This post was originally intended as part of a series on the NBA draft for the Wages of Win Journal . It began life as a long e-mail conversation between Andres Alvarez, Prof.Berri and myself. It grew into multiple pieces in the Wages of Win Journal, a Wall street Journal Article and this blog. The companion pieces are:
And without further ado here’s the piece.
Some quick background
This article uses Wins Produced and WP48 [Wins Produced per 48 minutes] to evaluate player’s performance.* This measure uses three key components to evaluate a player:
- The player’s per minute box score statistics
- The player’s team’s per minute box score statistics
- The average performance at the player’s position (PG, SG, SF, PF or C)
A full explanation can be found here. To give a general scale, an average player has a WP48 score of 0.100. The very best players in the league usually have a WP48 over 0.300. To put this in perspective; an average player who plays a full season at 24 minutes a game would generate around four wins for their team. In contrast, a player posting a 0.300 WP48 would generate more than twelve wins in this time on the court.
What makes a good draft Pick?
In our previous post on the draft we focused a lot of attention on evaluating rookies based on their immediate impact to their teams. GM and fans can be notoriously shortsighted in their goals. The low probability of quick fixes in the draft combined with ill-advised impatience has been the end of the line for many a front office. When dealing with draft picks it is important to remember that you are getting a low cost player for four and not one year and any evaluation of draft picks should go beyond the rookie year.
When looking back at draft picks over history it is easy to get confused. We “know” or are told who the best players are and what the best draft classes are. For example, we are constantly reminded that the 2003 draft class is the best in recent memory. But we already know that public accolades and public perception do not correlate with performance and value. A constantly repeated theme of this blog, is that the wrong guys often are lauded, paid and played and good talent goes to waste . Rarely does the best rookie get selected for Rookie of the Year (In fact the Yay!Points! rule should states that scoring equals pub) . So when looking at draft picks we want to use statistics and analysis to come to our conclusions.
The first and most critical question is how do we measure a good draft pick. We should consider the following factors:
- The players contract is important so we will look at players over the first four years of his career (i.e his rookie contract)
- Overall productivity (i.e. Wins Produced) is important
- Per minute performance (i.e WP48 ) is important
- A small sample size is bad so any player playing less that 1600 minutes ( about 4.9 minutes a game) in 4 years will be excluded .
- Where the player is picked is important. His value should be compared to relative value at his position. So average Wins produced per pick and average WP48 per pick will be important.
Ranking the Draft Picks
With these factors in mind let’s work on ranking the draft picks. We will be looking at four factors:
- Wins Produced
- WP48 -Average WP48 at Pick
- Wins Produced-Average Wins Produced at Pick
Wp48 and Wins produced have been worked out for all the players but we will need to work out average WP48 and Wins Produced by pick. For Wins Produced by pick we simply take the sum of all wins by the players drafted at that position and divide it by 30 drafts. For WP48 by Pick we take the previously calculated total wins by Pick multiply by 48 and divide it by total minutes played. For the first thirty picks this looks as follows:
Couple of interesting points jump out from this table :
- If we rank picks in terms of Wins Produced we get:
- Pick 1
- Pick 3
- Pick 5
- Pick 2
- Pick 4
- Pick 9
- Pick 7
- Pick 11
- Pick 6
- Pick 10
- If we rank picks in terms of WP48 we get:
- Pick 1
- Pick 3
- Pick 5
- Pick 9
- Pick 26
- Pick 2
- Pick 11
- Pick 4
- Pick 30
- Pick 24
So it seems like in general teams get good value from the number one pick. But the data points to a talent evaluation model for NBA teams that is not very efficient at delivering value.
So now that we have our base data set for the players and the picks what’s next? Simple, we will now rank our players based on the four factors (WP48,Wins Produced,WP48 -Average WP48 at Pick,Wins Produced-Average Wins Produced at Pick). Once we have his ranks for these four number we will average them and proceed to rank the players based on their composite rank (or Draft Rank). Draft Rank should give us an effective tool for measuring the value and “goodness” or “badness “ of a pick.
So to work an example, let’s pick a player . We’ll call him Player K . Player K’s numbers are as follows:
Ranking by draft score Player K is thus the 140th ranked draft pick out of the full population.
Everybody Loves a Table
So now all that left is to post the table. What follows is a link to the table featuring all the draft picks for 1977 thru 2006 (the last year to complete their 1st 4 years in the league).
Some key points:
- The top five rookies Charles Barkley, Magic Johnson, Larry Bird, David Robinson and Chris Paul were really, really good. The analysis suggests that it’s extremely improbable that we will see their like again. (Barkley is currently listed in the data as having a position of 3.41 which suggests he played SF 59% of the time, this may merit further analysis)
- The average pick of the top 25 is 12.24. Only 3 were the top pick (Magic, Robinson & Shaq) and only seven were in the top 3 picks (and none at number 2). This, and the fact that 8 of the top 25 were picked at 20 or later, strongly suggests the league in general is not very skilled at pinpointing incoming talent.
- Of the top 100 picks half (and 122 of the top 200 picks) were taken after pick 9 suggesting there is always value in the later part of the draft
- Of the picks since 1999, only 4 guys (Paul #5, Lee #15, Rondo #17, Howard #28) crack the top 30 and not a single one from the vaunted 2003 draft class (Wade at 36 and Lebron at 58 are the two best from that class)
- The number 1 pick can and has been flubbed massively. 10 of the 30 #1 Picks fall in the second half of the rankings. The list includes some old WoW friends:
- Mark Aguirre
- Allen Iverson
- Kenyon Martin
- Glenn Robinson
- Kwame Brown
- Joe Barry Carroll
- Joe Smith
- Kent Benson
- Michael Olowokandi
- Andrea Bargnani
- Drafting Players under 20 is an extremely dangerous game. Only 13 of the top 200 players were aged 19 or younger at the end of their first NBA season
- Dwight Howard
- Tracy McGrady
- Kevin Garnett
- LeBron James
- Andris Biedrins
- Luol Deng
- Tyson Chandler
- Josh Smith
- Rashard Lewis
- Chris Bosh
- Kobe Bryant
- Cliff Robinson
- Andrew Bynum
Ranking the Draft Classes
So now that we’ve ranked all the players let’s take a look at draft classes. If we calculate total wins per draft class for their 1st 4 years and rank them accordingly the results are surprising:
The 79 and 84 classes at #1 and #3 should surprise no one but the 77 ,99 & 04 classes at #2,#4 & #5 fly in the face of conventional wisdom (so does 03 at #14). The 77 class featured 16 of the top 200 draft picks but no superstars (Yes, Knick fans I know Bernard King was drafted in 1977). 1999 featured three of the top 50 (Marion, Kirilenko & Ginobli all steals). 2004 featured 8 of the top 100 picks.
Looking at this table, it is easy to see that the draft is an inexact science and even worse the value of a pick is poorly understood. The difference over the last ten years from the best to the worst (2004 to 2000) works out to 256 wins or 4.3 wins per pick. One would think that given the public expectations, it can be a high risk proposition for a GM with a lottery pick. What fans may see as a terrible pick may in fact be a good one. But given the economics of the draft and low cost labor, it can also be very much high reward. My final post around the draft will center around this fact.
Note: Tables were corrected. Initial Tables reflected 1 Year not 4 Year Average (carry on)