Roster Construction: Surveying the League

[PhoSita]

Over at Wages of Wins, they've been writing a lot lately vociferously arguing that tanking doesn't work.  Most recently, Devin Dignam responded to the "Anonymous GM" article from Jeff Goodman by reiterating David Berri's arguments against the efficacy of tanking. 

Part of the argument is that teams that draft high in the lottery don't often meet with success within 4 years, and many of the players drafted high in the lottery who go on to win titles do so on a different team than the one that drafted them.

It's a compelling argument, but I can't help thinking that it's a very skewed way to look at things.  It's looking at the value of individual draftees in terms of the immediate effect they had on their team's success.  It occurred to me to ask, what if we look at all the teams in the league currently on track to contend and ask how those teams were constructed?

54. Since 1985, only two teams (the Miami Heat in 2006 and the Houston Rockets in 1995) have managed to win a title without winning at least 66 percent of their games (which works out to 54 wins in an 82-game season). So it seems likely that a team needs to win at least 54 games to be considered a contender.

Using this definition of contention from Wages of Wins, let's take a look at the league.  I'm going to include teams that won at least 54 games last year, that appear to have a chance of winning that many games this year, or that have the pieces in place to perhaps win that many games in the next season or two.

Miami: High Draft (Wade @ #5) + Free Agency (James, Bosh etc)

Indiana: Mid-lottery Draft (George @ #10) + Mid-1st Draft (Hibbert @ #17) + Free Agency (West)

San Antonio: High Draft (Duncan @ #1) + Late 1st (Parker @ #28) + Late 2nd (Manu @ #57)

OKC: High Draft (Durant @ #2, Westbrook @ #4) + Late 1st (Ibaka @ #24)

Houston: Free Agency (Howard) + Trade (Harden) + 2nd Round Draft (Parsons @ #38)

Cleveland: High Draft (Kyrie @ #1, Bennett @ #1, Thompson @ #4, Waiters @ #4) + Draft Trade (Varejao @ #30) + Free Agency (Bynum, Jack)

Dallas: Mid-lottery Draft (Dirk @ #9) + Free Agency (Ellis, Marion etc)

Detroit: Mid-lottery draft (Monroe @ #7, Drummond @ #9) + Free Agency (Smith, Jennings)

Chicago: High Draft (Rose @ #1) + Mid-lottery Draft (Noah @ #9, Deng @ #7) + Free Agency (Boozer) + Late 1st Draft (Gibson @ #26, Butler @ #30)

Golden State: Mid-lottery Draft (Curry @ #7, Thompson @ #11) + Free Agency (Lee, Iguodala)

LA Clippers: Trade (Paul) + High Draft (Griffin @ #1) + 2nd Round (Jordan @ #35)

New York: Trade (Anthony) + Free Agency (Chandler, Stoudemire, Felton) + Mid-1st Draft (Shumpert @ #17)

LA Lakers: Draft-Trade (Kobe @ #13) + Trade (Gasol)*

Brooklyn: Trade (Williams, Johnson, Pierce, Garnett) + Mid Lottery Draft (Lopez @ #10)

Denver: Draft-Trade (Lawson @ #18) + Late 1st Draft (Faried @ #22) + Trade (Gallinari, Chandler, McGee)

Memphis: Trade (M. Gasol, Randolph) + High Draft (Conley @ #4) + Free Agency (T. Allen)


*assumes the possibility of adding a high lottery talent or big name free agents to Kobe + Gasol in next 1-2 years*


A quick tally:

High draft:11
Mid-lottery: 9
Mid-1st: 2
Late 1st: 5
2nd Round: 3
Draft Trade: 3 (#13, #18, #30)
Free Agency: 17
Trade: 13


Some thoughts:

- Drafting in the high to mid lottery, trading, and signing free agents appear to be the most prevalent ways to acquire core pieces. 

- Whether you're selecting in the mid-1st, late-1st, or 2nd round, your chances of finding a core piece aren't great.

- If you look at all of those trades, the majority of them involved lottery assets; this was definitely the case in the trades for Chris Paul, Carmelo Anthony / Gallinari, James Harden, Deron Williams, and Javale McGee.  The Pau Gasol / Marc Gasol trade, and the trade for Zach Randolph, are the only ones to break that trend.

[BballTim]

Over at Wages of Wins, they've been writing a lot lately vociferously arguing that tanking doesn't work.  Most recently, Devin Dignam responded to the "Anonymous GM" article from Jeff Goodman by reiterating David Berri's arguments against the efficacy of tanking. 

Part of the argument is that teams that draft high in the lottery don't often meet with success within 4 years, and many of the players drafted high in the lottery who go on to win titles do so on a different team than the one that drafted them.

It's a compelling argument, but I can't help thinking that it's a very skewed way to look at things.  It's looking at the value of individual draftees in terms of the immediate effect they had on their team's success.  It occurred to me to ask, what if we look at all the teams in the league currently on track to contend and ask how those teams were constructed?

54. Since 1985, only two teams (the Miami Heat in 2006 and the Houston Rockets in 1995) have managed to win a title without winning at least 66 percent of their games (which works out to 54 wins in an 82-game season). So it seems likely that a team needs to win at least 54 games to be considered a contender.

Using this definition of contention from Wages of Wins, let's take a look at the league.  I'm going to include teams that won at least 54 games last year, that appear to have a chance of winning that many games this year, or that have the pieces in place to perhaps win that many games in the next season or two.

Miami: High Draft (Wade @ #5) + Free Agency (James, Bosh etc)

Indiana: Mid-lottery Draft (George @ #10) + Mid-1st Draft (Hibbert @ #17) + Free Agency (West)

San Antonio: High Draft (Duncan @ #1) + Late 1st (Parker @ #28) + Late 2nd (Manu @ #57)

OKC: High Draft (Durant @ #2, Westbrook @ #4) + Late 1st (Ibaka @ #24)

Houston: Free Agency (Howard) + Trade (Harden) + 2nd Round Draft (Parsons @ #38)

Cleveland: High Draft (Kyrie @ #1, Bennett @ #1, Thompson @ #4, Waiters @ #4) + Draft Trade (Varejao @ #30) + Free Agency (Bynum, Jack)

Dallas: Mid-lottery Draft (Dirk @ #9) + Free Agency (Ellis, Marion etc)

Detroit: Mid-lottery draft (Monroe @ #7, Drummond @ #9) + Free Agency (Smith, Jennings)

Chicago: High Draft (Rose @ #1) + Mid-lottery Draft (Noah @ #9, Deng @ #7) + Free Agency (Boozer) + Late 1st Draft (Gibson @ #26, Butler @ #30)

Golden State: Mid-lottery Draft (Curry @ #7, Thompson @ #11) + Free Agency (Lee, Iguodala)

LA Clippers: Trade (Paul) + High Draft (Griffin @ #1) + 2nd Round (Jordan @ #35)

New York: Trade (Anthony) + Free Agency (Chandler, Stoudemire, Felton) + Mid-1st Draft (Shumpert @ #17)

LA Lakers: Draft-Trade (Kobe @ #13) + Trade (Gasol)*

Brooklyn: Trade (Williams, Johnson, Pierce, Garnett) + Mid Lottery Draft (Lopez @ #10)

Denver: Draft-Trade (Lawson @ #18) + Late 1st Draft (Faried @ #22) + Trade (Gallinari, Chandler, McGee)

Memphis: Trade (M. Gasol, Randolph) + High Draft (Conley @ #4) + Free Agency (T. Allen)


*assumes the possibility of adding a high lottery talent or big name free agents to Kobe + Gasol in next 1-2 years*


A quick tally:

High draft:11
Mid-lottery: 9
Mid-1st: 2
Late 1st: 5
2nd Round: 3
Draft Trade: 3 (#13, #18, #30)
Free Agency: 17
Trade: 13


Some thoughts:

- Drafting in the high to mid lottery, trading, and signing free agents appear to be the most prevalent ways to acquire core pieces. 

- Whether you're selecting in the mid-1st, late-1st, or 2nd round, your chances of finding a core piece aren't great.

- If you look at all of those trades, the majority of them involved lottery assets; this was definitely the case in the trades for Chris Paul, Carmelo Anthony / Gallinari, James Harden, Deron Williams, and Javale McGee.  The Pau Gasol / Marc Gasol trade, and the trade for Zach Randolph, are the only ones to break that trend.

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

[PhoSita]

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

You're correct that Cleveland skews the number a bit.  You're welcome to remove Cleveland from the list and take a look at how the numbers change; I think that'd be valid.

The "high draft" number then stands at 7 and the "free agency" number then stands at 15.

I'm not sure what you mean about the trades.

Looking at the trades I mentioned:

- Chris Paul was traded for Eric Gordon (#7) and a Minnesota 1st (probably expected to be a high pick; ended up at #10)

- Carmelo was traded for Danilo Gallinari (#6) and Wilson Chandler (#23).

- Harden was traded for Jeremy Lamb (#12) and a Raptors 1st (probably expected to be a mid-lottery pick; ended up at #12).

- Deron Williams was traded for Derrick Favors (#3) and two 1sts (ended up being #3 and #21).

- Javale McGee was traded for Nene, who was drafted at #7 in 2002 by the Nuggets (draft-day trade for the pick).

[BballTim]

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

You're correct that Cleveland skews the number a bit.  You're welcome to remove Cleveland from the list and take a look at how the numbers change; I think that'd be valid.

The "high draft" number then stands at 7 and the "free agency" number then stands at 15.

I'm not sure what you mean about the trades.

Looking at the trades I mentioned:

- Chris Paul was traded for Eric Gordon (#7) and a Minnesota 1st (probably expected to be a high pick; ended up at #10)

- Carmelo was traded for Danilo Gallinari (#6) and Wilson Chandler (#23).

- Harden was traded for Jeremy Lamb (#12) and a Raptors 1st (probably expected to be a mid-lottery pick; ended up at #12).

- Deron Williams was traded for Derrick Favors (#3) and two 1sts (ended up being #3 and #21).

- Javale McGee was traded for Nene, who was drafted at #7 in 2002 by the Nuggets (draft-day trade for the pick).

  I misunderstood what you meant by "lottery assets", I thought you meant the picks that were traded. I'd say that out of the (young) players mentioned, the players traded for the elite talent (Melo, Harden, Paul, Williams) were by and large expected to be fringe all-stars on average.

[PhoSita]

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

You're correct that Cleveland skews the number a bit.  You're welcome to remove Cleveland from the list and take a look at how the numbers change; I think that'd be valid.

The "high draft" number then stands at 7 and the "free agency" number then stands at 15.

I'm not sure what you mean about the trades.

Looking at the trades I mentioned:

- Chris Paul was traded for Eric Gordon (#7) and a Minnesota 1st (probably expected to be a high pick; ended up at #10)

- Carmelo was traded for Danilo Gallinari (#6) and Wilson Chandler (#23).

- Harden was traded for Jeremy Lamb (#12) and a Raptors 1st (probably expected to be a mid-lottery pick; ended up at #12).

- Deron Williams was traded for Derrick Favors (#3) and two 1sts (ended up being #3 and #21).

- Javale McGee was traded for Nene, who was drafted at #7 in 2002 by the Nuggets (draft-day trade for the pick).

  I misunderstood what you meant by "lottery assets", I thought you meant the picks that were traded. I'd say that out of the (young) players mentioned, the players traded for the elite talent (Melo, Harden, Paul, Williams) were by and large expected to be fringe all-stars on average.

You may very well be right about that; my point here is just to show that those players, regardless of the evaluation of their talent, were originally acquired by the teams that traded them via lottery picks.

My thesis here, to the extent that I have one, is that whether they turn directly into building block players, or are used to acquire such players, lottery picks are the most prevalent ingredient in the construction of contending teams.  That doesn't mean it's not possible to construct a contender without lottery picks.  It's just much more difficult to do so.

I do think that this "survey" supports the idea that all-out tanking is not necessarily an especially effective strategy.  Even for those teams listed here that had one or more bottom-of-the-barrel seasons, many more positive moves unrelated to tanking were needed in order to reach contention (e.g. smart trades, picks later in the 1st round, and free agency pickups).  Tanking alone won't get you to the promised land.  I think that's clear.

[D.o.s.]

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

You're correct that Cleveland skews the number a bit.  You're welcome to remove Cleveland from the list and take a look at how the numbers change; I think that'd be valid.

The "high draft" number then stands at 7 and the "free agency" number then stands at 15.

I'm not sure what you mean about the trades.

Looking at the trades I mentioned:

- Chris Paul was traded for Eric Gordon (#7) and a Minnesota 1st (probably expected to be a high pick; ended up at #10)

- Carmelo was traded for Danilo Gallinari (#6) and Wilson Chandler (#23).

- Harden was traded for Jeremy Lamb (#12) and a Raptors 1st (probably expected to be a mid-lottery pick; ended up at #12).

- Deron Williams was traded for Derrick Favors (#3) and two 1sts (ended up being #3 and #21).

- Javale McGee was traded for Nene, who was drafted at #7 in 2002 by the Nuggets (draft-day trade for the pick).

  I misunderstood what you meant by "lottery assets", I thought you meant the picks that were traded. I'd say that out of the (young) players mentioned, the players traded for the elite talent (Melo, Harden, Paul, Williams) were by and large expected to be fringe all-stars on average.

You may very well be right about that; my point here is just to show that those players, regardless of the evaluation of their talent, were originally acquired by the teams that traded them via lottery picks.

My thesis here, to the extent that I have one, is that whether they turn directly into building block players, or are used to acquire such players, lottery picks are the most prevalent ingredient in the construction of contending teams.  That doesn't mean it's not possible to construct a contender without lottery picks.  It's just much more difficult to do so.

I do think that this "survey" supports the idea that all-out tanking is not necessarily an especially effective strategy.  Even for those teams listed here that had one or more bottom-of-the-barrel seasons, many more positive moves unrelated to tanking were needed in order to reach contention (e.g. smart trades, picks later in the 1st round, and free agency pickups).  Tanking alone won't get you to the promised land.  I think that's clear.

Unless you're the Thunder... then you make it all the way to the biggest stage before crying poor and blowing the team apart.

[PhoSita]

Unless you're the Thunder... then you make it all the way to the biggest stage before crying poor and blowing the team apart.

Eh, I don't think that's fair to the Thunder.  I think you can easily argue that they chose the wrong guy to hold onto -- Ibaka versus Harden -- but I don't think it was realistic for them to hold onto all four of their core guys.

I think they decided their long term prospects were better if they prioritized 3 core stars and then used the remaining money to surround them with talent, rather than investing heavily in 4 guys well above the luxury tax and then hoping veteran minimum pickups and mini-MLE guys would get them the rest of the way.

[D.o.s.]

Unless you're the Thunder... then you make it all the way to the biggest stage before crying poor and blowing the team apart.

Eh, I don't think that's fair to the Thunder.  I think you can easily argue that they chose the wrong guy to hold onto -- Ibaka versus Harden -- but I don't think it was realistic for them to hold onto all four of their core guys.

I think they decided their long term prospects were better if they prioritized 3 core stars and then used the remaining money to surround them with talent, rather than investing heavily in 4 guys well above the luxury tax and then hoping veteran minimum pickups and mini-MLE guys would get them the rest of the way.

As a basketball in Seattle supporter, I kindly ask that  you get your logic out of here while I rip on their team and management decisions.

[PhoSita]

Unless you're the Thunder... then you make it all the way to the biggest stage before crying poor and blowing the team apart.

Eh, I don't think that's fair to the Thunder.  I think you can easily argue that they chose the wrong guy to hold onto -- Ibaka versus Harden -- but I don't think it was realistic for them to hold onto all four of their core guys.

I think they decided their long term prospects were better if they prioritized 3 core stars and then used the remaining money to surround them with talent, rather than investing heavily in 4 guys well above the luxury tax and then hoping veteran minimum pickups and mini-MLE guys would get them the rest of the way.

As a basketball in Seattle supporter, I kindly ask that  you get your logic out of here while I rip on their team and management decisions.

Oh!  Well in that case, fire away.

[BballTim]

  Over a third of your high draft picks come from a team that hasn't won 25 games in any of the last 3 years and didn't add any top level players to their roster. That skews your results quite a bit. Also, how many of those "lottery assets" involved in the trades were high draft picks, and how many of those were set when the trade was made (as opposed to trading picks in future drafts)? I'm not sure you've really done much to disprove the original claim.

You're correct that Cleveland skews the number a bit.  You're welcome to remove Cleveland from the list and take a look at how the numbers change; I think that'd be valid.

The "high draft" number then stands at 7 and the "free agency" number then stands at 15.

I'm not sure what you mean about the trades.

Looking at the trades I mentioned:

- Chris Paul was traded for Eric Gordon (#7) and a Minnesota 1st (probably expected to be a high pick; ended up at #10)

- Carmelo was traded for Danilo Gallinari (#6) and Wilson Chandler (#23).

- Harden was traded for Jeremy Lamb (#12) and a Raptors 1st (probably expected to be a mid-lottery pick; ended up at #12).

- Deron Williams was traded for Derrick Favors (#3) and two 1sts (ended up being #3 and #21).

- Javale McGee was traded for Nene, who was drafted at #7 in 2002 by the Nuggets (draft-day trade for the pick).

  I misunderstood what you meant by "lottery assets", I thought you meant the picks that were traded. I'd say that out of the (young) players mentioned, the players traded for the elite talent (Melo, Harden, Paul, Williams) were by and large expected to be fringe all-stars on average.

You may very well be right about that; my point here is just to show that those players, regardless of the evaluation of their talent, were originally acquired by the teams that traded them via lottery picks.

My thesis here, to the extent that I have one, is that whether they turn directly into building block players, or are used to acquire such players, lottery picks are the most prevalent ingredient in the construction of contending teams.  That doesn't mean it's not possible to construct a contender without lottery picks.  It's just much more difficult to do so.

I do think that this "survey" supports the idea that all-out tanking is not necessarily an especially effective strategy.  Even for those teams listed here that had one or more bottom-of-the-barrel seasons, many more positive moves unrelated to tanking were needed in order to reach contention (e.g. smart trades, picks later in the 1st round, and free agency pickups).  Tanking alone won't get you to the promised land.  I think that's clear.

  It's obviously true that it's easier to make moves when you have higher draft picks. By the same token though, a solid majority of teams in your contender list built their cores without using or trading a top 5 draft pick of their own.