10 first rounders in 5 years- What would that team look like?

[sofutomygaha]

Hi guys,

  I write Monte Carlo simulations sometimes. It's part of my job. It also leads to some big fantasy baseball time-wasting. Today I turned my attention on the draft because BasketballTim made me curious- what kind of team would we have if we stockpiled and used a whole roster's worth of 1st round picks?

  So here's the game- pretend that the draft position is a total crapshoot but imagine we'd had two first rounders in the last five drafts. Choose randomly. The team you get is the team you can compare to what we'll have in 2018 if we follow the same trajectory.

  It's mostly a game, but you get some perspective- these little sets of assets would be par for the course. Here's what turned up in seven tries, after you cut the dudes that didn't stick in the nba. I've ranked them by my preference, but that might not be yours...

Team 1, who has Anthony Davis
1. Cory Joseph
2. Tyreke Evans
3. Kawhi Leonard
4. Ed Davis
5. Anthony Davis
bench: Tony Wroten, Lucas Nogueira, Lazar Hayward, Andre Roberson

Team 2, who is intriguing
1. Michael Carter-Williams
2. Klay Thompson
3. Demar Derozan
4. Ed Davis
5. Larry Sanders
bench: Austin Rivers, Lucas Nogueira, Hasheem Thabeet, Chris Singleton

Team 3, who is also intriguing
1. Cory Joseph
2. Jimmer Fredette
3. Giannis Antetokoumpo
4. Blake Griffin
5. Miles Plumlee
bench: Dennis Schroeder, Omri Caspi, Jared Cunningham

Team 4, who is bad but nicely balanced
1. Avery Bradley
2. Klay Thompson
3. Gerald Henderson
4. Derrick Favors
5. Mason Plumlee
bench: Darren Collison, Meyers Leonard, Austin Rivers

Team 5, who is really bad
1. Jordan Crawford
2. Avery Bradley
3. Kawhi Leonard
4. Byron Mullens
5. Festus Ezeli
bench: Cory Joseph, Lucas Nogueira, Demare Carroll, otto Porter

Team 6, who has potential but is probably even worse
1. Toney Douglass
2. Bradley Beal
3. Evan Turner
4. marcus Morris
5. Tyler Zeller
bench: Roberson, Orton, Byron Mullens

Team 7, who is just a mess
1. Steph Curry
2.
3.
4. Kenneth Faried, Marcus Morris
5. Rudy Gobert
bench: james johnson, Lucas Nogueira, Festus Ezeli

[D.o.s.]

I don't think I understand what this means, but I love it.

[Cman]

Interesting. Did you randomly pick first rounders, or did you weight the chances of C's getting a top 5 pick (with their own 2014 pick), a pick in the low 20s (2014 Atlanta pick), etc?

[sofutomygaha]

Interesting. Did you randomly pick first rounders, or did you weight the chances of C's getting a top 5 pick (with their own 2014 pick), a pick in the low 20s (2014 Atlanta pick), etc?

I'm actually working on that. These were totally random. Do you have any suggestions for a reasonable weighting past the first year? It's such a mix of picks- 1 from the Clippers that'll be low, 2 from Brooklyn that will be...mid?...not sure..., 1 from philly that's protected so I could constrain that, one lesser of brooklyn and atlanta... also probably mid-range?

Maybe the best approach would be to assume that all but the Clippers pick and the lesser brooklyn/atlanta pick are in the top, say 20?


Hey... cool... there's a star by my name. Hey... cool... I love Bill Walton!

[PhoSita]

I think this would be more meaningful if you used a tighter range on the picks to be more realistic.  Ideally you could weight the probability so that a pick is more likely to end up in the middle of the range, making it possible to use a wider range.

In any case, it's an interesting exercise.

[BballTim]

 So, 60 or so picks and there's 3 players (Davis, Griffin and Curry) that are at least arguably better than Rondo. You could build a contender around that group, you just need to figure out how to get all of those 60 picks in a short amount of time so one star doesn't retire before you draft another.

[PhoSita]

So I tried to follow more or less what you're doing here on my own. 

I randomly selected picks for each of the Celtics' picks between 2014 and 2018, selecting within a range that seemed reasonable to me -- 25 to 35 wins this year, 25 to 45 wins next year, 30 to 45 wins the third year, and so on.

I did the same thing with the 2014 Hawks' pick, the 2015 Clips' pick, and the three Nets' picks (2017 being a swap option).  I assumed the Philly pick turned into two second rounders.  That means 9 picks over that 5 year period, with the right to swap picks with Brooklyn in 2017.

I took the players from the 2009 draft through the 2013 draft.  I didn't bother doing the lottery, so I just assigned the picks based on where the win total fell in the standings (this only really made a difference once: the #3 pick in Trial 3 which in real life ended up at #5 for Sacramento).  When there was a tie, I gave the tie breaker to my pick. 

I also honored playoff standings, so a worse record in the East sometimes resulted in getting the #15 or #16 rather than a late lottery pick.


Here's what I got --


Trial #1:

2014 --> #7 Stephen Curry, #19 Jeff Teague
2015 --> #11 Cole Aldrich, #19 Avery Bradley
2016 --> #15 Kawhi Leonard, #7 Bismack Biyombo
2017 --> #12 Jeremy Lamb
2018 --> #16 Giannis Antetokounmpo, #15 Lucas Nogueira

Team #1

Curry / Teague
Bradley / Lamb
Leonard / Antetokounmpo
Biyombo
Aldrich


Trial #2

2014 --> #8 Jordan Hill, #18 Ty Lawson
2015 --> #12 Xavier Henry, #22 Elliot Williams
2016 --> #16 Nikola Vucevic, #9 Kemba Walker
2017 --> #9 Andre Drummond
2018 --> #22 Mason Plumlee, #7 Ben McLemore

Team #2

Lawson / Walker
E. Williams / McLemore
Henry
Vucevic / Hill
Drummond / Mason Plumlee


Trial #3

2014 --> #7 Stephen Curry, #15 Austin Daye
2015 --> #3 Derrick Favors, #28 Greivis Vasquez
2016 --> #15 Kawhi Leonard, #23 Nikola Mirotic
2017 --> #12 Jeremy Lamb
2018 --> #24 Tim Hardaway, Jr., #15 Giannis Antetokounmpo

Team #3:

Curry / Vasquez
Lamb / Hardaway, Jr.
Leonard / Antetokounmpo
Daye
Favors



A fairly mixed bag, as you can see.  My methods are far from perfect, of course.  I think it illustrates that even without many high picks, you can get a pretty good haul of talent.  At the same time, you probably couldn't just go ahead and build a full competitive rotation out of 8 1st round picks over 5 years.

One factor worth mentioning here, of course, is that Danny may or may not be a better drafter than some of the GMs who chose the players that came up at certain spots in the sim.  In many places there were probably better players available.  It would be way too prejudicial to go ahead and play armchair GM with the benefit of hindsight, though.

[pokeKingCurtis]

 So, 60 or so picks and there's 3 players (Davis, Griffin and Curry) that are at least arguably better than Rondo. You could build a contender around that group, you just need to figure out how to get all of those 60 picks in a short amount of time so one star doesn't retire before you draft another.

This is like the Bizarro version of the hypothetical that some posit (in defense of tanking) where if we don't get a top 5 pick, we're not going to do anything else, at all. We're gonna be stuck with AB and stuck with Rondo for ever!

Your hypothetical is if you build solely through the draft and do nothing else XD.

Dunno if the (kind of) parallel is intended.



This simulation is pretty promising, though. This simulation is like if Ainge picked names out of a hat. Even the "really bad" one has Otto Porter, AB and Kawhi Leonard.

This kind of supports the hypothesis that the idea is not to get ONE pick and pray, even in a supposedly epic draft like this one, but get a bunch of picks. The list would be even less rosy with 5 picks (the normal number of picks).

I believe each and every one of the simulations has one top 5 pick as well and quite a few are busts.

TP for more data.

[PhoSita]

 So, 60 or so picks and there's 3 players (Davis, Griffin and Curry) that are at least arguably better than Rondo. You could build a contender around that group, you just need to figure out how to get all of those 60 picks in a short amount of time so one star doesn't retire before you draft another.

This is like the Bizarro version of the hypothetical that some posit (in defense of tanking) where if we don't get a top 5 pick, we're not going to do anything else, at all. We're gonna be stuck with AB and stuck with Rondo for ever!

Your hypothetical is if you build solely through the draft and do nothing else XD.

Dunno if the (kind of) parallel is intended.



This simulation is pretty promising, though. This simulation is like if Ainge picked names out of a hat. Even the "really bad" one has Otto Porter, AB and Kawhi Leonard.

This kind of supports the hypothesis that the idea is not to get ONE pick and pray, even in a supposedly epic draft like this one, but get a bunch of picks. The list would be even less rosy with 5 picks (the normal number of picks).

I believe each and every one of the simulations has one top 5 pick as well and quite a few are busts.

TP for more data.

The simulation that I ran -- and again, I recognize my methods are probably flawed -- did illustrate to some extent, I think, that it is possible to get a nice collection of talent without having a bunch of high picks.  Still, the only players in any of the trials with really transcendent potential were top 10 picks.

[pokeKingCurtis]

 So, 60 or so picks and there's 3 players (Davis, Griffin and Curry) that are at least arguably better than Rondo. You could build a contender around that group, you just need to figure out how to get all of those 60 picks in a short amount of time so one star doesn't retire before you draft another.

This is like the Bizarro version of the hypothetical that some posit (in defense of tanking) where if we don't get a top 5 pick, we're not going to do anything else, at all. We're gonna be stuck with AB and stuck with Rondo for ever!

Your hypothetical is if you build solely through the draft and do nothing else XD.

Dunno if the (kind of) parallel is intended.



This simulation is pretty promising, though. This simulation is like if Ainge picked names out of a hat. Even the "really bad" one has Otto Porter, AB and Kawhi Leonard.

This kind of supports the hypothesis that the idea is not to get ONE pick and pray, even in a supposedly epic draft like this one, but get a bunch of picks. The list would be even less rosy with 5 picks (the normal number of picks).

I believe each and every one of the simulations has one top 5 pick as well and quite a few are busts.

TP for more data.

The simulation that I ran -- and again, I recognize my methods are probably flawed -- did illustrate to some extent, I think, that it is possible to get a nice collection of talent without having a bunch of high picks.  Still, the only players in any of the trials with really transcendent potential were top 10 picks.

Considering the very definition of a transcendent player depending on who you ask, it only makes sense. The best players are hyped up the most so won't fall as much. The fact that there's only 10-20 of them at any given point in time means that (obviously) there isn't one every single draft. Having a top 10 pick helps, a lot.

But by that very same definition, banking on any one pick, even if it could potentially be very very good, could also very well be futile. You could end up with "just" a Camby or a J-Rich, and of course you could have a Michael Beasley or a Darko. Even in the very best drafts there's only very few "transcendent players", even in the top 5 of those very top drafts.

Being a 5 year period, of course the sample size would be a bit biased. Roy Hibbert is very much a difference maker right now (I think most would agree) but falls out of the scope of the simulations. Gilbert Arenas if he wasn't so troubled would have had a heck of a career.





Regardless, I think what you've said before re: this draft very much applies. The team will be in the "top 10". Then there's the LOTTERY. Let it be.

I also think the more picks, the better. It's a crapshoot. Ainge has obviously taken care of that.

A fantastic tidbit from the other night against the Mavs quoted by Mike: Pierce and Dirk are probably the only Hall of Famers from the '98 draft, and they're picked 9th and 10th.

Having too much choice with a higher pick may hurt more than help. I think that may have very well factored into us stealing Sully.

[Moranis]

Oklahoma City Thunder

2007 - 2 Kevin Durant, 5 Jeff Green
2008 - 4 Russell Westbrook, 24 Serge Ibaka
2009 - 3 James Harden, 25 Rodrigue Beaubois (traded on draft night with a 2nd rounder for BJ Mullens)
2010 - 18 Eric Bledsoe (traded for future 1st), 21 Craig Brackins, 26 Quincy Poindexter (Brackins and Poindexter were traded during free agency for Morris Peterson and Cole Aldrich)
2011 - 21 Reggie Jackson

They also drafted Landry with the first pick in the 2nd round in 2007 but traded him for a later pick and cash.
 
So if Oklahoma City had kept all picks and made no trades they would like this:

PG - Westbrook, Jackson, Beaubois
SG - Harden, Bledsoe
SF - Durant, Poindexter
PF - Green, Landry, Brackins
C - Ibaka

Even without Landry (who was a 2nd rounder anyway), I'd say that team would have worked out quite well.  A bit light at center, but that is what free agency is for and with assets like Jackson and Bledsoe riding the bench, they easily could have moved them (or traded Harden and moved one of them to the starting lineup).

[sofutomygaha]

Oklahoma City Thunder

2007 - 2 Kevin Durant, 5 Jeff Green
2008 - 4 Russell Westbrook, 24 Serge Ibaka
2009 - 3 James Harden, 25 Rodrigue Beaubois (traded on draft night with a 2nd rounder for BJ Mullens)
2010 - 18 Eric Bledsoe (traded for future 1st), 21 Craig Brackins, 26 Quincy Poindexter (Brackins and Poindexter were traded during free agency for Morris Peterson and Cole Aldrich)
2011 - 21 Reggie Jackson

They also drafted Landry with the first pick in the 2nd round in 2007 but traded him for a later pick and cash.
 
So if Oklahoma City had kept all picks and made no trades they would like this:

PG - Westbrook, Jackson, Beaubois
SG - Harden, Bledsoe
SF - Durant, Poindexter
PF - Green, Landry, Brackins
C - Ibaka

Even without Landry (who was a 2nd rounder anyway), I'd say that team would have worked out quite well.  A bit light at center, but that is what free agency is for and with assets like Jackson and Bledsoe riding the bench, they easily could have moved them (or traded Harden and moved one of them to the starting lineup).

Good point, Moranis. That's probably the gold standard for quality (or lucky) drafting. It's good to have the best-case-scenario in mind, but you'd probably have to agree if this little exercise has demonstrated anything, it's that the odds of coming through as well as Oklahoma did are low!

[sofutomygaha]

So I tried to follow more or less what you're doing here on my own. 

TP, Phosita, for taking it to the next level. Looking at the sim teams that we drafted, I'm thinking why don't we redo this for, say, the 2001-2005 drafts instead of the 2009-2013 drafts? That way we can say with much more confidence whether they produced competitive teams (two of your teams turned up Antetokounmpo, for example, to which my initial reaction was "home run!!" but of course he might not turn into a good player). I'm going to try to do that if I have some time later today

[PhoSita]

So I tried to follow more or less what you're doing here on my own. 

TP, Phosita, for taking it to the next level. Looking at the sim teams that we drafted, I'm thinking why don't we redo this for, say, the 2001-2005 drafts instead of the 2009-2013 drafts? That way we can say with much more confidence whether they produced competitive teams (two of your teams turned up Antetokounmpo, for example, to which my initial reaction was "home run!!" but of course he might not turn into a good player). I'm going to try to do that if I have some time later today

Rather than starting with the 2000 draft (a historically weak one), I figured 2003 would be a good starting point, seeing as that draft is so often compared to the upcoming draft.  Also, 2003-2007 is a period of rebuilding we're all very familiar with.


Here are my results.  I changed my ranges a bit to be wider in an effort to provide some more (realistic) variance.  As before, I ignored the lottery and just gave the team the pick that corresponded to standings (I don't have time to create a lottery simulation).



Trial #1:

2014 --> #8 TJ Ford, #16 Troy Bell
2015 --> #21 Pavel Podkolzin, #23 Sergei Monia
2016 --> #7 Charlie Villanueva, #21 Nate Robinson
2017 --> #23 Josh Boone
2018 --> #26 Aaron Brooks, #4 Mike Conley, Jr.

Team #1:

Conley / Ford
Robinson / Brooks
Villanueva
Boone

Note --> Wow, what an unfortunate haul.  Only Conley is a keeper.  This team is BAD.


Trial #2:

2014 --> #8 TJ Ford, #9 Michael Sweetney
2015 --> #15 Al Jefferson, #17 Josh Smith
2016 --> #17 Danny Granger, #21 Nate Robinson
2017 --> #6 Brandon Roy
2018 --> #21 Daequan Cook, #13 Julian Wright

Team #2:

Ford / Robinson
Roy / Cook
Granger / Wright
Smith
Jefferson / Sweetney

Note --> For a season or three, this team would've been pretty great thanks to some great mid-round finds. Beyond that, well, Smith and Jefferson are nice enough pieces.


Trial #3:

2014 --> #4 Chris Bosh, #9 Michael Sweetney
2015 --> #15 Al Jefferson, #25 Tony Allen
2016 --> #23 Francisco Garcia, #13 Sean May
2017 --> #12 Hilton Armstrong
2018 --> #2 Kevin Durant, #11 Acie Law

Team #3:

Law
Allen
Durant / Garcia
Bosh / Sweetney / May
Jefferson / Armstrong

Note --> Oh my goodness.  Besides the lack of a point guard, this team is awesome.  Pretty cool that it turns out the "2015" haul was the same as what the Celtics actually got in 2004.  This is as close as you get to a best-case scenario (except maybe the real-life Thunder).




Between Trial #1 and Trial #3, you see just how huge of a disparity there can be.  Good drafting and better luck are hugely important.

[Cman]

Interesting. Did you randomly pick first rounders, or did you weight the chances of C's getting a top 5 pick (with their own 2014 pick), a pick in the low 20s (2014 Atlanta pick), etc?

I'm actually working on that. These were totally random. Do you have any suggestions for a reasonable weighting past the first year? It's such a mix of picks- 1 from the Clippers that'll be low, 2 from Brooklyn that will be...mid?...not sure..., 1 from philly that's protected so I could constrain that, one lesser of brooklyn and atlanta... also probably mid-range?

Maybe the best approach would be to assume that all but the Clippers pick and the lesser brooklyn/atlanta pick are in the top, say 20?

What I had in mind was something similar to what PhoEdited.  Profanity and masked profanity are against forum rules and may result in discipline.a did. Basically ranges around certain numbers. Its all a bit fudged, but a bit more realistic than equal probability.

This all has been a fantastic exercise though. it does a nice job of highlighting *how difficult* it is to build solely through the draft.