The Milwaukee Bucks finished with a 50-32 mark in Don Nelson’s final season as head coach.  That season marked the 7^th consecutive season the Bucks finished with at least 50 wins.  One suspects that Nelson takes some credit for this record.  And although Wins Produced mostly credits Sidney Moncrief for Milwaukee’s success in the 1980s (a point made a few days ago), Nelson’s perspective is bolstered by the fact that Bucks have only have only reached 50 wins once in the 22 years since he left town. 

This past season was consistent with the post-Nelson era.  The Bucks finished with only 34 wins, a record that ranked 11^th in the Eastern Conference.  The team’s efficiency (offensive efficiency minus defensive efficiency), though, was -1.1; a mark that tied the Pacers for 9^th in the conference.  Translating this number into Wins Produced shows us that the Bucks should have expected to win 38 games, or just one game less than the Detroit Pistons.  In sum, the Bucks were nearly a playoff team last season.

Turning to the individual players – reported in Table One – one can see who was responsible for these wins.  Leading the way was Ramon Sessions, who finished the 2008-09 season with 9.0 Wins Produced and a 0.198 WP48 [Wins Produced per 48 minutes].  Sessions was not the only above average talent.  An average player posts a 0.100 WP48; and Luc Mbah a Moute, Andrew Bogut, Luke Ridnour, and Michael Redd all played more than 500 minutes and posted a WP48 mark that was above par.

Table One: Milwaukee Bucks in 2008-09

Bogut and Redd, though, played fewer than half the team’s games.  One suspects that if Bogut and Redd had been available the entire year, Milwaukee could have been a playoff team.  No, the Bucks may not have won 50 games.  But certainly they could have challenged the Hawks – a team with only 44.9 Wins Produced last year (or just seven more than the injured Bucks) — for the fourth seed in the East.

Of course, the injuries did happen and the Bucks missed the playoffs. In the off-season the team’s two leading scorers (in terms of total points scored) – Richard Jefferson and Charlie Villanueva – went elsewhere.  And now Ramon Sessions has signed an offer sheet with the Minnesota Timberwolves.  If Sessions leaves, a significant number of Wins Produced also departs.  Given these departures, we shouldn’t be surprised that the experts at ESPN expect the Bucks to finish the 2009-10 season ranked 14^th (out of 15 teams) in the Eastern Conference.

Then again, the Bucks haven’t just lost players.  A few new faces have also come to town.  And some of the new faces have actually been productive players in the past.  In fact, when we actually look at the team’s current depth chart (a depth chart from ESPN.com that ignores Hakim Warrick) there’s reason for hope in Milwaukee.

First String

PG: Luke Ridnour, 0.106 WP48

SG: Michael Redd, 0.133 WP48

SF: Luc Mbah a Moute, 0.116 WP48 [at PF; 0.194 WP48 if he played SF]

PF: Hakim Warrick, 0.082 WP48

C: Andrew Bogut, 0.204 WP48

Second String

PG: Brandon Jennings (rookie)

SG: Charlie Bell, 0.037 WP48

SF: Carlos Delfino, 0.180 WP48 [at SG in 2007-08; 0.141 if he played SF]

PF: Kurt Thomas, 0.191 WP48

C: Francisco Elson, 0.011 WP48

Of these ten players, six are above average performers.  If these players maintain this production, then the Bucks will not be one of these worst teams in the Eastern Conference in 2009-10.  In fact, it’s possible this team will challenge the Hawks and Wizards for the fourth seed.  In other words, the Bucks could actually reach the second round of the playoffs (where they will probably get blown out by the Cavaliers, Magic, or Celtics).

Before Milwaukee fans get too excited, there are a few bumps on the road to this vision. For all this to happen…

• Bogut and Redd need to be healthy. 
• Joe Alexander – the team’s lottery choice in 2008 – needs to stay on the bench (or get much better).
• Unless Jennings can come in and be above average as a rookie (not a common occurrence), Ridnour has to continue as the team’s starting point guard.
• Delfino has to play and be productive as a small forward (where ESPN.com currently lists him on the depth chart).

In sum, there are some issues that need to be resolved.  But if these issues are resolved the Bucks will be better than expected.  And if that happens, Scott Skiles – the team’s head coach – might be Coach of the Year; an honor not given to a Milwaukee coach since Don Nelson coached Sidney Moncrief. 

Let me close with another persepctive on the Bucks. Ty Willihnganz of Bucks Diary [mvn.com/bucksdiary] also expects Milwaukee to be in playoff contention in 2009-10.  Here is how he summarizes Milwaukee’s prospects:

All in all, I think 40 wins is the STARTING POINT as far as what to expect from the Bucks in 2009-10, based upon each player’s career norms.  And remember, NO ONE has ever accused me of being in the tank for the Bucks.  If anything, I bend over in an attempt to err on the side of caution and pessimism.  The bottom line is, I think, that the people who are projecting the Bucks among the worst in the NBA are Know-Nothing Pinheads who, like most analysts, wildly overvalue the contributions made by below average to average players like Charlie Villanueva and Richard Jefferson.  The truth is, there is simply no credible evidence to support the notion of a dramatic Bucks collapse… in fact, everything points to improvement (Skiles is loving it… he’s going to look like the overachieving hero).  So cheer up Bucks fans.  We are nowhere near championship ground yet, but the team ought to show improvement, ought to play respectable defense, and frankly, ought to make the playoffs in 2009-10.   Get your 10 packs ordered!

- 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.

Although it may not be obvious, most columns posted here have a coherent theme.  Today, though, I thought I would just comment on a few stories I have seen this past week.

Immigration Policy and Competitive Balance

Chris Cook of Financial Times makes an interesting connection – see Top clubs triumph on an uneven pitch — between immigration policy in Great Britain and competitive balance in the English Premier League.   Specifically, Cook argues that immigration restrictions in England have harmful effects in soccer.  The argument draws upon research we noted in The Wages of Wins. 

The Cook article also mentions the new book by Simon Kuper and Stefan Szymanski. The book – Soccernomics – will be released in the U.S. in October.  It’s already a big hit in England (where it was released last month).

Attractive Quarterbacks Again

About a year ago I published a short article in Play Magazine (from the New York Times) detailing how the physical attractiveness of a quarterback was related to his pay.  Research I conducted – with Jennifer VanGilder, Rob Simmons, and Lisle O’Neill – suggested that the symmetry of a quarterback’s face (as noted in our academic paper, we use a measure of attractiveness from Symmeter) has a statistically significant link to a quarterback’s compensation. 

This story has once again been noted in the media.  Reed Albergotti of the Wall Street Journal asked us to update our analysis. These updates are noted in Cute Quarterbacks? There’s a Stat for That, Too. Katie Leslie of Atlanta Journal Constitution also comments on the story (see Matt Ryan: The NFL’s face of ‘perfection’?).

My Childhood with Ernie Harwell

This past week we all learned that Ernie Harwell has incurable cancer.  Harwell was the Tigers announcer for about forty years and his voice is a fond memory from my childhood.  About the first twelve years of my life were spent in Michigan, primarily in the city of Detroit.  When I was four or five years old, the family’s black and white television set broke.  My father decided not to replace the TV, so for the next seven years we had no TV (yes, it was a tragedy).  So as my attention turned to sports, it was via the radio that I primarily followed my favorite teams.  This means that I spent quite a bit of time listening – like “the house by the side of the road” (here is a list of Harwell’s catchphrases) – to Ernie Harwell.  Although I enjoy listening  — via the Internet – to Jim Price and Dan Dickerson today, when I think of the Detroit Tigers on the radio I will always think of Harwell (and of course, Paul Carey).   So I was very sorry to hear about his illness.

Forecasting the Lions

And now for some more sad news (although not nearly as serious).

Next week the Detroit Lions will begin their 46^th season with William Clay Ford as sole owner of the team (and 49^th since Ford became team president).    Prior to Ford taking over, the Lions won NFL titles in 1935, 1952, 1953, and 1957.  The team came to Detroit in 1934, so in 27 years without Ford involved the Lions won four titles.  With Ford in charge, the team has won only one playoff game and last year finished 0-16.

A new season, though, is upon us and in football, hope springs eternal before the games begin.   In 2007 the Falcons and Dolphins combined to win five games.  Last season these two teams – led by new coaches — each finished with eleven wins and made the playoffs. 

The Lions enter this season with a new coach and coaching staff.  In addition, of the 25 players listed on the team’s offense at the moment, thirteen were not on the team last year (including five new starters). On defense there are even more new faces.  Of the 25 players on defense, sixteen are new to Detroit.   And at least eight of the eleven starters were not with the Lions last year.

Despite all the new faces, forecasts for the Lions seem quite low.  No one seems to think this team has any chance of being good in 2009.  It’s almost as if people think the clothes worn in Detroit lose games.  After all, the coaches have changed.  Many of the players have changed.  One would think, then, that at least some of the forecasts would change also.  But that hasn’t happened.  Every forecast I have seen expects the Lions to be losers in 2009.

At this point I would like to say that there is some statistical model that contradicts the conventional wisdom.  But it doesn’t seem to me that any such model exists.  In basketball we have data on every player that is somewhat consistent from season to season.  Even with such data, surprises still happen.  In football, no such data exists for players.  Consequently, forecasting football seems quite difficult.

Brian Burke – of Advanced NFL Statistics – has recently had some fun with this point.  His Koko Fantasy Ranking highlights how difficult it is to construct a winner in fantasy football.

Phil Taylor at Sports Illustrated also comes clean on this point.  His column in the latest issue of Sports Illustrated captures the truth of NFL Forecasts: Trust Us: We’re Wrong.

The words of Burke and Taylor should give some hope to fans of the Lions.  There is a possibility the Lions will be good this year.  How much of a possibility, though, is unclear to me.  And I suspect, everyone else as well.

- DJ

The WoW Journal Comments Policy

Andre Miller came to the Philadelphia 76ers –via the Allen Iverson trade — in December of 2006.  At the time of the trade, the Sixers were 5-19. With “the Answer” departing Philadelphia, the future looked bleak in Philadelphia. Across the remainder of the season, though, the Sixers were 30-28 (a mark predicted in this forum).

The next season – as E. James Beale noted in the Philadelphia City Paper – the media expected the Sixers to be very bad.  After all, Iverson had left.  Sure the team managed to play 0.500 ball without Iverson.  But such evidence, prior to the 2007-08 season, was ignored.

When the 2007-08 season ended, though, the Sixers were once again average (40-42 final record).  Yes, that isn’t good.  But it isn’t “historically bad” (as Beale quoted people saying) either.

During the summer of 2008 the Sixers added Elton Brand.  In the past, Brand has been very good.  But he only played 29 games last year and consequently didn’t make much of a contribution.  Despite Brand’s inability to contribute, though, the Sixers were once again average (record of 41-41).

The Impact of Miller

When we look back at the last three editions of the Sixers, we see the same three names leading the team in Wins Produced.  In 2006-07, 27.9 of the team’s 35 wins could be traced to Andre Iguodola, Samuel Dalembert, and Andre Miller.  In 2007-08, 30.4 of the team’s 40 win were linked to these players.  And last year – as Table One indicates – 31.1 of the team’s 41 wins were linked to Iguodala, Dalembert, and Miller. 

Table One: The Philadelphia 76ers in 2008-09

The rest of the roster, though, only produced 10 wins last season.  And a similar story is told for 2006-07 and 2007-08.  Across the past three seasons, players not named Iguodala, Dalembert, and Miller have only produced 26.6 wins for the Sixers (or about nine wins per season).

 Miller will be 33 years old before the next season ends.  So he is rapidly approaching “ancient” for an NBA guard.  Given his age, the Sixers decided to allow him to depart for Portland.  Assuming Miller’s production doesn’t decline dramatically, this move really helps Portland.  But Philadelphia has a problem.  Eleven wins have left the team, and if this team is going to maintain its “average” status someone is going to have step-up and produce.

Forecasting Philly

The experts at ESPN apparently believe the Sixers can easily replace Miller.  This week the consensus forecasted 39 wins for Philadelphia.  In other words, losing Miller only costs this team about two wins. Oddly enough, these same experts think adding Miller will cost Portland two victories. Obviously, just as we saw when he joined the Sixers, Andre Miller is under-valued by the experts (a perspective that is probably due to the fact Miller has only averaged 14.0 points per game in his career).  Nevertheless, past history suggests Miller does help quite a bit and the Sixers will need to replace his contribution if Philadelphia is going to return to the playoffs.

To see how they can do this, let’s consider the current depth chart in Philadelphia (stats from 2008-09; WP48 = Wins Produced per 48 minutes):

First String

PG: Jrue Holiday (rookie)

SG: Willie Green [-0.3 Wins Produced, -0.008 WP48]

SF: Andre Iguodala [13.1 Wins Produced, 0.193 WP48]

PF: Elton Brand [1.5 Wins Produced, 0.081 WP48]

C: Samuel Dalembert [6.9 Wins Produced, 0.162 WP48]

Second String

PG: Royal Ivey [-0.7 Wins Produced, -0.037 WP48]

SG: Louis Williams [1.8 Wins Produced, 0.046 WP48]

SF: Jason Kapono [-1.7 Wins Produced, -0.045 WP48]

PF: Thaddeus Young [2.2 Wins Produced, 0.040 WP48]

C: Marreese Speights [0.6 Wins Produced, 0.023 WP48]

The Sixers also have Jason Smith [-1.5 Wins Produced, -0.067 WP48 as a rookie in 2007-08] and Primoz Brezec [career -0.002 WP48, but who posted a 0.088 WP48 in 2004-05].

When we look at the entire roster we only see two players – Iguodala and Dalembert – who were above average last season.  And these players only produced 20 wins last year.  For this team to get to average status, someone else is going to have to produce 20 more wins.  The remaining veterans on the team, though, only produced 3.5 wins last season.  And after Elton Brand (who we will discuss in a moment) the highest WP48 from last year was the 0.046 mark posted by Louis Williams.  Such numbers tells us that this roster doesn’t have an abundance of productive problems. 

It does, though, have Elton Brand. From 2001-02 to 2006-07, Brand averaged 14.5 Wins Produced per season. In 2007, though, Brand was hurt.  Since this injury, Brand has only played 37 games and produced 1.8 wins.  And he is now 30 years old.  So although it is possible Brand can return to form, there is evidence that this won’t happen. But if Brand does return to form, the Sixers can get back to “average” status. 

After Brand, the picture does look bleak.  Again, most of the players on the roster have never been productive NBA players.  It’s possible that some might look to the one player without prior NBA experience.  The player ESPN.com lists as the starting point guard – Jrue Holiday – is a rookie.  Of the 47 players drafted out of a college last year, Holiday was ranked 30^th in Position Adjusted Win Score per 40 minutes (PAWS40).  Right after Holiday was selected, the Denver Nuggets chose Ty Lawson; who was ranked 3^rd in PAWS40.   Those numbers suggest the Sixers made the wrong choice, and Philadelphia is not going to Miller-like production from the point guard spot.  Of course, that’s just a suggestion. College numbers do not project perfectly to the NBA (although there is a correlation).

Best Case vs. Worst Case

Okay, so what can fans of the Sixers expect?  The Best Case scenario is that Brand returns to what we saw two years ago and Iguodala and Dalembert remain productive (and maybe Holiday is not as bad as his college numbers suggest).  If that happens, the Sixers will get about 35 wins from their top three players and the team might make the playoffs.  The Worst Case scenario, though, is that Brand doesn’t return to form.  Given what’s left on this roster, the Sixers will then struggle to get past 30 wins. 

If the worst case scenario occurs, I suspect some people will blame Eddie Jordan (the team’s new coach).  And I suspect others will argue that the team misses Allen Iverson.  For the first argument I would note that coaches do not generally change player performance in the NBA.  As for the second argument we can talk about Wins Produced.  But also consider the following:  The Sixers were 111-111 with Andre Miller.  From 2003-04 to 2005-06 (or the last three full seasons Iverson played in Philadelphia), the Sixers were 114-132.  Yes, the Sixers did reach the NBA Finals with Iverson.  But the team’s record with the Answer (throughout his career in Philly) was actually below 0.500 (really, it was).  So Iverson – as Wins Produced suggests – was never really “the Answer”.  

- 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.

Stephen Jackson stated the following a few days ago: “It’s just things are in the air right now. I really can’t get too much into it right now, but I’m just looking to go somewhere where I can go and win a championship.”

Translation: Jackson doesn’t think the Golden State Warriors are very good and he wants to be traded to a contender.

Evaluating Jackson

Jackson averaged 20.7 points per game last year, a mark that led Golden State.  If all one cared about was scoring, then Jackson would be considered a very good player.

Of course, there are other aspects to the game of basketball.  If we look at Jackson’s stats (at Basketball-Reference.com) a few deficiencies stand out.  An average small forward has a 48.2% adjusted (or effective) field goal percentage. Jackson’s career mark is 47.5% and last year he only shot 46.6%.  With respect to rebounds, net possessions (rebounds + steals – turnovers), and blocked shots, Jackson is also below average.  Yes, he can get assists and steals.  But his deficiencies overwhelm his few advantages.

As a consequence, although he has played 19,470 minutes in his career he has only produced 15.1 wins.  His career WP48 is 0.037, well below the average mark of 0.100.  In fact, in nine seasons Jackson has yet to be above average.

Last season the Warriors won 29 games.  The team’s Wins Produced of 31.2 indicates Golden State was a slightly better than their record indicated.  Looking at the players – reported in Table One – we can see who was responsible for these wins.  Or in the case of Jackson, who was not really responsible. Last year Jackson was paid $6.6 million and only produced 2.4 wins.

Table One: Golden State Warriors in 2008-09

Beyond a relatively low level of productivity, Jackson is also old and expensive.  Before the next season ends, Jackson will be 31 years of age.  He is also scheduled to receive $35 million across the next four seasons.

Jackson on a Better Team 

When we consider the entire Jackson picture, it seems a contender wouldn’t be helped much if they acquired his services.  It’s also apparent that the Warriors shouldn’t care much if he departs.  As it stands now, Jackson is going to get $10 million dollars in 2013-14.  Given the age profile of NBA players, at that point he will be a very unproductive 35-year old player.  So if the Warriors do find a contender that wants Jackson, now is the time to make a deal. 

Let me close by noting the other argument made by Jackson.  As noted a few weeks ago,  the Warriors could improve next season. In other words, it’s possible that Jackson will get to play on a better team if he just stays where he is.  Jackson is not likely to be the major reason the team improves, but he can still enjoy the additional wins.

Of course, I made this statement before the Warriors signed Mikki Moore.  As I have noted in the past, Moore is not a very productive NBA player. And if Moore takes minutes from Andris Biedrins, Anthony Randolph, or Brandan Wright; Golden State’s attempt to turn this team around will not be helped.  Nevertheless – despite the signing of Moore and the comments of Jackson – the Warriors could be much better next season.  All it would take is a few different decisions (and yes, that does make it sound far easier than it is).

- 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.