What Does a Draft Pick Get You? Part III: First-Year Impact


For Part III of our as-yet-to-be-numbered series analyzing the NBA Draft, we’re going to focus on the first year.  Using some of the ideas first presented in Part I and Part II, this will look at the “instant gratification” that may or may not come from the Draft.

One thing that must be reiterated is that this will focus on the players’ performance the year they were drafted.  There were 17 players from the 2009 Draft class that did not play last season.  Some of these, like Blake Griffin and Ricky Rubio, are almost certain to play in the NBA in the future.  Others, like Robert Vaden and Robert Dozier, are far more likely to never see NBA game action.  In this collection of draft classes, there have been 179 players (besides the 17 from the 2009 class) who played their “rookie” year later than the rest of the draft class.  Seventeen of these players ended up earning All-Rookie honors in later years, and two — Larry Bird and David Robinson — even won Rookie of the Year.

However, since we’re theoretically trying to look at what might be expected of the 2010 Draft class next season, all of these players show a 0.00 AdjPR100 for their first year.  While some were calculated decisions (Bird, Robinson, Toni Kukoc, Manu Ginobili) and others were not (Greg Oden, Griffin, Rubio), none contributed to their teams on the court the year they were drafted.

An Overview

Let’s start things up again with a look at the Simple Average Adjusted PR per 100 by Draft Slot.

sa first year

The bars represent the first-year AdjPR100, while the line shows career average.  I don’t find it particularly surprising that the career average is higher in most cases, particularly as you move later in the Draft.  The #3 pick is the one lone outlier, and a quick check shows a number of players whose career failed to match the expectations set by their first-year performance — usually due to injury.  Among these are Bill Cartwright (22.15 vs. 11.78), Christian Laettner (21.67 vs. 14.64),  and Penny Hardaway (20.20 vs. 11.32).

The draft slot that showed the greatest increase after the first year was the #11 pick, and Pacer great Reggie Miller had one of the best improvements (9.42 to 17.11).

I’m going to do this one a little differently from the first two.  Rather than going through each draft grouping in varying levels of detail, I’m going to give you a look at first the 5-Star statistical analysis, then the First-Year Honors.  As in Part II, I’ll use spider charts, which will hopefully give you some sense of motion as you scroll through this post.  For a complete list of each draft grouping, simply click on that group’s header in the 5-Star Rating section.

The 5-Star Ratings

A more detailed explanation of this can be found in Part I of this series.  The charts below represent the AdjPR100 for the year that the player was drafted. Again, this is basically production, adjusted for Pace and Reliability.

Picks #1 to #3

5Star 1 to 3

The top three grouping shows, in my opinion, the kind of dramatic production the people hope for out of this area in the draft.  Over 60% of the players selected in this group turned in first-year numbers that rated them at 3-Stars or above (out of a possible 5-Stars).  Of course, the fact that these players are usually being added to teams lacking in talent provides ample opportunity for them to put up numbers.  Later in this post, I’ll break down how much playing time each of these groupings have seen the year they were drafted, but for now, I want to try to move quickly through each of the groups to keep that sense of motion, or “reading the clock” for these spider charts.

Picks #4 to #6

5Star 4 to 6The clockwise motion begins with a sizable swing towards the bottom of the dial.  The 3-Star and above ratings drop to about 40 percent, and only Chris Paul earns 5 stars.  The Rifleman’s — Chuck Person — first year of 19.87 was the second best from this draft grouping, and marks the high-water mark of his career.  This year’s Rookie of the Year, Tyreke Evans, earns a 4-Star rating with his 18.34.  Pacer bust Jonathan Bender’s 0.42 AdjPR100 marks the worst campaign of the 97 draftees that played with their draft class.

Picks #7 to #9

5star 7 to 9

The 3-Star and above ratings drop to under 30 percent, but the median remains at or above 2-Star.  Indiana Pacer Clark Kellogg posted the lone 5-Star season with a 21.64, but George McCloud’s 1.11 was better than only three of the 98 draftees who played.

Picks #10 to #12

5Star 10 to 12

No more 5-Star first-year campaigns, and fewer than 10% are 3-Star or above.  We know that some good players come out of this area of the draft, it’s just that very few of them make an immediate impact.  Reggie’s rookie year had 2-Star production that put him in the top third of this grouping.

There’s more to be gained from the trending of the charts, than there is any comments on each draft group, so just follow the clock for the rest of the sample, and I’ll hit the high points at the end of the section.

Picks #13 to #15

5star 13 to 15

Picks #16 to #18

5star 16 to 18

Picks #19 to #21

5star 19 to 21

Picks #22 to #24

5star 22 to 24

Picks #25 to #27

5star 25 to 27

Picks #28 to #30

5star 28 to 30

Picks #31 to #40

5star 31 to 40

Picks #41 to #50

5star 41 to 50

Picks #51 to #60

5star 51 to 60

A quick way to get a feel for the above charts is to center on the spiderweb for the #1 to #3 picks, then simply page down at a steady rate.  It will give the charts an animation, bringing the clockwise rotation towards “Never Played” to life.  Basically, history says that the chances of your team getting a significant first-year contribution after about the middle of the first round are pretty small.

From 16 to 60, the “best” first year performance was turned in by Mark Jackson for the Knicks in 1988.  As the #18 pick, Jackson posted a 4-Star 20.87 on his way to being the latest Rookie of the Year drafted in this sample.   However, that is far from representative.  Again, using our own A.J. Price for perspective, his 1-Star 4.51 rating was in the top 20% of all first-year performances under this rating system.  Of the 272 players drafted between 51st & 60th in this sample, he had the 12th “best” first-year performance.

One thing that is worthy of further study is whether this is an ongoing phenomenon or if this is actually changing.  The 2009 rookie class saw immediate impact from players taken in the late first or early second round.  Among these were Darren Collison, Taj Gibson and Omri Casspi from late first round, and Jonas Jerebko, DeJuan Blair and Marcus Thornton from the second round.  While it’s likely to remain true that the chances are slim with these picks, it would be interesting to see if there has been a significant increase in the hit rate over the last decade or so. That will have to wait for another part in this series.

Awards and Honors

The “Awards and Honors” we’ll talk about here are Rookie of the Year and First Team and Second Team All-Rookie.  (Note:  Second Team All-Rookie was not awarded until the 1989 season.)  Again, I want to use visuals as opposed to commentary.

Picks #1 to #3

Honors 1 to 3

Picks #4 to #6

Honors 4 to 6Picks #7 to #9

Honors 7 to 9

Picks #10 to #12

Honors 10 to 12Picks #13 to #15

Honors 13 to 15Picks #16 to #60

honors 16 to 60

Consistent with the 5-Star system, charting out the awards gets a bit pointless after the 15th pick, so I just condensed picks #16 through #60 on one chart.  Of the 1,427 players drafted between the 16th and 60th pick in this study, only 14 made First Team All-Rookie.  Another 33 made Second Team, and one (Mark Jackson) was named Rookie of the Year.  Given that less than 3% of these players were even a blip on this radar, the #16 to #60 chart equates to the “Line of Death” — the near-flat, bold line running at a 45% angle from upper left on the chart to the bottom right, tracing from “Did Not Play” to “None.”  If I were to show you all of the individual Draft Groups, they would all be virtually identical.

Still, rookie awards are not the final word on a player’s career, so while “Line of Death” is fun to say, it’s not 100% accurate.  Of the 209 players discussed in Part II of this series, 99 of them received no Rookie Honors.  Of the 263 players earning All-Rookie honors, 143 (excluding this year’s group) have received no other honors.  Like the Awards and Accolades over the career, these tell only a portion of the story for players

The Super Rookies

Over the last 33 years, some players have been able to achieve non-rookie honors during their rookie season.  Here’s a look at those “Super Rookies.”

All Stars (16) – Walter Davis, Larry Bird, Magic Johnson, Bill Cartwright, Isiah Thomas, Buck Williams, Kelly Tripucka, Ralph Sampson, Hakeem Olajuwon, Michael Jordan, David Robinson, Dikembe Mutombo, Shaquille O’Neal, Grant Hill, Tim Duncan, Yao Ming

All Defense (3) – Hakeem Olajuwon (Second Team), David Robinson (Second Team), Tim Duncan (2nd Team)

All NBA (6) – Larry Bird (First Team), Tim Duncan (1st Team),  Michael Jordan (Second Team), Walter Davis (Second Team), Phil Ford (Second Team), David Robinson (Third Team)

There were no rookies between 1978 and 2010 who won the MVP and Rookie of the Year, but Wilt Chamberlain did it in 1960, and Wes Unseld repeated the feat in 1969.

Playing Time

The standard fan mantra for their new rookie’s playing time is “more,” so I’m not even going to try to address the issue of what’s “enough.”  Each situation is unique, but here’s a little overview of what kind of action these players have seen.


There’s nothing particularly revelatory here: high draft picks play more during their first year than later picks.

Still, here are some nuggets about first-year playing time:

  • Only 21 players drafted over the last 33 years have started all 82 games in the year they were drafted.  Only two were selected outside of the Top 10 — Kelly Tripucka (#12) and Mario Chalmers (#34).  Larry Bird started all 82 games his rookie year, but he did not play with his draft class.
  • In this sample of 1,922 players, only three played more than 3,200 minutes in their first year.  Surprisingly, Tim Duncan (#1) was the only “high” draft pick, playing 3,204 minutes.  Michael Finley (#21) played 3,212 minutes, and Mark Jackson (#18) led everyone in this group with 3,249.
  • Pacers of interest: Only 13 of 272 players drafted between #51-#60 over the last 33 years played more during the year they were drafted than Price’s 865 minutes.  Of the 99 players drafted between #13-#15, only 13 played more minutes than Brandon Rush did in the 2009 season.  Tyler Hansbrough is near the bottom with 511 minutes, but he spent most of his 29 games operating under either a 15- or 22-minute medical limitation.  In effect, his 17.6 minutes per outing arguably indicates that he played almost every minute he was available to play.

The End of the Beginning

Over the first three parts of this series,  we’ve more or less laid the foundation of myriad discussions about the draft.  I’ve got some in the works (including an analysis of the #10 pick and a ranking of the last 33 draft classes) but I’m open to ideas on what other subjects to broach.  The feedback from the first two has provided some ideas, and I’m willing to try anything — provided I have the ability to get the data.

The draft is a month away, so there’s plenty of time to fill.

Tags: 2010 NBA Draft NBA Draft What Does A Draft Pick Get You?

  • http://ilevy.wordpress.com Ian

    I hope I didn’t miss it somewhere in the last three posts, but it might be interesting to add in the variable of position. Are point guards generally more successful than power forwards when taken at similar draft picks?

  • http://www.twitter.com/statcenter Statcenter

    Given that both first-year minutes and production seem to correlate so well with a player’s draft slot, I wonder if we’re seeing that the NBA draft is actually incredibly efficient overall. Yes, there are obviously misses (both among high and low picks), but overall there seems to be a huge amount of promise for top-3 picks, an opportunity to contribute heavily for picks 4-9, a chance to play for 10-15, and a chance to collect a paycheck from 16 onward. Is that fair?

  • http://www.eightpointsnineseconds.com Tim Donahue

    Ian – No you didn’t miss anything. That is a really good idea, but I’d have to see if there was an easy way that I could map the players to their positions. That data isn’t listed in my database at the moment.

    Statcenter – I think you’re right. I’m going to be doing “sidebars” to this series, shorter posts looking at specific tangents. I have one that will be called “Damned if you do…” which will talk about overall ability that is demonstrated by NBA front offices and scouting departments. The draft is an unbelievably difficult thing to master, and I think (most) do a pretty good job.

    However, the risk factor in the draft makes it a wonderful breeding ground for second guessers.

    Thanks for reading these guys. I hope you’re finding them worth your time.

  • DSMok1

    Nicely done again, Tim.

    It seems to be surprising to many people how many “busts” there are, even at higher picks of the draft.

  • They chose me

    You made a mistake– I believe 5 star was the highest accolade and 1 the lowest, but on your spider graphs the 12 o’clock position should be the premier, 5 star. It is noted as 1 star.

  • They chose me

    What an idiot– you have it correct. I misread the notation

  • http://ilevy.wordpress.com Ian

    Tim, I have definitely enjoyed these posts. They certainly give me hope for the Pacers draft this year. As far as looking at positions go I have resource you might be able to use. I am not sure if you read David Berri, but recently one of his readers came up with a way to automate Wins Produced calculations. A big component of automating these calculations was to find a formula to calculate the position designations to allow for the position adjustments. Andres Alvarez, who put it together came up with a way to automate the positions by looking at a player’s listed position, height, body-mass index, and assists. Here is the link explaining this automation:


    The positions are listed in the Wins Produced database he created:


    This might be a way to eliminate those difficult judgements calls and use a consistent formula to designate the position a player plays at. Thanks for all the great writing and analysis!

  • http://www.eightpointsnineseconds.com Tim Donahue

    They Chose Me – No…that’s my fault. I had put the #1 in my database to capture the order, but meant to remove it for the charts. I was just a little sloppier than I had intended in my preso.

    Ian – thanks for the info. I’ll have to look at it to see what I can do with it.

    Regarding Berri: I must admit that I really need to try to understand what he’s doing. I have his STUMBLING ON WINS book. Some of the results give me some heartburn, but it wouldn’t be honest of me to say that I don’t trust his work until I gave due diligence. A lot of smart people use it, so the right answer here is that I just need to do some homework. That’s basically why I haven’t used it much.

  • DSMok1

    Most of the advanced stats people don’t trust Berri’s work, Tim, including people like Dan Rosenbaum and Dean Oliver. Some discussion over at the Association of Pro Basketball Research boards: Stumbling on Wins: http://sonicscentral.com/apbrmetrics/viewtopic.php?t=2564
    Wins Produced: http://sonicscentral.com/apbrmetrics/viewtopic.php?t=877&highlight=wages+wins
    Win Scores: http://sonicscentral.com/apbrmetrics/viewtopic.php?t=1232&highlight=wages+wins

  • http://www.eightpointsnineseconds.com Tim Donahue

    DSMok1 – The analysis made me realize that there were fewer busts than I thought, but I hear what you’re saying. I’ve always had a much more pessimistic view of how “sure” high draft picks are than most others. (Understandable, given my team’s selection of George McCloud at #7 and Jonathan Bender at #5.)

  • http://www.eightpointsnineseconds.com Tim Donahue

    The thing that bothered me about Stumbling on Wins was the comment about playing in the Final Four boosting a player’s draft position by 12 slots. When I read the section, I was looking for methodology, and the way he talked, it made me suspicious. So, I went back and did a rough analysis, and found that from like the mid-80′s to present, the average draft position for players who had played in the Final Four the year they were drafted was 18, while the average for players who didn’t (read: everybody else) was 30.

    Now, I can’t say with 100% certainty that it was that simple, but it sure looks that way, and that seems grossly misleading to me. if nothing else, he was sloppy with the writing, either by making an improper claim about the impact, or by not providing enough support to make that claim.

    Thanks for the info…I looked around that board yesterday, but then lost internet connection last night.

  • DSMok1

    The image here: http://sonicscentral.com/apbrmetrics/viewtopic.php?p=31637#31637
    covers the likelihood of each type of result, in terms of wins added over the first 4 years (under rookie contract).

    30+ WSoRP: Superstar (~15 in the league in any 4 year span)
    20-30 WSoRP: Star (~30 in the league over a 4 year span)
    12-20 WSoRP: Solid Starter (~60)
    5-12 WSoRP: Role Player
    0-5 WSoRP: Bit Player
    <0 WSoRP: Worthless Player

  • DSMok1

    A number of the people criticizing Berri’s work in the “Wins Produced” thread above (which is the older thread from when his first book came out) are now employed by NBA teams…. which kind of indicates to me who was probably right.

  • http://www.eightpointsnineseconds.com Tim Donahue

    DS – Am I reading your chart correctly if I see that the mid-point (forgive me if I use the wrong terminology – just a lowly accountant) of a good chunk of the Top 10 is basically in the 5-12 Wins added, or Role Player level? That is to say, that’s where you get 50% above and 50% below.

  • DSMok1

    The term is median for midpoint. I’m a civil engineer by trade, so our numbers training isn’t that different! I never even took statistics.

    Yes, the next graph below that is more clear. From 4 to 10, you’ve got a greater than 50% chance of drafting a role player or worse, when looking cumulatively over their first 4 years. Some may develop towards the end of that period into something more and their average is still in the role-player level, though.

    By the end of the top 10, you’re seeing Patrick O’Bryant and Mouhamed Sene being taken. Ouch!

  • http://alwaysmillertime.com Josh Dhani

    Another fantastic piece Tim! This is probably my favorite one. Great job with the series and the statistical analysis. Enjoying read. Looking forward to the draft more and more as it gets closer

  • Pingback: Dutchman’s Links | Always Miller Time | An Indiana Pacers Blog

  • Dawar

    Great post, but as a huge Tim Duncan fan, i was a little steamed at how you forgot to mention him as a rookie to make an all nba team.(specifically the first team baby!)

  • http://www.eightpointsnineseconds.com Tim Donahue

    Wow…that’s embarrassing. I fixed it in the post. My apologies, and thanks for the catch. He also made Second Team All Defense.

  • Pingback: What Does a Draft Pick Get You? Part V: Ranking the Draft Classes — The 10 Worst

  • Pingback: What Does a Draft Pick Get You? Part VI: Ranking the Draft Classes 11-20