NBA Draft Analysis IV: Ranking the GMs

This is the fourth installment of the StatDance.com NBA Draft Analysis series. In Part I, I established the basis of how the process works - how to value players and establish an Expected Value based on their draft position. In Part II, I ranked the last ten draft classes by strength. And Part III was ranking the best-drafting teams over the last ten drafts.

Now, we get to the decision-makers and analyze the General Managers themselves. Originally, this was why I started the NBA Draft analysis project. I looked back at who was the GM for every team at every draft over the last ten years (2002-2011 drafts) and assigned each of them credit (or blame) for their team's draft that year. I do realize that every GM has help and/or orders on draft night - the team owner or another executive might force some picks and the GM has no choice but to take responsibility for it.

For example, Michael Jordan has been widely ridiculed for running the Bobcats poorly, but he has never actually been the GM in Charlotte. Therefore, he is not eligible for the rankings, despite most likely having a lot of influence over who gets drafted. Another example is Pat Riley, who was hired as the head coach and team president of the Heat in 1995, but has only had the General Manager title since the 2009 draft. While I assume that Riley was calling the shots since he got there, if I make a judgement call for any team, I have to look in to every team - and it still wouldn't be fair. Even Pat Riley still has a boss, and it could have been the owner's decision to draft Wade in 2003. The point is, there is no fair way to look at this unless we use a black-and-white system of who had the GM title.

I highly recommend looking back at Part I and Part III for further information about how these numbers were generated if you are interested. Now, on to the rankings!

The Top 10 Drafting Tenures as GM (2002-2011)

 The formula I came up to rank the top GM tenures with is overly complicated, but basically it values drafting well over a large sample size. To qualify, you must have at least 5 draft picks. This is a ranking I plan on taking another look at in the future with a wider historical window, since ten years doesn't really do this justice, especially if you consider how many players can still change their evaluation.

The Worst 10 Drafting Tenures as GM (2002-2011)

Much simpler ranking - total raw value short of Expected Value. I did not include Jerry Krause since Jay Williams' motorcycle accident is the only reason he would have made the list. This list favors failed GMs from early in the 2002-2011 window, but the more recent GMs might argue their picks still might turn out, and I tend to agree.

Ranking the Current GMs from 1 to 30

The simplest  rating so far - just ranking by percentage of Expected Value they have drafted as GM from 2002-2011. Some of these executives have multiple tenures during this period, but they have all been combined to give a overall drafting performance over the last decade.

Before Pat Riley emails me and complains that he doesn't deserve to be on the bottom of this list, let me just say that its not really fair to say Riley has drafted worse than Kahn and Pritchard. They have a much larger raw deficit, this is just simple percentage based ranking.

Some other interesting notes:

• Kevin Pritchard has a new job for the Pacers after he showed up in the top 10 worst tenures in the last decade ranking. He made the mistake in Portland a lot of people would have made, drafting Oden over Durantula. He was not far below average other than that pick.
• Michael Jordan was only GM for one year that I looked at in Washington (2002) - he got 70% of his expected value with four picks, two in each round. 
• 17% of the league's current GMs were not in charge of a team from 2002-2011
• Donnie Nelson looks like a poor drafter in this analysis, which might be true - but he has been in charge of the Mavericks for 10 years and has only had 1 pick in the top 24 in that time (he picked Devin Harris 5th, who has outperformed his EV). During that tenure they are second in the NBA in winning percentage. That's hard to do even if you find talent late in the draft - which he obviously can't do.

Thanks, as always, to Basketball-Reference.com for their data (both player stats and executive listings), this would be even more time-intensive if not for that resource.


This was Part IV of the StatDance.com NBA draft analysis.
Part I: Determining the expected value of a draft pick
Part II: Ranking the Strongest NBA Drafts

Part III: The best and worst drafting teams
Part V: Who did they miss? Looking at the undrafted free agents in the NBA - Coming Soon

NBA Draft Analysis Part III: Top Drafting Teams of the Last Decade

This is part III of the StatDance.com NBA Draft Analysis series. In part I, I went over how to fairly evaluate a draft pick. Basically, the contribution of each pick is measured using Player Efficiency Rating and the number of minutes played every year. The first eight years of a career are weighted, and then measured against the Expected Value of that draft pick. The Expected Value is a smooth-line historical average, based on years after being drafted and the draft position. So, if a player performs better than the average player drafted at his spot, he gets a positive evaluation, and vice versa. In part II, I looked at which drafts over the last ten drafts (2002-2011) were the strongest and the weakest.

While it is true that there are other ways to build your team from year-to-year, the draft is the only organic way to acquire talent. Free agency is great - you get a known commodity (usually!) but it is dangerous to count on, since you never know for sure what players you are able to sign. In order to field a strong team, you need to acquire talent through the draft - if you trade them away to land other assets, it still took an astute evaluation of the talent to get the players you need.

How are championship teams actually put together? Lets look at the last 5 NBA champions and see how they built their teams. We will look at the top 3 or 4 players in PER*MP for each team.

2012 - Miami Heat
LeBron James - 71408 - Free Agency
Chris Bosh - 37932 - Free Agency
Dwane Wade - 42737 - Drafted

2011 - Dallas Mavericks
Dirk Nowitzki - 58594 - Drafted
Tyson Chandler - 37886 - Trade
Shawn Marion - 38301 - Trade
Jason Terry - 40768 - Trade

2010 Los Angeles Lakers
Andrew Bynum - 39935 - Drafted
Kobe Bryant - 62086 - Drafted
Pau Gasol - 55029 - Traded

2009 - Los Angeles Lakers
Kobe Bryant - 72224 - Drafted
Pau Gasol - 66578 - Trade
Lamar Odom - 38445 - Trade

2008 - Boston Celtics
Kevin Garnett - 58898 - Trade
Paul Pierce - 56330 - Drafted
Ray Allen - 43033 - Trade

2007 - San Antonio Spurs
Tim Duncan - 71149 - Drafted
Manu Ginobili - 49646 - Drafted
Tony Parker - 53479 - Drafted

Only 8 of the 17 players that were the major contributors to a title had been drafted by the team that they took to a championship, but only 2/16 were actually signed in free agency.  This just shows that (at least over the last six years) it is vitally important to acquire assets in the draft so you can play them, or at least trade them for who you want.

In today's NBA, free agency greatly favors the stronger teams - great players want to win. And in order to trade for the players you need to complete your team, you need to have assets - thats where drafting wins you championships. The Heat don't land LeBron without drafting Wade. Every championship team is built by acquiring talent, and the biggest part of that is on draft night.

I've analyzed the NBA draft for the last ten drafts - 2002-2011 (2012 would be useless to analyze, since they haven't played yet). I compared the results for each team with their winning percentage over the last ten years. For each team, I compared the value they got out of their draft with the Expected Value of the picks they got "credit" for. To get "credit" for a draft pick, the team must either draft with their own pick and keep the player, or acquire a player's draft rights near draft time (usually on draft night, but occasionally afterwards).

Over the ten year span, I have winning percentages for each team, ranking from .358 (Charlotte) to .706 (San Antonio). Then I have the draft successes. Of the 30 teams, 15 have gotten over 100% of their Expected Value, and 15 have gotten less. The most successful team (Boston) has gotten 149% of its expected value. The least successful team drafting, the Clippers, has only gotten 68.5% of their Expected Value.

Can You Win Without Good Drafting?

While everyone knows that drafting better players correlates with winning basketball games, its nice to know that the system works, so lets analyze the results. Only 4 teams have a winning record over the last 10 years without also getting over 100% their Expected Value on draft night. The teams are: Dallas, Denver, Phoenix, and San Antonio Houston (Luis Scola was credited to Houston mistakenly and is now credited to the Spurs - once I switched him, the Rockets went below 100% and the Spurs went over 100%). The Mavs (Dirk), and the Spurs (Duncan/Parker/Manu) all drafted huge stars before the first year of my analysis. If I expanded my analysis another five years, the Spurs would be one of the strongest drafting teams. The Mavericks are only at 81%, after adding Nowitski they might still be close. The rest of that team is assembled with pieces that they got at a discount in trades.

Denver has only had two bad drafts in the last decade: 2002 (picking Nikoloz Tskitishvili 5th) and 2005 (picking Julius Hodge 20th). Since then, they have consistently had good drafts or gotten rid of all their picks (you can't mess up a draft if you trade the picks for proven commodities). They also turned Carmelo Anthony into a lot of assets. Still, definitely an exception with overall poor draft performance and a .551 winning percentage.

Phoenix is an even better example of poor drafting and yet still winning, having had only two good drafts in the last decade. Their Nash aqcuisition really propelled them a long ways, with a .595 winning percentage and one great pick (Amare).

Houston has drafted just below 100% of their Expected Value and yet has a winning percentage of .562. They have been good enough since they drafted Yao Ming in 2002 to avoid any high draft picks. While they are at 98%, they are only 13966 points away from being at 0%, the closest of any team to breaking even.

So, to recap - only 4 teams posted winning records without good draft results, and one of them was very close to breaking even.

Can You Lose While Drafting Well?

There are ways to win in the NBA without drafting well - so of course there are other ways to lose, too. It only takes a few really horrible free agent signings to completely tank a franchise. Fortunately for my analysis, it seems the teams that draft well are more likely to run their team well - there are only 5 teams that have a positive draft record and have managed to post a losing record: the Wizards, Knicks, Hornets, Kings, and 76ers.

The Wizards are barely positive with the draft at 102% of Expected Value, but they suffered from Michael Jordan's mismanagement from 2000-2003 and then the Gilbert Arenas era (four productive years for the Wizards and two contracts signed worth a total of $170 Million).

The Knicks are the classic case of great drafting and horrible management. To quote Bill Simmons (pretending to quote Isiah) "If you look at what I've done over the years, I always drafted well: Stoudamire, T-Mac, Camby, Frye, Ariza … you want to stockpile as many assets as possible, only because it gives you more options to do something dumb." What more can I say?

The Hornets (winning percentage: .488) have gotten an impressive 131.8%  of their Expected Value over the last ten years. Their winning percentage the last five years (after they moved back to New Orleans) is .530 - they had some bad years when they were in Oklahoma City.

Sacramento, despite their winning percentage over the last ten years of .456, has drafted from the second round or late in the first round the first five years of our analysis, 2002-2006. Since then, their first picks have been 10, 12, 4, 5, and 10. Their drafting record is stellar, having had only 1 year the past 10 under 100% of Expected Value (2006 when they picked Quincy Douby 19th with their only pick). Their winning percentage the last 5 years is an atrocious .320, even worse than the 10 year mark. Either the Kings are going to start winning titles now, or they are one of the worst-managed teams in the history of the league.

Philadelphia is the best example of good drafting and a bad record - boasting a ridiculous 147.3% value from drafting while posting a .474 winning percentage. The 76ers are a study of mediocrity - always playing well enough to avoid drafting too high (Iguodala 9th in 2004 being their only top 10 selection), but never having the talent to really start winning.

Ranking The Best Drafting Teams

So, to summarize the last two sections: 3 teams drafted poorly with winning records, and 5 teams drafted well and had losing records. That means 23 teams either posted winning records with positive draft results, or posted losing records with negative draft resuluts (under 100% Expected Value).

So we are left with the results! Here are the teams that have gotten the best value for their picks from 2002-2011. Of course, these rankings could all change a lot since the majority of the players are still playing, but this is how it stands today.

1. Boston Celtics 148.9% 
2. Philadelphia 76ers 147.2% 
3. Sacramento Kings 138.0% 
4. New Orleans Hornets 131.8% 
5. Cleveland Cavaliers 131.5% 
6. Miami Heat 128.1% 
7. Los Angeles Lakers 117.8% 
8. New York Knicks 116.2% 
9. Indiana Pacers 113.6% 
10. Utah Jazz 112.8% 
11. Detroit Pistons 111.6% 
12. Orlando Magic 105.5% 
13. San Antonio Spurs 104.4%
14. Washington Wizards 102.2% 
15. Chicago Bulls 101.4% 
16. Houston Rockets 98.32% 
17. Milwaukee Bucks 97.37% 
18. Atlanta Hawks 93.61% 
19. Oklahoma City Thunder 91.45% 
20. Memphis Grizzlies 90.72% 
21. Phoenix Suns 89.28% 
22. Charlotte Bobcats 88.19% 
23. Denver Nuggets 82.86% 
24. Toronto Raptors 82.82% 
25. Brooklyn Nets 80.83% 
26. Dallas Mavericks 80.55% 
27. Portland Trail Blazers 79.77% 
28. Golden State Warriors 77.49% 
29. Minnesota Timberwolves 72.65% 
30. Los Angeles Clippers 68.46% 

And here is each team, with every pick they get credit for over the last ten years. 

(note: Houston no longer has credit for Luis Scola)

(note: San Antonio now has credit for Luis Scola)

This was Part III of the StatDance.com NBA draft analysis.
Part I: Determining the expected value of a draft pick
Part II: Ranking the Strongest NBA Drafts
Part IV: We evaluate every NBA GM since 2002 - Coming Soon
Part V: Who did they miss? Looking at the undrafted free agents in the NBA - Coming Soon

NBA Draft Analysis Part II: Ranking the Draft Classes

In Part I of the NBA Draft Analysis series, I went through the methodology of determining a players worth, and listed some of the best value picks of all time. In this second installment (of five) I'll go through each of the last ten drafts (2002-2011) and look at some of the best and worst picks from each draft, and then rank the drafts in order of overall strength.

To briefly recap the value system, for every pick in the last ten drafts we average out the performance in an eight-year weighted average and determined the expected value from each of the picks. If you haven't read Part I yet, I wrote a pretty detailed explanation of the system.

You can view a gallery of the drafts (from Part I) directly on imgur from here. The overall rankings are based on the total production of all players drafted divided by the total expected value.

2002
Overall: 83.16%

Best Picks: Yao Ming (1), Amare Stoudamire (9), Carlos Boozer (34)
Worst Picks: Nikoloz Tskitishvili (5), Dajuan Wagner (6)

Only 11 players performed over 125% of their expected value. This draft was pretty weak across the board with the top 10, 11-30, and 31-57 slices under-performing.

2003
Overall: 120.21%

Best Picks: LeBron (1), Carmelo Anthony (3), Chris Bosh (4), Dwyane Wade (5)
Worst Picks: Darko Milicic (2), Mike Sweetney (9)

There were lots of standout picks in this draft, but with the four greats I have listed, it doesn't seem fair to list the others. David West at 18 performed at the Expected Value of a #2 pick, Josh Howard (29) at a #3 pick, and Mo Williams (47) close to #4 value.

Interestingly, this draft was not exceptionally deep. The overall 120% value mostly due to the players I've already mentioned. Overall, over a third (23/58) of the picks gave less than 25% of their Expected Value overall, an average number over the last decade.

It should be noted that the Darko pick by the pistons was made exponentially worse by the other members of the top 5. The Pistons had just won the NBA Championship - imagine if they had added a Carmelo or a Wade to that team.

2004
Overall: 100.28%

Best Picks: Dwight Howard (1) Andre Iguodala (9), Josh Smith (17), Kevin Martin (26), Al Jefferson (15)
Worst Picks: Shaun Livingston (4), Rafael Araujo (8), Luke Jackson (10)

This draft is exceptional for having a very shallow talent pool. With the names listed above, there was a lot of talent trafted in the first round. Three players went on to have production consistent with being drafted first overall (Howard, Iguodala, and Josh Smith), and four more produced number two overall values - Okafor (2), Ben Gordon (3), Luol Deng (7), and Al Jefferson (15). Thats seven players that could have been drafted #1 or #2 and been a worthy selection. But unlike most drafts, there were almost no players drafted in the second round that went on to have significant careers - Trevor Ariza (43) and Chris Duhon (38) were the only two exceptions.

2005
Overall: 114.04%

Best Picks: Chris Paul (4), Danny Granger (17), David Lee (30), Monta Ellis (40)
Worst Picks: Yaroslav Korolev (12), Julius Hodge (20)

The high overall rating of this draft is amazing considering the careers of the first and second picks (Andrew Bogut and Marvin Williams) - both have under-performed their draft position. The draft is bolstered by Deron Williams (3) and Chris Paul (4), and an exceptionally solid 17-40, with players like Monta Ellis and Louis Williams getting drafted in the second round.

2006
Overall: 77.91%

Best Picks: LaMarcus Aldridge (2), Rajon Rondo (21)
Worst Pick: Adam Morrison (3)

This draft was so weak it seems hard to call many of the picks bad - there just wasn't that much talent available. Only 4 players have given the Expected Value of a #4 pick, compared to a similarly weak 2002 draft when 7 players performed at that level. Only 7 players have given the Expected Value of a top 10 player.

2007
Overall: 95.75%

Best Picks: Kevin Durant (2), Marc Gasol (48)
Worst Pick: Greg Oden (1)

The only player to really stand out in this draft is Durant, who was half of the obvious #1/#2 pairing with Oden. Gasol as a late round pick was a great pick since he wasn't going to play the next year, which obviously paid off and he has given the third highest value of his draft class so far, despite missing a year.

2008
Overall: 117.06%

Best Picks: Westbrook (4), Love (5), Brook Lopez (10)
Worst Picks: Joe Alexander (8), Alexis Ajinca (20)

An exceptionally deep draft, with only the two "worst" picks not producing well among the first 29 picks. A very high 114% overall performance, without a group like the 2003 draft (LeBron/Bosh/Wade/Carmelo) makes this draft unique among the last ten. An amazing 34 players drafted performed at least at 75% of their Expected Value, the most in the ten years of this survey.

2009
Overall: 108.32%

Best Picks: Brandon Jennings (10), Jrue Holiday (17), Ty Lawson (18), Darren Collison (21), Marcus Thornton (43)
Worst Pick: Hasheem Thabeet (2)

With only three seasons to look at, many of these picks are still works-in-progress. Ricky Rubio has given almost nothing back to the Timberwolves, but the glimpse we saw of him last year shows he could still be a good investment of a #5 pick. This draft, much like 2008, appears to be very deep with 32 players overall performed at 75% or better of their Expected Value, the second-most in the last ten drafts. However, no player has given, so far, performance equal to that of a #1 pick. Blake Griffin is closest, having been drafted in the spot, and he missed a season due to injury - so all expectations are that he will exceed his Expected Value soon.

2010
Overall: 76.73%

Best Picks: Greg Monroe (7), Landry Fields (39)
Worst Picks: Evan Turner (2), Cole Aldrich (11)

These players have only had two seasons to perform, so its not very fair to be evaluating the draft already. But so far, it is remarkable that Landry Fields has been able to contribute the Expected Value of a #4 pick from the 39th pick. Greg Monroe has put up exceptional value, producing more than John Wall - and both of them over the EV of a number one overall.

2011
Overall: 96.39%

Best Picks: Kyrie Irving (1), Isaiah Thomas (60)
Worst Picks: N/A

While this is way too early to evaluate the draft using the metrics that I have designed, it should be noted that what Thomas did, as the 60th pick in the draft, is pretty remarkable - having the second-most productive rookie season of the draft class.

Here is a summary of the overall results:

This was Part II of the StatDance.com NBA draft analysis.
Part I: Determining the expected value of a draft pick
Part III: Team-by-team NBA draft performance - Coming Soon
Part IV: We evaluate every NBA GM since 2002 - Coming Soon
Part V: Who did they miss? Looking at the undrafted free agents in the NBA - Coming Soon