NBA Draft Results Last 14 Years

[OxAndFox]

This is an add on to the tanking thread where I don't really see a problem with it right now, and if anything the NBA has done a good job in that regard.
Anyway in the spoiler is a list of each teams lottery selections, by draft position over the last 14 years. I'm not suggesting this is the best way, or that there shouldn't be a deeper analysis, but this is just a quick version of putting a points system to draft picks.

The points system I went with is a simple #14-6 in the draft being from 1 points to 9 points. A top 5 pick jumps to 20 points for #5 and goes up 2 points each position until #1 is worth 28 points. In hindsight it probably needs to be in 3 categories, 1-4, 5-8, 9-14??
Anyway the below are the results and it's more to get a dialogue going of competitive balance etc.
Also we must understand, obviously, that not all of these picks are from the team itself, it includes trades etc so some teams may have had more during that period anyway.

What are your thoughts?

Cleveland – 200 points
Sacramento – 158 points
Orlando – 157 points
Philadelphia – 144 points
Minnesota – 144 points
Charlotte – 142 points
Washington – 135 points
Detroit - 134 points
Phoenix – 120 points
LA Lakers – 108 points
Houston – 101 points
New Orleans – 89 points
Utah – 71 points
San Antonio – 70 points
New York – 70 points
Atlanta – 68 points
Boston - 59 points
Portland - 59 points
Toronto – 59 points
Memphis – 57 points
Golden State - 55 points
OKC – 52 points
Chicago – 43 points
Milwaukee – 37 points
Dallas – 33 points
Indiana – 28 points
Brooklyn – 24 points
Denver - 23 points
LA Clippers – 12 points
Miami – 8 points

Spoiler:

Atlanta
#1 - 1
#3 - 1
#6 - 1
#8 - 1

Boston
#3 - 2
#6 - 1
#14 - 2

Brooklyn
#3 - 1

Charlotte
#2 - 2
#3 - 1
#4 - 1
#6 - 1
#9 - 3
#11 - 3
#12 - 1
#13 - 1

Chicago
#4 - 1
#7 - 2
#11 - 1
#14 - 1

Cleveland
#1 - 3
#3 - 1
#4 - 2
#5 - 2
#8 - 1
#14 - 1

Dallas
#5 - 1
#9 - 1
#10 - 1
#13 - 1

Denver
#7 - 2
#11 - 1
#13 - 1
#14 - 1

Detroit
#1 - 1
#5 - 3
#7 - 2
#8 - 3
#9 - 1
#12 - 1

Golden State
#2 - 1
#6 - 1
#7 - 2
#11 - 1

Houston
#2 - 1
#3 - 2
#4 - 1
#12 - 1
#14 - 2

Indiana
#6 - 1
#7 - 1
#10 - 1
#11 - 1
#13 - 1

LA Clippers
#8 - 1
#12 - 1
#13 - 1

LA Lakers
#2 - 3
#4 - 1
#7 - 1

Memphis
#2 - 1
#4 - 1
#9 - 1
#12 - 1

Miami
#10 - 1
#13 - 1
#14 - 1

Milwaukee
#2 - 1
#10 - 2
#14 - 1

Minnesota
#1 - 2
#2 - 1
#4 - 1
#5 - 1
#7 - 1
#9 - 1
#11 - 1
#13 - 1

New Orleans
#1 - 2
#6 - 2
#8 - 1
#10 - 2
#11 - 1
#13 - 1
#14 - 1

New York
#3 - 1
#4 - 1
#8 - 2
#9 - 1
#11 - 1

OKC
#2 - 1
#6 - 1
#11 - 1
#12 - 4
#14 - 1

Orlando
#1 - 1
#2 - 1
#4 - 1
#5 - 2
#6 - 3
#8 - 1
#11 - 1
#12 - 1

Philadelphia
#1 - 2
#2 - 1
#3 - 2
#10 - 2
#11 - 1

Phoenix
#1 - 1
#4 - 2
#5 - 1
#6 - 1
#8 - 1
#10 - 1
#13 - 3
#14 - 1

Portland
#3 - 1
#6 - 1
#7 - 2
#10 - 1
#11 - 1
#14 - 1

Sacramento
#2 - 1
#4 - 1
#5 - 3
#6 - 1
#7 - 2
#8 - 1
#9 - 1
#10 - 1
#12 - 1
#13 - 2

San Antonio
#1 - 1
#4 - 1
#8 - 1
#9 - 1
#11 - 1
#12 - 1

Toronto
#4 - 1
#5 - 1
#8 - 1
#9 - 1
#13 - 2

Utah
#3 - 1
#5 - 1
#9 - 2
#10 - 1
#12 - 3
#14 - 1

Washington
#1 - 1
#2 - 1
#3 - 2
#6 - 1
#8 - 1
#9 - 2
#10 - 1

[runtmc]

This is not the correct way to value draft slots -- your values are all over the place. You're using a linear function for 6-14, but by #14 being 1 and #6 being 9, you're essentially saying the #6 is 9 times more valuable than #14, but also that #13 is twice as valuable as #14. Neither of those are correct. You then jump in value to 20 at #5 for some unknown reason -- why #5 being the break point? But after that, for #1-5 you again start using a linear function, just with a slightly higher slope. This means #5 is more than twice the value of #6 (9 vs 20), but #4 is only 10% more valuable than #5 (22 vs 20). None of these values make any sense, nor is there any evidence or reason as to why you chose the values you did, nor any attempt to correlate the values assigned with the actual value of the pick. What it means is your results have essentially no meaning or value.

However, all is not lost, as this has been done before, quite a few times. Here's a good example article:

https://quantimschmitz.com/2023/04/02/how-valuable-is-each-nba-draft-pick/

He uses the formula y = -17.3ln(x) + 71.5

I would suggest re-compiling your list with something like this in mind. If you dont like his formula, there are others -- Kevin Pelton's for example. Anyhow, good luck.

[OxAndFox]

This is not the correct way to value draft slots -- your values are all over the place. You're using a linear function for 6-14, but by #14 being 1 and #6 being 9, you're essentially saying the #6 is 9 times more valuable than #14, but also that #13 is twice as valuable as #14. Neither of those are correct. You then jump in value to 20 at #5 for some unknown reason -- why #5 being the break point? But after that, for #1-5 you again start using a linear function, just with a slightly higher slope. This means #5 is more than twice the value of #6 (9 vs 20), but #4 is only 10% more valuable than #5 (22 vs 20). None of these values make any sense, nor is there any evidence or reason as to why you chose the values you did, nor any attempt to correlate the values assigned with the actual value of the pick. What it means is your results have essentially no meaning or value.

However, all is not lost, as this has been done before, quite a few times. Here's a good example article:

https://quantimschmitz.com/2023/04/02/how-valuable-is-each-nba-draft-pick/

He uses the formula y = -17.3ln(x) + 71.5

I would suggest re-compiling your list with something like this in mind. If you dont like his formula, there are others -- Kevin Pelton's for example. Anyhow, good luck.

Thanks for that. Yes it is completely flawed, as is any draft pick measurement because there are so many factors that are required and unless you have factored in everything, including trades, then it's impossible to get it right. This was more a basic understanding of what each team has done in the last 14 years with draft picks, rather than the picks themselves, but as you mention that's when you start running into trouble and circle back to the value of a pick.

[zimpy27]

The scoring system confuses me..

Why not just score 14 as 1 point, 13 as 2 points, all the way to 1 as 14 points?

The random jump at 5 isn't appropriate. If you wanted to scale to pick value there are some attempts based on real world outcomes

[JujitsuFlip]

This is not the correct way to value draft slots -- your values are all over the place. You're using a linear function for 6-14, but by #14 being 1 and #6 being 9, you're essentially saying the #6 is 9 times more valuable than #14, but also that #13 is twice as valuable as #14. Neither of those are correct. You then jump in value to 20 at #5 for some unknown reason -- why #5 being the break point? But after that, for #1-5 you again start using a linear function, just with a slightly higher slope. This means #5 is more than twice the value of #6 (9 vs 20), but #4 is only 10% more valuable than #5 (22 vs 20). None of these values make any sense, nor is there any evidence or reason as to why you chose the values you did, nor any attempt to correlate the values assigned with the actual value of the pick. What it means is your results have essentially no meaning or value.

However, all is not lost, as this has been done before, quite a few times. Here's a good example article:

https://quantimschmitz.com/2023/04/02/how-valuable-is-each-nba-draft-pick/

He uses the formula y = -17.3ln(x) + 71.5

I would suggest re-compiling your list with something like this in mind. If you dont like his formula, there are others -- Kevin Pelton's for example. Anyhow, good luck.

[leolozon]

The other problem I have is that if the goal is to see what team was "most advantaged" by draft positions, you just can't say that all #1 picks are worth the same. For exemple, the #1 pick in 2013 wasn't worth much even predraft. The #1 pick in 2023 was worth way more.

[brackdan70]

Thanks for sharing. Interesting. In that time span every team has had a top 10 and at least 3 lotto picks. And all but 2 teams have had a top 5.

[JujitsuFlip]

The other problem I have is that if the goal is to see what team was "most advantaged" by draft positions, you just can't say that all #1 picks are worth the same. For exemple, the #1 pick in 2013 wasn't worth much even predraft. The #1 pick in 2023 was worth way more.

That was my point in the other thread. In 2012 the #1 pick was super valuable. 1 year later, the 2013 #1 wasn't valuable at all.