The Wages of Wins Journal

Twice in the past month ESPN.com’s Bill Simmons has made reference to our book.  Generally when members of the media mention our book I get pretty happy.  Of course, that's because typically the response to our work – by people who have actually read our work – has been really positive. Simmons, though, has been an exception.  Not in the reading the book part (which I don’t think he has) but in saying positive stuff about our work (which he doesn’t). And this is odd. 

Bill Simmons clearly likes Malcolm Gladwell (see here and here).  And Gladwell clearly like The Wages of Wins (see here).  And I really like the writing of Bill Simmons (I have no link to prove this, so you'll just have to believe me).  But despite all this “liking” going around, Simmons seems none to fond of our book.

A few weeks ago he wrote the following: “Next time, Malcolm Gladwell needs to consult with me before he recommends an NBA book that argues Jerome Williams was among the strongest players of his generation.”

A few days ago Simmons said, “One more note on this: we get carried away with basketball statistics nowadays, as evidenced by the new book that rated Allen Iverson as the 90th best player in the league during his MVP season. Why make it so complicated? Just add up the point, rebound and assist averages for franchise guys during the playoffs: If the number tops 42, you're probably talking about a pantheon guy. You could even call it the 42 Club, just as exclusive as the Five-Timer Club on SNL, only without the NBA equivalent of Elliott Gould.”

Now one should notice, that's two references to our book without saying “The Wages of Wins”, or even “Dave Berri, Martin Schmidt, and/or Stacey Brook.”  Would it kill the guy to say the name of the book?

Okay, he doesn’t like what our book says about Jerome Williams or Allen Iverson.  What we say is of course based on our Wins Produced measure, which as we detail in the book is a simple, complete, and accurate measure of player performance in the NBA.

Simple is of course a relative term. What Simmons proposes is really, really simple.  Simmons says we should just add up points, rebounds, and assists.  So Simmons has cornered the market on simple.

His metric also has other advantages.  First of all, Simmons stated after the season that Kobe Bryant was the MVP.  Although during the playoffs he fell in love with Dirk Nowitzki, and then later Dwyane Wade, the measure Simmons has developed does list his first love, Kobe, as the best player in the 2005-06 season.  As for Allen Iverson, he joins both LeBron James and Kobe as the only members of his 42 Club in the regular season.

Of course, critics might note a few problems with what Simmons proposes.  The weaknesses of Iverson – poor shooting and turnovers – are magically whisked away by this measure.  And if we look at the relationship between team wins and its points, rebounds, and assists we uncover another difficulty.  A team’s summation of points, rebounds, and assists – by itself – only explain 17% of team wins. 

Okay, the 42 Club misses some stuff so it can’t be called complete.  And it can’t explain much of wins, so it’s not accurate.  But it is simple and it does allow Simmons to remain in love with Kobe.  And it let’s Simmons maintain his belief that Iverson is great.  So that’s at least something.

Of course we have proposed a measure that is still pretty simple, much more complete, and certainly more accurate.  And Simmons can read all about it in The Wages of Wins.  But to read what we have to say, doesn’t Simmons have to fork over some cash?  Actually that’s true for most people living on the planet.  But for famous members of the media – and Bill Simmons is in that group — the book comes free.  Yep, all Simmons has to do is ask, and I will see that a book is delivered free of charge to either his home or office.  And you thought exchanging e-mails with Malcolm Gladwell was the only perk Simmons had in his job (having exchanged e-mails with Malcolm Gladwell I can confirm that this is a pretty cool perk).

With all that said I want to emphasize that I really like what Bill Simmons writes. Okay, maybe not the stuff where he takes shots at our book (I mean, can he at least say “The Wages of Wins”?).  And the 42 Club stuff was kind of silly.  But the rest of the stuff is gold.  Really.

- DJ

The Knicks yesterday bid good-bye to Larry Brown.  After a season where the Knicks spent more than any other team, and lost more than all but the Portland Trailblazers, it is expected that New York would make changes.  But did they make the right change?

There are two actors in this drama.  Brown was hired to coach the talent.  Isiah Thomas was hired to pick the talent.  Now did the Knicks lose because Brown couldn’t coach the players or because Thomas couldn’t choose the players?

Let’s start with evaluating what the Knicks did in 2004-05 and 2005-06.  If you go here you can see the Wins Produced for each player the team employed in the past two seasons.  There were seven players who played both seasons for New York.  One of these players, Stephon Marbury, clearly played worse under Larry Brown.

Had Marbury played the same number of minutes in both seasons — and maintained his Wins Produced per 48 minutes (WP4 observed in 2004-05 — the Knicks would have posted virtually the same number of wins this past season as they did in 2004-05.  So one could make the argument that the Marbury-Brown feud cost the Knicks wins.  Of course, going from 23 wins to 33 wins is nice, but the Knicks would have simply matched the Celtics in the standings.  In other words, New York would have still missed the playoffs.

And focusing on Marbury does not tell the entire story.  Other players – like Jamal Crawford, David Lee, Qyntel Woods, and Eddy Curry – played better with Larry Brown than one would expect give their past performance (of course Curry was still below average, just not as far below as we detail in the book). If we look at every player the Knicks employed in 2005-06, and forecast what we would expect the player to do given their past performance, one would have expected the Knicks to win about the same number that they did in 2005-06.  In sum, despite the problems with Marbury, Brown got as many wins from this team as one would have forecasted.

In the end, the problem doesn't seem to be with Brown’s coaching but with the player’s Isiah hired.  Consider one statistic – turnovers.  The Knicks led the league in turnovers.  How do you address that problem? The one major move Isiah made during the season was the acquisition of Steve Francis.  Francis is much like Isiah was as a player.  Francis is a scoring guard who tends to commit turnovers.  At the time Francis was acquired the Knicks led the league in turnovers per game, so Isiah had to know this was a problem.  Given this weakness, why would you acquire a player who averages nearly four turnovers per game for his career when your team problem is turnovers?

In the Wages of Wins we detail how the primary factor that determines player salary is scoring.  Turnovers and shooting efficiency are not found to impact how much a player is paid.  The Knicks are the best illustration of this tendency.  The team is led by a collection of high-priced scoring guards – Marbury, Francis, Quentin Richardson, Jamal Crawford – who achieve high scoring totals without shooting efficiently.  And these players tend to couple inefficient shooting with high turnovers.  So often a Knick possession consisted of the ball going to one of these scoring guards and then – after a turnover or a missed shot – the next player with the ball tended to be someone not recently hired by Isiah.   And this is why the Knicks finished towards the bottom of the league in offensive efficiency.  Coupled with the team’s woeful defensive efficiency, it is not a surprise the Knicks ended up with a worse record than the league's recent addition, the Charlotte Bobcats.

Perhaps Larry Brown should have said some magic words to stop this behavior.  But for many of these players, the problems seen in their performance in 2005-06 have been exhibited throughout the player’s careers.  Given the cast he was given, Brown would have had to be immensely magical to convert a team of frogs into a prince of a franchise.

Now Isiah has been given the job of coaching this collection.  Given that the primary determinant of current performance is past performance, it seems unlikely that Isiah is going to achieve dramatically different results.  Of course, if someone else could pick different talent, maybe Isiah could become the coach in New York that Brown now wishes he could have been.

- DJ

The United States – who entered the World Cup with supposedly the fifth best team in the world – has been eliminated from the World Cup.  Once again the United States has failed to find much success in the world’s most popular sport.  How can it be that a nation that so consistently succeeds in a wide variety of athletic competitions fails so miserably in soccer? Is it just because we refuse to call this sport football?

Okay, the answer goes a bit deeper than this.  I think the answer begins with the Olympics. In 2000, two economists — Andrew Bernard & Meghan Busse — wrote an article in the New York Times predicting how many Olympics medals each country would win in the 2000 Sydney games.  The economists did amazingly well.  There were 36 nations who fielded teams in Sydney who in the 1996 games won at least five medals.  Forecasting medals won for these 36 nations before the Sydney games, the Bernard-Busse model was right on target once, within one medal nine times, and only off by three medals 23 times. 

How could two economists predict the Olympics so well?  Well the answer did not lie in a detailed study of each sport and the athletes each nation sent to the games.  No, the economists simply focused on past Olympic performance, the size of each nation’s population, and each nation’s per-capita GNP. 

One might think predicting the World Cup would be similar.  At least, that is the approach taken by Goldman Sachs’ in their The World Cup and Economics 2006.  Interestingly the results indicate that soccer is not quite the same as the Olympics.  Specifically, a nation’s GNP is not related to success on the soccer field.  This is not surprising given the training cost of fielding an Olympic team, which is huge, and the training cost for producing soccer talent, which is relatively small.

Although money is not the key, the size of a nation’s population does seem to matter.  At least population explains the success of nations in Europe and Latin America.  Unfortunately, if it all it took was people the United States should dominate nations like Italy, Czechoslovakia, and Ghana.  But that is not what happened this year. 

The lack of success of the United States appears difficult to explain given it relatively large population. I think, though, that the answer can be found in The Wages of Wins.  And this is surprising, since at no point in the book do we ever talk about the World Cup.

We do talk about the role population size plays in sport.  Specifically, we note the impact the size of a population available to play a sport has on competitive balance in the sport.  Think about basketball for a moment.  Basketball probably follows soccer on the list of most popular sports in the world.  Although many people have played basketball, the population that the NBA draws upon is actually quite small.  To play in the NBA one must be extremely tall.  And as we note in the book, there is a short supply of extremely tall people.

As a consequence, NBA teams will employ very talented tall people like Shaquille O’Neal and Michael Jordan.  But because there is a short supply of tall people, teams must fill out their roster with less talented players.  The result is large differences in playing talent and a lack of competitive balance in the NBA. Applying this argument to soccer we see why the United States struggles playing the world’s game.  Although the U.S. has a large population, it does not seem to have large numbers of people who like to play soccer.   It is estimated that, in the United States, 8 million play soccer at all levels, with 3 million youngsters under 10 playing competitively [see Oliver (2000)]. 

Given that 38% of the US players are under the age of 10, something must happen along the way.  By the time they reach middle school or high school, they may be playing football - but that football involves a pigskin, pads, and touchdown passes.  And one might wonder if those who continue to play soccer in the U.S. play with the same intensity children play this sport in other nations.

With a relatively small population dedicated exclusively to playing the sport in the U.S., we expect the supply of extraordinary athletes is also small. And when a team that lacks many extraordinary players faces a nation with a larger supply of such players, the outcome is generally the less talented get to be unhappy.  Of course as soccer becomes more popular in the United States this will change.  And although it may be hard to believe after the U.S. failed to defeat Ghana, someday the U.S. will be able to compete on soccer’s – I mean football’s — biggest stage.

- Stacey

After Wade averaged nearly 35 points and 8 boards in the NBA Finals the media has now crowned him the “next Jordan.”  Unlike the previous “next Jordans” – Grant Hill, Jerry Stackhouse, Vince Carter, etc. – this “next Jordan” has now clearly led his team to an NBA Championship.  So is Flash really the next MJ? 

Let me begin by noting that in the first two games in Dallas Wade was a below average player.  Not below average for Wade.  Below average for an NBA shooting guard (average Win Score for a shooting guard is 0.128).  In Miami’s four victories, though, Wade was amazing.  It was not just his 39.3 points per game.  Or his 1.04 points scored per field goal attempt.  Wade also created 2.5 steals per contest and perhaps most importantly, grabbed more than eight rebounds per game.  In sum, Wade was indeed Jordan-like in his performance.  In these four games Wade posted a per-minute Win Score of 0.323.  Jordan himself only bested this performance twice in his playoff career.  In 1989-90 Jordan had a per-minute Win Score of 0.326 in the playoffs.  The next season he was a bit better with a per-minute Win Score of 0.348. 

A Brief Interruption: For those who do not know what Win Score is, please go here.  And please remember, this is just a simple model we created to look at performance in a game quickly. Win Score is not the same as Wins Produced, the more complex algorithm we explain in the book.  Now back to our story.

Of course, I am comparing four games of Dwyane Wade to what Jordan did across an entire playoff season. If we consider all the games Wade played in the playoffs, we see that Flash is not quite MJ yet.  Wade played in 23 playoff games.  In these games his per-minute Win Score was 0.237.  Well above average for an NBA guard, but is it Jordan-like?  In Jordan’s first three seasons his team didn’t make it out of the first round.  Over his next ten years, though, the Bulls often went quite far in the playoffs.  In these ten years Jordan had a per-minute Win Score of 0.257, besting Wade’s 2006 performance six different times.

In the regular season Jordan also was the better player.  In Wade’s first three regular seasons he has produced 36.3 wins.  Very few players produce ten wins in a season, and Wade is averaging twelve.  Jordan, though, was much, much better early in his career. Ignoring Jordan’s second season, when he didn’t play much due to an injury, we see that in MJ’s first three full seasons he produced an amazing 71.3 wins.  In other words, Wade in the regular season – although clearly above average at his position — has only been about half an MJ so far.  That is not to take anything away from Wade.  He was a key reason why the Heat took the title.  And for four games, he was very much like Jordan. Still, when we look at who MJ was on the court – game after game and season after season – we see that Flash, like everyone else, still has a way to go. So Wade is not quite Jordan.  Still, if he were, one might ask…

Who Were the Wade-anaires? (In other words, who helped Wade win Miami’s first title?)

The obvious answer is Shaq, although the numbers say he was not as much help as we have come to expect.  Okay, Shaq is still an above average center.  And if we look at the entire playoff season that is the story we would tell.  In the Finals, though, Shaq only posted an above average per-minute Win Score in Games Three and Four.  Although he rebounded well in Game Six, his below average performance from the field and the line contributed to an overall productivity level that was again below average.  For the entire post-season Shaq did post an above average Win Score – for a center average is 0.225 – of 0.247. So he did help the Heat win the title.  His performance just wasn't what it once was.  To put his 0.247 Win Score in perspective, Shaq posted a higher per-minute Win Score every year he was with the Magic and the Lakers in the playoffs.  Again, Shaq was not quite Shaq in the 2006 playoff season.

Beyond Shaq, who else played well?  In the deciding game, there was a surprise. Just a few days ago I said the following:

Now that does not mean Walker has never had an above average game in his career.  But if we look at the typical performance Walker offers, it is typically below average and thus, that is what we should expect to happen going forward. 

So we should not expect Walker to be above average, but there he was in Game Six, playing above average.  Although he missed all six three-pointers he took, his eleven rebounds were definitely helpful.  In addition to Walker, the Heat received above average performances in the deciding game from Udonis Haslem and Alonzo Mourning.  Mourning is especially notable.  In fourteen minutes he grabbed six rebounds, blocked five shots, and made three out of four field goal attempts.  On a per-minute basis he was actually the most productive player on Miami.  If Mourning could have played 45 minutes like Wade – which he probably couldn’t – and if he could have maintained the productivity we observed in 14 minutes – which he definitely couldn’t – then Mourning would have scored 26 points, grabbed 19 rebounds, blocked 16 shots, and committed 13 personal fouls.  Okay, we would need a rule change for that to work.  Still, Mourning played well in winning his first NBA title.

Enough of the Heat.  What about the Mavericks? 

Specifically, what happened to Nowitzki?

Previously I noted that Nowitzki was MIA in Miami.  That was not the case in Game Six.  Although he again had problems shooting, his fifteen rebounds definitely helped the Dallas cause.  For the game he posted a per-minute Win Score of 0.351, which is above the average NBA power forward (0.215).  No, unlike Games Three, Four, and Five, Nowitzki played well in Game Six.And he did get some help.  Josh Howard, Erick Dampier, and Marquis Daniels were all above average performers in Game Six. 

Jason Terry and Jerry Stackhouse, though, were not as helpful.  The two took 41% of the Maverick’s field goal attempts.  Despite taking so many shots, per field goal attempt these two did not perform well, only scoring 0.74 points per shot from the field.  Again, average is 0.98.  Not surprisingly, Terry and Stackhouse posted below average Win Scores. For the series, Stackhouse had one above average Game (#2), while Terry played incredibly well twice (Games One and Five).

Unfortunately, in the other games these two players were less than helpful. So the 2006 NBA Finals is in the books and what have we learned? 

1. Dwyane Wade can play really well, although he is not yet “Like Mike.”  He might be someday, though. 

2. Shaquille O’Neal looks like he is coming closer to the end of an amazing career.  That does not mean he should retire tomorrow.  But his performance in the regular season -  which was below what we normally see from Shaq – does not seem like such an anomaly any more. 

3. Dirk Nowitzki is a really good player – at least in Dallas.  Okay, that’s not fair.  Three games does not a sample make.  Still, Nowitzki did not play well in Miami.  Had he played a bit better, then we might of at least had one more NBA game to look forward to this year.

4. Final lesson: Role players matter.  In each game players other than the stars made a difference – for better or worse. And that is a point I think needs to be made again and again. The term “the Wade-anaires” is perhaps cute, but not actually fair to Flash’s teammates.  For the playoffs Shaq, Posey,and Mourning were above average performers.  And without these players, Wade’s amazing efforts might not have translated into Miami’s first championship. 

– DJ

1.            Half a Hero is better than None

At the end of Game Five Dwyane Wade was a hero.  If all you saw was the first half, though, that is not the story you’d tell.  At the half the Mavericks led 51-43.  Miami was definitely in trouble, since a loss would have forced the Heat to win two games in Dallas to take the series.  As the Pistons discovered last year, winning two games on the road to close a series is difficult.  In fact, since going to a 2-3-2 format, no team has ever accomplished this feat.

Why was the Heat down by eight at the half?  If we look at Wade, we see part of the answer.  In the first half Wade took thirteen shots and only made three.  While the average NBA players scores 0.98 points per field goal attempt, at the half Wade only had scored 0.46 per shot from the field.  Obviously when your shooting efficiency is less than half the league average, there’s a problem.  Certainly his free throw shooting and three steals helped, but with a Win Score in the negative range, it is clear that Wade’s problems hitting field goals was a key to the Heat’s deficit on the scoreboard at halftime.

A Brief Interruption: For those who do not know what Win Score is, please go here.  And please remember, this is just a simple model we created to look at performance in a game quickly. Win Score is not the same as Wins Produced, the more complex algorithm we explain in the book.  Now back to our story.

In the second half Wade – as he did in Games Three and Four – once again became “Like Mike.”  Wade scored 30 points in the second half, hitting on eight of his fifteen field goal attempts and additionally, making fourteen of seventeen free throw attempts. Beyond scoring he added four rebounds, three assists, and only one turnover.  His per-minute Win Score for the second half was 0.379, well above the average for his position.

For the game, Wade was an above average performer, but not quite as productive as he was in Games Three and Four.  Now some will say that he was great when it mattered most.  As we note in the book, such thinking is flawed.  Had Wade performed better in the first half it is likely the Heat would have won easily and the late game and overtime heroics wouldn’t have been necessary.  As coaches teach, the game is 48 minutes long.  You can’t just play part of the game.  The whole game counts in the final score and that is why for Game Five, Wade was only half a hero.

2.            Shaq at the Charity Stripe

Shaquille O’Neal entered the league with the Orlando Magic in 1992-93.  Over the next fourteen seasons Shaq produced 244.7 wins, for an average of 17.5 wins per season.  So he has been one of the most productive players since the early 1990s.  Still, there is one deficiency in his game, and in Game Five it was once again displayed. From the free throw line he has problems.  Last night he took twelve free throws and made only two.  Obviously this was bad, even by Shaq standards.

For Shaq’s career he has taken 9,744 free throws in the regular season and only made 5,147, for a conversion rate of 53%.  Clearly this is below the league average if 75%. 

What if Shaq was average from the charity stripe? At the average rate Shaq would have made 2,160 additional free throws across his career.  What would 2,160 points scored with no additional expenditure in shot attempts have meant to Shaq’s career production of wins?  Each point creates 0.0328 wins.  So 2,160 additional points – again, with no change in free throw or field goal attempts – would create 70.9 additional wins.  Or per season, Shaq as an average free throw shooter would produce five more wins.  Again, that assumes nothing else changes about Shaq’s performance. 

Producing 17.5 wins per season is excellent.  But if Shaq could hit his free throws at an average rate, he would have been worth 22.5 wins per season across his career.  To put that in perspective, in Michael Jordan’s 11 full seasons with the Bulls he produced 26.6 wins per campaign.  So Shaq would not quite be MJ if he hit his free throws, but he sure would be closer.   

3.            Did it Matter that Stack was not in the House?

Ric Bucher of ESPN.com began a column on June 14 with the following statement: “NBA commandant Stu Jackson could have very well done the Dallas Mavericks favor by suspending Jerry Stackhouse for Sunday's Game 5 of the 2006 NBA Finals.”  

In reading this sentence I thought Bucher was going to explain that the Mavericks would be better off without a player whose lack of productivity I detailed just a few days earlier.  Bucher, though, was arguing that the loss of Stackhouse could serve as motivation for the Mavericks.  Given that the Mavericks are in the NBA Finals, it is not clear why Dallas needs additional motivation.  Still, such was the argument Bucher made.

To Bucher’s credit, he did note that Stackhouse’s game has limitations. “(Stackhouse) doesn't provide anything they can't get elsewhere. Keith Van Horn is a considerably better 3-point shooter. Adrian Griffin is a better defender. Marquis Daniels is, at times, an equal slasher.”  Bucher went on to add, “Stackhouse's shot has been short and his drives a tick off.” 

Of course, Stackhouse’s shot has been off since he first entered the NBA in 1995-96.  He has never had a season where he was even close to the average NBA player in shooting efficiency. 

Did the Mavericks miss Stackhouse?  In Game Four Dallas attempted 79 field goals and from these scored 53 points.  Stackhouse contributed to this dismal efficiency by taking eighteen field goals and only making six.  In Game Five Dallas took 81 field goal attempts and from these scored 79 points.  So the Mavericks were more efficient scorers in Game Five.  Unfortunately, Stackhouse will be back in Game Six, a development that will not help this team even the series.

4.            MIA in Miami

Before Dallas fans despair the return of Stackhouse we have to remember that his deficiencies did not prevent the Mavericks from winning 60 games in the regular season.  When a team has Dirk Nowitzki – who produced about 30% of the team’s wins in the regular season — it can overcome a lack of productivity from a bench player.

Unfortunately, in Miami Dirk was MIA.  In Games One and Two Nowitski lived up to his star billing.  In Dallas Nowitzki was a more efficient scorer, turned the ball over less often, and generated more steals.  Most importantly in Dallas he averaged one rebound every three minutes, while in Miami it took Nowitzki more than five minutes to capture a missed shot.  Had Nowitzki captured one rebound every three minutes in Miami he would have grabbed nearly twenty more boards in the past three games. And Dallas would be in a much better position to win its first championship.

A few days ago Bill Simmons of ESPN.com was singing the praises of Nowitzki.  We were told that Nowitzki was the next Larry Bird.  Of course this was the same Simmons that only a few weeks ago was telling us that Kobe Bryant was the league MVP.  So the love Simmons feels for a player doesn't seem everlasting.

Although there is no more hope for Kobe this year, Nowitzki still has a chance to show us that the love Simmons has for him is not without merit.  Perhaps a return to Dallas is all Nowitzki needs.  When we look at the numbers one thing becomes certain.  The Nowitski that played in Miami is not going to bring a championship parade to Dallas in 2006.

– DJ

The series is now tied and the story is that NBA Finals have only just begun. I am not sure I buy that story.  Yes, Miami is playing better and not coincidently, the Mavericks are playing worse.  The Mavericks, though, are still in the best position to win this series. Looking at the schedule Miami still has to win a game in Dallas to take the title.  If Miami wins Game Five, then it only needs one win in Dallas.  If the Miami loses Game Five, then it has to do what the Pistons failed to do last year: Win two games on the road to close out the finals.  So the pressure is on Miami to win on Sunday.

So that is the situation looking forward. Let’s talk about Game Four. 

The big news is that Shaq finally offered a fair impersonation of Shaq.  In 30 minutes he shot 75% from the field and grabbed 13 rebounds.  After seven turnovers in Game Three, he only lost the ball three times in Game Three.  With a per-minute Win Score of 0.483, he was easily above average (the average NBA center posts a per-minute Win Score of 0.225). In sum, Shaq was back. 

A Brief Interruption: For those who do not know what Win Score is, please go here.  And please remember, this is just a simple model we created to look at performance in a game quickly. Win Score is not the same as Wins Produced, the more complex algorithm we explain in the book.  Now back to our story.

Dwyane Wade also played extremely well, although not quite as good as he was in Game Three.  Still, with a per-minute Win Score of 0.295 – average for a shooting guard is 0.128 – Wade was very good. 

But the most productive player, relative to the average at his position, was James Posey.  Posey had a per-minute Win Score of 0.500.  Average for a small forward is 0.152.  Why was Posey so productive? Posey played 26 minutes, hit on 56% of his field goal attempts, and grabbed 10 rebounds.  

Before moving on to the Mavericks one more player deserves to be mentioned.  Alonzo Mourning only played 13 minutes, but in this brief time grabbed six rebounds, blocked three shots, and had a per-minute Win Score of 0.462.  In sum, the center position for the Heat was extremely productive in Game Three. 

Now who was not productive on the Heat?  That answer to that question is everyone else.  Let me be specific.  Everyone else on the Heat was below average.  Antoine Walker was the closest to being average, but still fell short.  For those keeping score, Walker had a negative Win Score in Game One, a positive Win Score (yet still below average in Game Two), a negative Win Score in Game Three, and a positive Win Score in Game Four.  Will this pattern hold for Game Five?  Well, one shouldn’t draw inferences from a sample of four games.  If we look at Walker’s entire career, though, we should not expect him to be above average in the next game.  Now that does not mean Walker has never had an above average game in his career.  But if we look at the typical performance Walker offers, it is typically below average and thus, that is what we should expect to happen going forward. 

Of course one might ask if the Heat have a better option than Walker. Walker appears to be splitting time at small forward and power forward. At small forward the Heat have Posey.  Posey was great in Game Four, although in the regular season Posey was only slightly more productive than Walker.  At power forward the Heat have Udonis Haslem, who was much more productive than Walker in the regular season, but has played very badly in the Finals.  Given that, is it too late to make a trade? Or , can the Heat call in sick for a few weeks, draft a new power forward at the end of this month, and then finish the Finals in July? Barring those solutions, I guess the best Pat Riley can do is hope Walker or Haslem play better.  Then again, if Shaq is going to be Shaq, and Wade is going to impersonate Michael Jordan, maybe it doesn’t matter.

– DJ

This just in – the Mavericks did not play well in Game Four.  Even in a bad performance, though, someone usually plays well.  For the Mavericks, that someone was Adrian Griffin.

There was one sequence in the first quarter that I think captures the productivity story of Adrian Griffin and the Mavericks. With 3:40 left in the first quarter and the Mavericks trailing by one point, Alonzo Mourning missed a shot.  Dirk Nowitzki grabbed the defensive rebound.  The ball went to Jerry Stackhouse who missed a jumper.  Adrian Griffin grabbed the rebound.  The ball went to Jason Terry who missed a lay-up. Stackhouse grabbed the rebound but missed a three point shot.  Griffin grabbed the rebound.  The ball went to Terry who missed a three point shot.  Griffin grabbed the rebound.  Terry then missed another jumper and this time Alonzo Mourning grabbed the rebound.

Summarizing that one possession:

The Mavericks scored no points.

Stackhouse grabbed one rebound but missed two shots.

Terry missed three shots.

Griffin grabbed three rebounds.

The team did very badly, but not because of Adrian Griffin.  During this sequence ABC’s announcer Mike Breen referred to Griffin as Mr. Intangible.  Intangible means “incapable of being realized or defined.”  Offensive rebounds, though, are tangible events.  So it not clear how what Griffin did could be thought of as “intangible.”

Wait, it can be made clear.  Listening to the broadcast the focus of the announcers – and this should not surprise those who have read our book – was on scoring.  As we note in the book, scoring is often the only tangible event people notice, and hence all other actions – even those recorded in the box score – are labeled as “intangible.”

To illustrate, Hubie Brown argued that Terry and Stackhouse played well.  These two did combine for 33 points, but took 36 shots.  The pair shot 20% from three point range (2 of 10) and only scored 0.83 points per field goal attempt.  Average for an NBA player is 0.98 points per field goal attempt.  Since neither Terry nor Stackhouse offered much more to the Mavericks beyond scoring – and the scoring offered was produced inefficiently — it is not surprising that neither posted an above average Win Score.

A Brief Interruption: For those who do not know what Win Score is, please go here.  And please remember, this is just a simple model we created to look at performance in a game quickly. Win Score is not the same as Wins Produced, the more complex algorithm we explain in the book.  Now back to our story.

What about Griffin?  With 4:24 left in the first quarter Griffin entered the game.  The score was 21-18 with the Heat in the lead.  By the time the quarter ended the Heat had scored nine points and the Mavericks seven.  So with Griffin on the floor the Mavericks lost two points.  But looking at what happened, Griffin played very well.  As noted, in one possession he grabbed three offensive rebounds.  Unfortunately his teammates missed all four shots in the possession.

Should Griffin’s value be reduced because his teammates failed to convert on any of the shots Griffin provided?  Ah, the beauty of our approach.  Griffin is credited for his offensive rebounds.  The value of his teammates is reduced for their missed shots.  In a possession that yielded no points, we see that not all players on the floor were equal in value.  Griffin was very good.  Unfortunately, the Mavericks could not capitalize because on that one possession, Griffin’s teammates were not so good.

In the time Griffin was on the floor at the end of the first quarter he grabbed three rebounds, made his only shot, and committed a personal foul.  His per-minute Win Score in that time was 0.795. For the remainder of the game his per-minute Win Score was still above average.  Unfortunately the same cannot be said for his teammates.  No other Maverick was above average in Game Four.

Now this was not the case at halftime of the game.  In the first half Erick Dampier, DeSagana Diop, and even Jerry Stackhouse were all like Adrian Griffin – at least in the sense that all were above average in productivity.  The second half, though, did not go so well for any of the Mavericks (except Griffin).  And since the second half began with the team already down by ten, their performance in quarters three and four did not help.

Of course, going forward that may not matter.  If the Heat do not win Game Five, then they will have to win twice in Dallas.  Dallas knows this, Miami knows this.  So Dallas can erase the memory of Game Four by taking Sunday’s game.  If that happens, the title will probably go to Dallas.  If not, Miami still has to win one game in Dallas.  And hopefully we have not all forgotten what happened the last time the Heat tried to win a game in the Finals in Texas.

– DJ

Three games into the NBA Finals and each team has held serve on their home court.  I have looked at the data from these games and read a fair sample of the media’s coverage. What are the myths (some of which I just made up) – and corresponding measurements (which I didn't just make up) – from these first three contests?

Myth One:            Jerry Stackhouse is a “Super Sub”

Let me begin with the end of the Game Three.  With three second left Dirk Nowitzki, a career 86% free throw shooter, stepped to the line and missed a free throw that would have tied the game. Somehow Dwyane Wade, who was lined up in the third position behind Jerry Stackhouse, still ended up with the rebound.  Yes, Wade somehow moved past Stackhouse to grab the miss.  That one play captures why a player like Wade is extremely productive – a point I will touch upon momentarily — and a player like Stackhouse is well, not quite as productive.

Let’s delve a bit deeper into the game of Jerry Stackhouse.  In the second game of the NBA Finals Stackhouse led all non-starters with 19 points.  And he scored efficiently, making six of eleven shots, including four of five from beyond the arc.  After this game, Jamie Aaron of the Associated Press labeled Stackhouse the “super sub.”   Now if this was the only game you ever saw Stackhouse play you would clearly think he was a super sub.  He was an efficient scorer and in terms of Wins Score, which takes into account all of his stats, Stackhouse was an above average player. 

The problem with focusing just on Game Two is that it clearly was not a typical Stackhouse effort.  Obviously no one in the NBA typically makes 80% of his shots from beyond the arc.  And Stackhouse would not have been the one you would expect to post such numbers.  For his career he has taken 2,576 three point shots in the regular season.  He has only made 774 of these, for a conversion rate of 30%.  The average NBA player made 36% of his three point shots in 2005-06, so Stackhouse – despite his Game Two effort — is not too dependable from long distance.

In Game Three, Stackhouse reverted to form. In 31 minutes Stackhouse took nine shots.  From two point range he made one of eight, and from beyond the arc he missed the only shot he took.  For all three games in the NBA Finals Stackhouse has now taken 31 field goal attempts, and from these scored 26 points.  So per field goal attempt Stackhouse is only scoring 0.84 points, which is a bit off his career mark of 0.89.  In the 2005-06 regular season the average NBA player scored 0.98 points per field goal attempt.  In sum, despite a career scoring average of close to 20 points per game, Stackhouse has been, and remains, an inefficient scorer.

Beyond inefficient scoring, what else does Stackhouse give his team?  Remember, it was Wade who grabbed the rebound.  This is not surprising because Stackhouse, for his career, is a below average rebounder at his position.  Beyond rebounds, he is also below average in generating steals.  Where is he above average? Again, looking at his career numbers we see that he turns the ball over at a per-minute rate that is more than 50% higher than the average player at his position.  In sum, Stackhouse scores inefficiently, gives the ball away at an alarming rate, and doesn’t do much else to help his team win games. If we put it all together, it’s hard to see how Stackhouse is a “super sub.”

Myth Two:            Dwyane Wade is the next Michael Jordan

Wade’s performance in Game Three was definitely Jordanesque.  In 43 minutes he took 26 field goal attempts, converting these into 29 points.  He added 13 points from the free throw line, on admittedly below average shooting from the charity stripe.  Still, he scored 42 points, the best scoring output seen in the Finals this year.  Beyond scoring, though, Wade grabbed 13 rebounds, including the one noted earlier with three seconds left.  He also added two steals – again one of these to close the game — and only committed one turnover.  If you put all his numbers together we see that Wade had a per-minute Win Score of 0.453.  The average shooting guard posts a per-minute Win Score of 0.128.  In the playoffs Jordan posted a career average per-minute Win Score of 0.255.  So Wade was very far above the average player at his position and even bested the average Jordan playoff performance. 

And it is important to emphasize why Wade was so good.  Yes he scored and he scored efficiently, just like Jordan.  But beyond this he contributed rebounds, steals, and limited his turnovers.  And this is also just “Like Mike”.

Does all this make Wade the next Jordan?  Well in Game One Wade had a per-minute Win Score of 0.105.  In Game Two his per-minute Win Score was 0.075.  Each of these performances was below average.  Of course in the regular season Wade’s per-minute Win Score was 0.244 — that would be far above average – and more importantly, he produced 18.2 wins.  So Wade is a very good player.  Is he the next Jordan? 

Jordan produced more than 18 wins in every full season he played before 1997-98, his last season with the Bulls.  In sum, Wade is not quite Jordan. But if keeps playing like Game Three, one can someday see people wondering if so-and-so is the next Dwyane Wade.

Myth Three:          Erick Dampier is better than Shaq 

Erick Dampier, for the third consecutive game, has posted a better Win Score than Shaq.  Does this mean that Dampier is better than Shaq?  Well let’s put these three games into some perspective.  Since Dampier entered the league in 1996-97 he has produced 32.8 wins, total.  Since 1996-97 Shaq has produced 164.6 wins.  Yes, Shaq is more than a bit more productive than Dampier. 

Yet in the Finals, Dampier has offered more.  In game three Dampier played 29 minutes.  In this time he grabbed nine rebounds, created three steals, and made six of the seven field goals he attempted.  Shaq played well, but his overall production was reduced by his seven turnovers.  So much to my surprise Shaq has been outplayed by Dampier.  Given the history of these two players, though, it is not a good bet that the trends in the first three games will continue.

Yes, I expect Dampier to remain above average.  He certainly was in the regular season.  But Shaq has always been better, and it's likely that is what we will see in the next few games.

Myth Four:           Miami is Now Back in this Series    

Miami overcame a thirteen point fourth quarter deficit to win Game Three. Does this put the Heat back in contention for the NBA Title?  The road for the Heat is still very steep.  The 2-3-2 format the NBA employs for the finals appears to put the team without home-court advantage at a significant disadvantage.  Since the NBA went to this format in 1985, the team with home-court advantage took a 2-0 lead eight different times.  In all eight instances, that team – who is Dallas this year – took the series.  In fact, only the Pistons last year managed to drop the first two games on the road and still came back to force a game seven, which Detroit lost.

So can the Heat comeback? The odds are stacked against Wade, Shaq, and company.  Of course, if Wade can keep impersonating Jordan, and Shaq ever decides to impersonate Shaq, then I guess anything is possible.

– DJ

In an earlier post I noted that baseball and golf were probably my two favorite sports.  But for pure excitement and intrigue I must admit that there exists no greater month of sports for me than Fifa’s World Cup.  Maybe because it comes around only once every four years.  Maybe because nations, rather than individuals or teams, are represented.  For whatever reason, most things are put on hold during the next four weeks.

I do have one big gripe (actually, as my wife tells me, I have quite a few but that takes us off the subject). The constant diving or flopping, which seems to have become entrenched in the game, really takes away from the enjoyment.  Yes, I may be getting old, but what ever happened to sportsmanship or winning the right way? 

The only sport that seems to still run on the mantra of sportsmanship is golf.  Only a few weeks ago Darren Clarke chose to follow the code and may have cost himself a chance to win the Nissan Irish Open:

“When he returned to his ball in the rough on the ninth he reported to a referee that his lie had been improved by grass being flattened down.He was told to play the ball as it lay but, feeling he was gaining an unfair advantage if he went for the green from his improved lie, chipped sideways from the rough, as he would have had to do, he felt, with his original lie, and took bogey - CNN.”

This type of behavior does not seem to live in most other sports, where the mantra seems to be “if your not caught its not cheating.”  Now it’s easy to understand why.   Once the first person that ‘dives’ is rewarded with a free kick or penalty kick, someone else will try.  The more rewards, the more of this behavior we will see.  This is what economics predicts:  if someone comes up with an innovation that is successful, others will follow.  For example – and it is an example off the topic but near and dear to my heart — the first successful brewpub has lead to, thankfully, many more!

Now, Nike is trying to combat flopping by introducing its Joga Bonito ad campaign.  Nike hopes to introduce (re-introduce?) an honor code into the game.  While I think this is a noble attempt, its unlikely to work – the gains to this new “innovation” are simply too high.

Here is my idea.  Have the referees continue to call games the best that they can.  After the game, have a panel of three judges look at the game film.  If two of the three agree that the player dove or was faking injury, then charge the yellow or red card to that player instead of the opposing player.  This would introduce a cost for choosing to adopt the flopping “innovation” and with modern technology will not unnecessarily slow down the game.  In other words, let’s not hope people will learn to behave with honor.  Let’s give players an incentive to behave and maybe then everyone will "play beautiful."

- MBS

Once upon a time I studied international trade.  In fact, my dissertation with Terry Ozawa at Colorado State consisted of an examination of trade flows between Asia and the United States.  Although I have spent much of my career researching the economics of sports, I still remember a bit of international trade theory. 

For example, there is the Law of One Price.  This law, as defined in the textbook by Paul Krugman and Maurice Obstfeld (yes, I had to go look up the exact definition), states that “… in competitive markets free of transportation costs and official barriers to trade, identical goods sold in different countries must sell for the same price… .”

Tests of this law with international data reveal it’s not quite a law.  One might expect, though, that the same good sold on the Internet would have the same price.  Well, one might expect this, but one would be wrong. A few days ago Stacey sent me the following book price comparison from bookprice.com.  According to this site, the price of our book varies quite a bit across the Internet.  At Amazon.com our book sells for $18.87.  At Barnes&Noble.com the price rises to $23.96.  Wal-Mart is charging $26.96 while Best Prices is selling 30 copies of our work for $36.95 each.  One of my students, Ryan Barry-Souza (who, by the way, actually took an astounding seven different classes from me) tells me that on E-Bay the price for the book ranges from $16.90 to $34.20.

Same book, written by the same wonderful people, and we have very different prices.  Perhaps Best Prices is offering more than just the book for $36.95.  Maybe they are promising that if you buy our book from their website that one of the three authors will come to your house and read it to you. 

I just want to say that this is not going to happen.  No, for your money you will get the same wonderful book enjoyed by the people paying $18.87 at Amazon.com.  Now we do not object to you paying more if you like.  Still, we would hate for people to pay close to $40 only to find out later that our book could have been had at half the price.

I have two motivations in telling this story.  First, I wish to demonstrate to Professor Ozawa that I still remember some international trade.  Okay, I had to look it up, but at least I remembered where to look.

My second motivation is to help those who have not read our book. Clearly it is in our interest to reduce the population of people who have never read The Wages of Wins.

Now that I think about it, I have one more motivation. There is one specific group who have not read our book that I really want to help.  I am talking about the very tiny group of people who have felt compelled to disagree with our findings in the blogosphere despite – and yes we can tell from the comments — not reading what we actually said.  To us, this seems to be odd behavior.  How can one form an opinion on a work that you never read?

For these people, I have now lowered the cost of searching for and buying our book.  You can now clearly see the best places to buy. All you have to do now is place an order, and in a few days, find out whether you really do disagree.  We think we can change your mind, but if we can’t, we can help your opinion move from an uninformed opinion, to a well-informed opinion.  And isn’t that worth $36.95? Or do I mean $18.87?

- DJ