People who study statistics in basketball often look with envy at the numbers in baseball.  A hitter in baseball stands at the plate by himself.  Consequently, the numbers he generates is primarily about the actions he takes.  In contrast, basketball players run with four other players. The interaction between those players appears to diminish the meaning of the numbers generated.

The sport of football highlights the importance of interaction effects. The numbers we see for quarterbacks and running backs are very inconsistent across time.  This tells us that numbers in football are not just about the player, but also about the player’s teammates.  Consequently, forecasting the future in football – as Brian Burke recently noted – is very difficult.

People often argue that basketball is very much like football.  To illustrate this argument, just recently we discussed the impact Trevor Ariza will have on the Houston Rockets.  Ariza posted a 0.192 WP48 [Wins Produced per 48 minutes] for the Lakers in 2008-09.  Such a mark is nearly double what we see from an average player, indicating that Ariza will help the Rockets in 2009-10.  People argued, though, that Ariza had Kobe Bryant as a teammate last year; and without Kobe in Houston, Ariza won’t be nearly as productive.

Although this story was told, the evidence suggests otherwise.  Here is what Ariza did in 2007-08 and 2006-07 (much of which was spent in Orlando, without Kobe).

2007-08: 0.225 WP48

2006-07: 0.217 WP48

These numbers suggest that what Ariza does on the court is really about Ariza.

Now let’s think about baseball.  The Detroit Tigers just signed Aubrey Huff.  Last season Huff posted a 0.912 OPS, a mark that ranked 15^th among 147 qualified Major League Baseball hitters.  In other words, Huff ranked in the top 10% in baseball.  This suggests that Huff is one of the most productive hitters in all of baseball, and therefore, fans of the Tigers – like me — should be thrilled.

When we look at the numbers from this year, though, it’s a very different story.  Huff’s OPS in 2009 (prior to coming to Detroit) is 0.726.  This mark ranks 136^th out of 147 qualified hitters, placing Huff in the bottom 10% in the league.

If we look over Huff’s career we see a similar pattern.  Here is Huff’s OPS and ranking from 2003 to 2007:

2003: 0.922 OPS, 19^th (top 17%)

2005: 0.853 OPS, 45^th (top 41%)

2006: 0.813 OPS, 77^th (top 60%)

2007: 0.779 OPS, 103^rd (top 65%)

The past six years of Huff’s career demonstrates a great deal of inconsistency.  So which Huff did the Tigers add? Are they getting the player ranked in the top 10% in 2008? Or is it the player ranked in the bottom 10% in 2009? It seems likely that even Huff isn’t sure.  Huff’s job is to hit a round ball with a round stick, and that’s simply not an activity that can be predicted easily.

In the Wages of Wins we noted that the stories of Ariza and Huff are not unique.  The numbers attached to players in basketball are simply more consistent than the numbers we see in baseball. And this means that decision-making should be easier in basketball.  For example, the Portland Trail Blazers just signed Brandon Roy – a player ranked in the top 10% in the NBA [in WP48]– to a long-term contract.  If Roy stays healthy, the Blazers can count on him remaining a top player in the game.  And that will probably be true, regardless of his teammates.  A similar story can be told about Chris Paul, LeBron James, and Dwight Howard.  If these players stay healthy, teammates can come and go and these players will still rank towards the top of the league.  Likewise, a player like Jamal Crawford – who has consistently placed in the bottom half of the league rankings – is not going to transform into one of the top players in the game now that he is with the Hawks. 

If only people in baseball had the data we see in basketball.  Maybe the recent history of the Tigers would then be quite different.  The Tigers finished in second place with 88 wins in 2007.  They then added $40 million in payroll, and proceeded to lose 88 games and finish in last place.   Given what we know about inconsistency in baseball, this result doesn’t necessarily mean that the Tigers didn’t know what they were doing.  The Tigers might have simply suffered the wrath of baseball’s inconsistency.

In contrast, the Pistons added Allen Iverson in 2008-09 and announced that such a move would help the team contend.  The data, though, suggested fans of the Pistons were about to be disappointed (and eventually they were).

Hmmm…let’s think about this again.  The Tigers have a built in excuse when an entire season goes to hell.  The Pistons – because their data is better – don’t get to use this same excuse (although they try).  So maybe people in basketball should hope for the data we see in baseball.  

-DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Back in 2004-05, the Atlanta Hawks won 13 games with a -10.2 efficiency differential (offensive efficiency minus defensive efficiency).  Looking at the ten players who led this team in minutes played, we can see why Atlanta failed so miserably (WP48 = Wins Produced per 48 minutes).

Al Harrington: 2,550 min., -0.1 Wins Produced, -0.003 WP48

Josh Childress: 2,376 min., 10.8 Wins Produced, 0.218 WP48

Antoine Walker: 2,128 min., -0.1 Wins Produced, -0.002 WP48

Josh Smith: 2,050 min., 8.1 Wins Produced, 0.189 WP48

Tyronn Lue: 1,528 min., 2.4 Wins Produced, 0.076 WP48

Pedrag Drobnjak: 1,435 min., -2.7 Wins Produced, -0.091 WP48

Tony Delk: 1,340 min., 0.3 Wins Produced, 0.012 WP48

Boris Diaw: 1,201 min., 1.1 Wins Produced, 0.044 WP48

Jason Collier: 942 min., -3.1 Wins Produced, -0.158 WP48

Royal Ivey: 809 min., -0.7 Wins Produced, -0.039 WP48

The Wins Produced of this collection sums to 16.0, telling us that this team failed because the players it employed were generally (although not always) unproductive. 

Over the next three seasons the Hawks improved, winning 26, 30, and then 37 games.  Last season the Hawks took another step forward and won 47 games, advancing to the second round of the NBA playoffs. 

When we look at the players who led the Hawks in minutes played we can see who is responsible for this outcome.

Joe Johnson: 3,124 min., 8.6 Wins Produced, 0.131 WP48

Mike Bibby: 2,740 min., 7.8 Wins Produced, 0.136 WP48

Josh Smith: 2,421 min., 5.9 Wins Produced, 0.116 WP48

Al Horford: 2,242 min., 9.4 Wins Produced, 0.201 WP48

Marvin Williams: 2,093 min., 6.5 Wins Produced, 0.149 WP48

Ronald Murray: 1,975 min., 1.4 Wins Produced, 0.035 WP48

Maurice Evans: 1,840 min., 2.1 Wins Produced, 0.055 WP48

Zaza Pachulia: 1,473 min., 2.7 Wins Produced, 0.088 WP48

Solomon Jones: 675 min., -0.2 Wins Produced, -0.012 WP48

Acie Law: 560 min., 0.5 Wins Produced, 0.040 WP48

If we look over these two lists we note that only Josh Smith appears both times.  This suggests that the Hawks improved because different players got to call themselves Hawks.

Now the Hawks wish to take the next step.  Steve Aschburner of Sports Illustrated, though, tells us — in Can Hawks evolve into contender? — the plan is now changing.

The Hawks, for the most part, have had a “stay-cation” summer. With Rick Sund passing his one-year anniversary as GM this offseason, Atlanta diligently has kept intact the nucleus assembled by predecessor Billy Knight. …”Yeah, I like our club,” Sund told the AJC. “The only reason I say that is there’s still growth from within.”

The Hawks have not entirely stood pat.  The team traded Acie Law and Speedy Claxton to the Golden State Warriors for Jamal Crawford.  On draft night the Hawks chose point guard Jeff Teague. And in the past few days the Hawks signed Joe Smith.

Judging by what he did last year in college, Teague is not expected to be a major producer of wins next season.  The following numbers from 2008-09 suggest the same story can be told about Crawford and Smith.

Jamal Crawford: 2,479 min., 1.2 Wins Produced, 0.023 WP48

Joe Smith: 1,102 min., 1.2 Wins Produced, 0.051 WP48

Given the quality of players added, it does look like the Hawks are simply expecting the players currently on the roster to get better.  Now this is possible.  Marvin Williams, Josh Smith, and Al Horford are still in their early twenties.  Improvement at that age does happen. 

But we should be realistic about how much improvement is likely.  Basketball players – relative to what we see in baseball and football – are very consistent over time.  So although there is a standard age profile in basketball (players first improve and then decline), the slopes up and down are gradual.  Consequently, a dramatic change in player productivity is not something one should count upon.

Let’s imagine, though, that a dramatic change did happen.  Specifically, what if Williams, Smith, and Horford all boosted their WP48 by 50% (a fantastic percentage I just made up)?  These three players produced 21.7 wins last year.  Therefore, a 50% jump would result in 10.9 additional wins.  Such a leap moves the Hawks from a 45 win team (that is what their efficiency differential said they should have won in 2008-09) to a 56 win team.  Had this happened last season, the Hawks would have moved from the 4^th seed in the Eastern Conference all the way to…. okay, the 4^th seed.  Yes, the Cavaliers, Celtics, and Magic all won more than 56 games last year.  And these teams have all made moves this summer that will likely increase their win totals in 2009-10.  

So a 50% improvement in the three young players the Hawks are counting on to evolve (as Aschburner put it) would not be enough to overtake the top teams in the East. And even if that improvement happened, the Hawks are still likely to lose production from players like Mike Bibby, who are on the wrong side of the age profile. 

If we put it all together, the Hawks plan to contend in the East appears flawed.  Improvement from within is probably not going to close the gap between Atlanta and the top teams in the East.  This team might as well hope the Cavaliers, Magic, and Celtics suffer major injuries.  Certainly if such a hope was realized, the Hawks could also rise to the top.

As the saying goes, though, hope is not a plan.  And at this point, it doesn’t look like Atlanta really has much of a plan.  This is a team that has risen from the ashes because it acquired more productive players. To move on, more roster changes are needed. 

It’s possible fans of this team might disagree.  Two years ago Boston Celtics needed seven games to eliminate the Hawks in the playoffs.  This year, injuries to the Hawks appeared to derail their post-season run.  Despite such evidence, we must remember that the 82 game regular season is the better measure of a team’s quality.  In 2009-10 the Hawks finished with a win total in the mid 40s.  To seriously contend in the East, the team needs at least 15 more wins.  Improvement from existing players should simply not be expected to close this gap.  Yes, one can hope.  But again, that’s not really a plan.

- DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

The teams considered playoff contenders in the Western Conference include Portland, the LA Lakers, San Antonio, Dallas, Denver, New Orleans, Utah, and maybe Phoenix (and maybe Golden State).  With the exception of Phoenix (and Golden State), all these teams made the playoffs in 2009.  Only one Western Conference playoff team – the Houston Rockets – are generally omitted from the list of 2009-10 contenders.

The Rockets were dropped from the list when it was learned Yao Ming would miss the next season.  Coupled with the injury to Tracy McGrady, the Rockets now face the prospect of entering the next season without a “star” player.  Here is how Chris Mannix describes the team in Sports Illustrated: Houston is now a pale imitation of the team that took the Lakers to seven games in the Western Conference semifinals.

Mannix goes on offer the following quite from Daryl Morey: “When the previews come out, we’ll probably be in the back of the magazines, under the cologne ads.”

In John Hollinger’s preview at ESPN.com (insider access required), ten teams are listed ahead of the Rockets.  And when I listed the contenders earlier in the week, the Rockets were ignored.

The Above Average Rockets

Then I looked at the following list of veteran players who should play for the Rockets this season. 

Luis Scola: 2,448 min., 9.8 Wins Produced, 0.189 WP48

Shane Battier: 2,031 min., 6.2 Wins Produced, 0.147 WP48

Trevor Ariza: 1,998 min., 8.0 Wins Produced, 0.192 WP48

Aaron Brooks: 1,998 min., -0.5 Wins Produced, -0.012 WP48

Carl Landry: 1,467 min., 5.4 Wins Produced, 0.175 WP48

Chuck Hayes: 858 min., 2.5 Wins Produced, 0.139 WP48

Brent Barry: 857 min., 2.0 Wins Produced, 0.113 WP48

Kyle Lowry: 608 min., 2.3 Wins Produced, 0.182 WP48

An average NBA player posts a 0.100 WP48.  Of the eight players listed above, seven were above average last season. 

The Wins Produced from these players sum to 39.7.  The minutes, though, only sum 13,486.  An NBA team that avoids overtime will play 19,680 minutes in a season.  Consequently, it seems likely some (if not all) of these players will play more minutes in 2009-10.  And if per-minute performance doesn’t change (an issue I will address at the end of this column), increasing the minutes of these players (except for Brooks) will lead to more than 40 wins. 

The eighth seed in the West last year won 48 games.  So if the West is the same in 2009-10, the Rockets will have to come closer to 50 wins to make the playoffs.  This may be difficult.  But I am suggesting – contrary to the perceptions of this team – that it’s possible the Rockets can make the playoffs without McGrady or Ming.

A Collection of Non-Scorers

Once again, perceptions say this is not possible.  When we look at the scoring of the above eight players we can see why expectations are so low.

Luis Scola: 12.7 Points per game, 20.1 Points per 48 minutes

Shane Battier: 7.3 Points per game, 10.3 Points per 48 minutes

Trevor Ariza: 8.9 Points per game, 17.5 Points per 48 minutes

Aaron Brooks: 11.2 Points per game, 21.5 Points per 48 minutes

Carl Landry: 9.2 Points per game, 20.9 Points per 48 minutes

Chuck Hayes: 1.3 Points per game, 5.1 Points per 48 minutes

Brent Barry: 3.7 Points per game, 11.6 Points per 48 minutes

Kyle Lowry: 7.6 Points per game, 16.9 Points per 48 minutes

Scola leads this group with 12.7 points per game.  When we turn to scoring per 48 minutes we see only three players who exceed the league average of 19.8.  Scoring drives perceptions in the NBA, and the lack of scoring numbers from this group suggests the Rockets are doomed.

But wins are about more than scoring.  And when we measure these players contributions to wins (via Wins Produced), we see that Houston has fewer problems than people believe.  Consequently, I think it’s possible that Morey’s considerable reputation is about to become further enhanced. 

If the Rockets do make the playoffs without McGrady and Ming, the following will also occur:

• Morey is going to be considered for Executive of the Year
• Rick Adelman will be considered for Coach of the Year
• Aaron Brooks – who may lead this team in scoring – will be considered one of the best point guards in the game.

All of this is possible because this collection of non-scorers is also a collection of above average performers.  Unfortunately, it seems likely that none of the non-scorers will get much credit if this team exceeds expectations.

Performance History

Let me close by noting that the seven above average performers listed above have a history of productive play.  For example, consider the recent history of these players.

Luis Scola:

2007-08: 4.8 Wins Produced, 0.113 WP48

Shane Battier:

2007-08: 6.5 Wins Produced, 0.107 WP48

2006-07: 6.4 Wins Produced, 0.103 WP48

Trevor Ariza:

2007-08: 2.6 Wins Produced, 0.225 WP48

2006-07: 5.8 Wins Produced, 0.217 WP48

Aaron Brooks

2007-08: 0.3 Wins Produced, 0.020 WP48

Carl Landry

2007-08: 4.3 Wins Produced, 0.292 WP48

Chuck Hayes:

2007-08: 6.7 Wins Produced, 0.206 WP48

2006-07: 7.8 Wins Produced, 0.217 WP48

Brent Barry:

2007-08: 2.7 Wins Produced, 0.231 WP48

2006-07: 7.7 Wins Produced, 0.225 WP48

Kyle Lowry

2007-08: 4.3 Wins Produced, 0.101 WP48

With the exception of Aaron Brooks, these players were above average before 2008-09.  This suggests that these players will be above average in 2009-10.  If that happens, the Rockets have a good chance of being an above average team.  That means the Western Conference might have as many as ten teams contending for eight playoff spots.  And I haven’t even thought much about the Clippers, Thunder, Kings, Grizzlies, or T-Wolves (okay, I have thought about the T-Wolves).

- DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

The Golden State Warriors traded Jason Richardson to the Charlotte Bobcats during the 2007 draft.  In return, the Warriors received the rights to power forward – and lottery pick – Brandon Wright.  In 2008 the Warriors “earned” their own lottery pick, and selected Anthony Randolph, yet another player who plays power forward.

Given two lottery picks at power forward, it was not surprising to see the Warriors play at this position in 2008-09… Corey Maggette and Stephen Jackson?  Randolph and Wright only played a combined 1,817 minutes last season.  In their place, Don Nelson – head coach of the Warriors – turned to two players who are generally considered a small forward or a shooting guard.

When we turn to Table One we can see the problem with this choice.  Relative to the average power forward, Maggette and Jackson were above average scorers with an ability to get to the free throw line.  Once we move past scoring, though, we see quite a bit of red ink.  And that means, these two players were generally below average with respect to most aspects of the game. Consequently, as a power forward, Maggette and Jackson posted WP48 [Wins Produced per 48 minutes] marks that were not only below average, but actually in the negative range.  In contrast, both Randolph and Wright were above average with respect to WP48.

Table One: Comparing Corey Maggette, Stephen Jackson, Anthony Randolph, and Brandon Wright

The Warriors only won 29 games last year, and the problem at power forward was not the only reason for this result.  The injury to Monta Ellis and play of Jamal Crawford also hurt this team.  Still, one wonders what would happen if Nelson was a bit more traditional.  Because if he was, I think the outlook for this team could change.

The current outlook for this team is hardly positive.  John Hollinger at ESPN.com sorted the Western Conference teams into five groupings (inside access required).  The Warriors are in the fourth grouping, with ten teams ranked higher in the conference.  Hollinger’s explanation for this low ranking notes Nelson’s propensity to play a multitude of shooting guards every night.

What if Nelson, though, embraced tradition and decided to play actual power forwards (like Randolph and Wright) at the power forward position?  Imagine the following rotation, with WP48 from 2008-09 (at the listed position) reported:

First String (according to ESPN.com)

PG: Monta Ellis [0.165 WP48 in 2007-08]

SG: Stephen Jackson [0.106 WP48]

SF: Corey Maggette [0.071 WP48]

PF: Anthony Randolph [0.166 WP48]

C: Andris Biedrins [0.277 WP48]

Second String (according to ESPN.com)

PG: Stephon Curry [Rookie player, but let’s conservatively guess 0.050 WP48]

SG: Anthony Morrow [0.115 WP48]

SF: Kelenna Azubuike [0.083 WP48]

PF: Brandan Wright [0.141 WP48]

C: Ronny Turiaf [0.053 WP48]

An average player posts a 0.100 WP48.  Of the ten players above, six are above average at the listed position.  If each starter plays 32 minutes per night, and each reserve plays the remaining 16 (yes, I am guessing), then the Warriors would be expected to win more than 50 games next season.  In sum, if Ellis returns to his 2007-08 form, and Nelson learns to respect NBA tradition, this team can improve 20 games in the standings (assuming player’s performance doesn’t change, no injuries, etc…).

And that would mean there is one more contending team in the Western Conference.  This list already includes Portland, the LA Lakers, San Antonio, Dallas, Denver, New Orleans, Utah, and maybe Phoenix.  If we throw in the Warriors, at least one team that is thinking playoffs in 2010 is going to be disappointed. 

Of course, all of this depends upon Nelson’s willingness to keep his big man rotation confined to Randolph, Wright, Biedrins, and Turiaf.   Hollinger’s assessment of this team depends on Nelson expanding this rotation and staying true to what he has done in the past.  If Nelson follows Hollinger’s expectation, and keeps on looking for little guys to come up big, the Warriors can look forward to another trip to the NBA lottery (and perhaps another power forward Nelson won’t play).

Let me close with a brief comment on something else Hollinger said.  In his discussion of the Blazers he noted the following: “Nobody talks about the Blazers as serious contenders in the West, but we should.”  

When I read this I wondered… does this mean I am “nobody”?

- DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.