produce ‘synergy’ not expected by their individual stats.
G
July 27, 2007
David,
Do you have an email address at which you can
receive specific questions about different
aspects of your work?
If so, what is it?
Thanks, in advance
dberri
July 27, 2007
G
dberri@csub.edu works. I am a bit behind in responding to my e-mail, but I generally get to everything eventually.
G
July 27, 2007
Thanks. I will contact you at that address.
bjk
July 28, 2007
It’s a real mystery why Ben Wallace wasn’t drafted. He may be only 6’7, but what was Barkley? He was drafted by the worst GM in NBA history, Unseld, and then promptly traded away for nothing after showing what he could do. And he probably deserves more playing time on the Bulls.
John
July 29, 2007
Hi David,
A couple thoughts: I’d be curious to see a similar sort of analysis for second round players. I’m also curious if there are metrics regarding the success of teams at drafting or discovering quality roll players in the second round or undrafted free agent ranks.
Bruce Bowen
July 30, 2007
I am not overrated are you out of your minds, I break more ankles than AI in his prime and I don’t even dribble!
Hows that for efficiency!