Thursday, June 21, 2012

NBA Draft - What is the best pick to have?

With the NBA Draft Lottery behind us and the actual NBA Draft a few days away I got to thinking.  What are the advantages of one draft pick over another?  Which pick has proved to be the most successful?  Which draft pick do I want for my team based on historical data?  Is there any benefit to winning the NBA Draft Lottery?  So, I started digging.  I went back through each draft since 1985 (the infamous Patrick Ewing lottery) to see just how well each pick has done. 

Comparing draft picks is difficult.  Positions are inherently different.  Scoring points is different than grabbing rebounds or dropping assists.  Teams have different philosophies that might not show off a certain players talents.  Injuries are always a menacing specter.  And of course some guys are simply more hype than talent.  Later picks go to better teams and might not ever get playing time, earlier picks my be pushed too hard too early and never round out before they are discarded.  There is no real good apples-to-apples comparison.  So, I decided to use a unit that is somewhat arbitrary, sometimes outright wrong, but overall is a good way to see how a draft pick panned out: All-Star appearances.  It ain't perfect, but it is as good a benchmark to having the NBA seal of approval as there is out there. 

This is all based off of past numbers, so it is in no way a prediction of how things will go in the future, but it does give us a sense of what is available and what your chances are of picking up a franchise type player.  Combing through the numbers I focused on 3 main questions: What percentage of the time has pick X produced an All-Star (what are the odds of getting an All-Star)?  How many All-Star appearances does pick X average (what should a team expect from their pick position)?  What pick has the most bang for the buck (what pick has had the best outcome IF a team does get an All-Star with that pick)?

One thing is clear, far and away the best pick to have is #1.  Over 27 years (actually 25 since the 2010 and 2011 draft classes have yet to produce any All-Stars) the #1 pick has produced 18 All-Stars, giving it a 67% All-Star Rate (ASR).  And those 18 player have had a total of 107 All-Star Appearances (ASAs), light years ahead of the next closest pick, pick #5 with 58 ASAs.  Shaq (15 ASAs) and Tim Duncan (13 ASAs) led the way for most appearances by the #1 picks.

Another point that jumps out is that there is a significant drop off after the top-5.  The top-5 picks have 274 ASAs while the rest of the draft total have 204 ASAs.  There ore only 3 picks that come close to the top 5 in terms of ASAs, pick #9 (30), pick #13 (29), and pick #10 (23).  The top-5 plus those three picks represent 74% of all ASAs in the Lottery era.

That is not to say that there aren't opportunities in the latter part of the draft.  Picks #18, #17, and #24 have higher ASRs (19%, 15%, and 15% respectively) than picks #8, #15, and #11 (7%, 4%, and 4% respectively).  But, most "non-lottery" picks (11 of 16) have an ASR under 10%.

Let's dive into the numbers.  First off, what percentage of the time has pick X produced an All-Star?  This is represented by the All-Star Rate (ASR), these are the top-10 picks by ASR. (In the case of a tie I put the lower pick ahead, because it proved to be a better value.)
  1. Pick #1 - 67%
  2. Pick #3 - 44%
  3. Pick #5 - 37%
  4. Pick #4 - 37%
  5. Pick #2 - 33%
  6. Pick #10 - 22%
  7. Pick # 9 - 22%
  8. Pick #6 - 22%
  9. Pick #18 - 19%
  10. 4 tied at 15%
View the full list here

Next, how many ASAs does a pick average?  Or, How many ASAs should a fan expect from a certain draft position?  Think of this almost as potency.  How strong a pick has shown to be.  We get this from the total number of ASAs divided over time (27 years).
  1. Pick #1 - 3.96
  2. Pick #5 - 2.15
  3. Pick #3 - 1.41
  4. Pick #2 - 1.33
  5. Pick #4 - 1.29
  6. Pick #9 - 1.11
  7. Pick #13 - 1.07
  8. Pick # 10 - 0.85
  9. Pick #17 - 0.52
  10. Pick # 7 - 0.44
View the full list here

Now this is a fun one.  Which draft pick has the most bang for the buck?  Meaning, IF my team does find that diamond in the rough, what might the longevity of our All-Star be?  We get to this by dropping all the years that a pick had 0 ASAs.  So while pick #13 has only produced 3 All-Stars, those players have garnered a total of 29 ASAs (thanks to Kobe, and Karl Malone going to 14 All-Star games a piece).  Giving it a whopping 9.67 Appearance Per All-Star rate (A/AS).  So, 24 times the #13 pick struck out, but when it has been an All-Star it tends to be a damn good one.  Here are the top-10 picks according to A/AS:
  1. Pick #13 - 9.67
  2. Pick #15 - 8.00
  3. Pick #1 - 5.94
  4. Pick #5 - 5.80
  5. Pick #9 - 5.00
  6. Pick #2 - 4.00
  7. Pick #28 - 4.00
  8. Pick # 10 - 3.83
  9. Pick #14 - 3.67
  10. 3 tied at 3.50
View the full list here

So what does this all mean?  Ultimately nothing, the sample is to small too have statistical relevance, and there is no way to predict where tomorrows All-Stars will be picked.  But there are some take aways.  The #1 pick is statistically the best place to be.  And being in the top-5 significantly increases a teams chances of getting a franchise player.  Also, until the streak is broken I think I would want my team to steer clear of Pick #12 (1 ASA) or Pick #8 (7 ASAs), but who knows, maybe they're due.  Anyway, it is some good info to throw around at happy hour, and to keep in mind during the draft next week.
Print Friendly and PDFPrintPrint Friendly and PDFPDF

No comments:

Post a Comment