The Wages of Wins Journal

Entries from December 2006

The Return of the Lakers?

December 31, 2006 · 6 Comments

In 2005-06 the LA Lakers won 45 games. This past summer the Lakers made very few changes to their roster, leading one to expect very little change in the outcome we were going to observe this season. After 20 games, though, the Lakers were 14-6 and appeared to be on-pace to win more than 55 games. And then Lamar Odom was hurt. Since this injury the Lakers have been playing 0.500 basketball and are no longer keeping pace with the giants of the Western Conference. Still the Lakers do appear to be a better team. What exactly has changed?

If we look HERE we can see the Wins Produced for each player on this team after 30 games. Last season the Lakers consisted of Kobe Bryant, Lamar Odom, and not much else. This year Kobe is again leading the team in Win Produced and Odom was contributing before he got hurt. The big two, though, are now getting a bit more help.

The Improved Play of Walton and Bynum

Luke Walton entered this season with a career Wins Produced per 48 minutes [WP48] of 0.139 in limited minutes. This year, playing more than 30 minutes a contest for the first time, his WP48 has increased to 0.198. Such a result is consistent with what we report in The Wages of Wins where we found increased minutes were associated with increases in productivity, even after we controlled for past performance. Of course one still wonders if players play better with more minutes, or players demonstrate to coaches in practice that they have improved, and therefore are given more minutes. Either way, Walton is playing better this season.

Like Walton, Andrew Bynum has also improved. Bynum was taken out of high school in the 2005 draft, and if the NBA maintains its minimum age requirement might be one of the last players to ever make the jump from high school to the pros. In his first season he offered very little production in only 340 minutes of playing time. This season Bynum has already played 539 minutes and posted a WP48 of 0.206. Coupled with the improved play of Kwame Brown, the Lakers have turned the center position from one that was quite weak into a team strength.

The Lack of a Major “Star”

The Lakers now have an impressive collection of “very good” players. In addition to Bryant, Odom, Walton, Bynum, and Brown, LA is also getting above average play from Brian Cook, Ronny Turiaf, and Jordan Farmar. Such a collection is a testament to the scouting and coaching skills of the Lakers (by the way, like Don Nelson, Phil Jackson also appears to be a coach who can get more from his talent).

Despite this impressive collection, the Lakers – even if Odom was healthy – do not appear to be one par with teams like the Spurs, Mavericks, or Suns. Each of these teams has a player or players with a WP48 far in excess of 0.200. As I noted a few months ago in my discussion of the Washington Wizards, without that one player who can offer a very high level of productivity, it’s very difficult for an entire team to join the ranks of the NBA elite.

One might suspect that Kobe is that kind of player. Entering the 2006-07 season Kobe had produced 103.1 wins and posted a career WP48 of 0.196. This year he is offering 0.203. This is very, very good, but not quite what we see from players like Tim Duncan (0.292 in 2005-06), Dirk Nowitzki (0.281 in 2005-06), Steve Nash (0.319 in 2005-06), Shawn Marion (0.331 in 2005-06), Dwayne Wade (0.301 in 2005-06), and Ben Wallace (0.335 in 2005-06). Basically, each team at the top of the final standings in 2005-06 had a “star” player (or players) whose WP48 was at the very top of the league rankings.

Is Bryant a “star”? If we think about popularity, he’s most definitely a star. But if we compare what he offers on the court to players like Wade, we see that Bryant comes up a bit short. At no point in his career has Bryant offered a WP48 equal to what Wade offered last season. And without such a player, it’s simply very difficult for an entire team to produce enough to rank among the league’s elite.

- DJ

Our research on the NBA was summarized HERE.

Wins Produced and Win Score are 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

Categories: Basketball Stories

Artest for Maggette?

December 30, 2006 · No Comments

It has been reported that the Sacramento Kings and LA Clippers have discussed the trade of Ron Artest for Corey Maggette. Although it has also been reported this trade is not going to happen it doesn’t stop us from looking at who would get the better of this hypothetical deal.

Each of these players primarily plays small forward and is currently playing his eighth season in the NBA. Artest, though, is much younger than Maggette. Artest was born on November 13, 1979 while Maggette’s birthday is November 12, 1979. As you can see, Maggette is one day older so such a trade would be a bit of a youth movement for the Clippers.

In addition to being the same age, each player offers a similar level of productivity on the court. Entering this season each player has averaged about fifteen points and five rebounds per contest. On a per-minute basis Maggette is a more prolific scorer and better on the boards. Maggette is also a more efficient scorer, but a bit more prone to commit turnovers.

If we look at each player’s career wins production we again see similarities. Coming into the 2006-07 campaign Maggette had played seven seasons and produced 33.9 wins. His career Wins Produced per 48 minutes [WP48] was 0.132, which is a bit above the average mark of 0.100. Artest has also played seven seasons. Across these years he has produced 28.7 wins and has a career WP48 of 0.100. In other words, for his career Artest has been an average NBA player (given this, why is his misbehavior tolerated?).

This season Artest is playing a bit better. His WP48 after 27 games stands at 0.134. Maggette, though, has posted a WP48 of 0.144 after 29 contests. So Maggette, whether we look at this season or each player’s career, appears to be slightly more productive. These players, though, are not major stars and the differences we see are simply not large enough to dramatically impact final outcomes for either team.

So what would motivate this trade? Each team is floundering in the Western Conference. The Clippers expected to contend this season, but as noted yesterday, have not met this expectation. This trade, though, doesn’t fix the problem this team is having (which I noted yesterday was with Chris Kaman’s performance).

The Kings have a slightly better record than the Clippers, but are also not contending. Looking at the productivity of the King’s players – which one can see HERE – we see a team led by Kevin Martin. Martin’s WP48 of 0.242 actually exceeds the production of Kobe Bryant this season. That does not mean that Bryant is playing poorly, but that Martin is truly becoming one of the better guards in the game. Unfortunately he plays on a team that is going to struggle to even make the playoffs.

In the past the Kings were led in wins production by Mike Bibby and Brad Miller. Both of these players have been below average this season. Shareef Abdur-Rahim, who is capable of being above average, has also played very poorly. The Kings are getting above average performances from John Salmons, Kenny Thomas, and Artest. But it’s not enough to turn this team into a contender.

Would Maggette change this outcome? He might offer a bit more than Artest, but again, it’s not enough to significantly change where the Kings will finish in the final standings. In sum, the key for both the Kings and Clippers is to get players who produced in the past to once again contribute. Having Maggette and Artest switch places will probably do nothing to change the fortunes of either team.

- DJ

Our research on the NBA was summarized HERE.

Wins Produced and Win Score are 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

Categories: Basketball Stories

The Decline of Kevin Garnett?

December 30, 2006 · 4 Comments

In the first seven seasons of Kevin Garnett’s career he produced 108.7 wins and posted a Wins Produced per 48 minutes [WP48] of 0.262. An average NBA player only offers a WP48 of 0.100, so Garnett was clearly well above average. In the past four years, though, Garnett has gone beyond “well above average.” From 2002-03 to 2005-06, Garnett produced 118.5 wins and had a WP48 of 0.450. In each of these years he led the league in wins production. Unfortunately, the team around him offered very little. In fact, as noted in “The Tragedy of Kevin Garnett,” his teammates last year were the least productive teammates in the league.

Last year we saw a small decline in Garnett’s productivity. After three seasons where his WP48 ranged from 0.453 to 0.462, last year his WP48 “slipped” to 0.430. Again, he was the most productive player in the game. But he was not quite as good as he had been.

In 2006-07 we see a further decline in Garnett’s production. Through 28 games his WP48 stands at 0.380. Yes, he is still among the very best in the game. But he’s not quite as far above the competition as he was in years past (as noted a few days ago, Jason Kidd has a WP48 of 0.370 this season).

As a team the Timberwolves have not improved very much over last year. After Garnett the team is only receiving above average production from Marko Jaric and rookie Craig Smith. Mike James, an above average player last year, has thus far been a disappointment. And Ricky Davis, Mark Blount, and Trenton Hassell continue to offer very little. In sum, as we saw last year, this team is still Garnett and not much else.

Consequently, despite having the most productive player in the game, the Timberwolves appear destined to miss the playoffs once again.

For those wishing to see the entire analysis of the Timberwolves in 2006-07, please click HERE. As this table indicates, rookie Randy Foye has fit in quite nicely in the general theme in Minnesota. Yes, he is not offering much either.

Does all this mean that the Timberwolves should trade KG? Such a trade would almost certainly make another team very happy. But it’s hard to see anyone giving the Timberwolves equal value for Garnett. After all, he’s probably still the best player in the game.

Perhaps this is a better suggestion. Maybe the Timberwolves could trade general manager Kevin McHale. McHale is quite tall, and surely some team needs another big man in their front office.

- DJ

Our research on the NBA was summarized HERE.

Wins Produced and Win Score are 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

Categories: Basketball Stories

Why are the Clippers back to being the Clippers?

December 29, 2006 · 4 Comments

Last year was a historic season for the LA Clippers. For the first time in the history of the Clippers organization (not including the Buffalo Braves years) this team won 47 games in a single season. Furthermore the Clippers advanced to the second round of the NBA playoffs and even pushed the Phoenix Suns to seven games before bowing out. Given this success, many people thought the Clippers would contend with the likes of Phoenix, Dallas, and San Antonio this year.

So far, though, this has not come close to happening. After 28 games the Clippers have only won 12 games and are currently in 12th place in the very competitive Western Conference. What exactly happened?

In October I noted that the Clippers had done very little to improve themselves last summer. The main addition to the roster was Tim Thomas. With the Phoenix Suns last year Thomas played less than 1,300 minutes and was a bit above average. For the rest of his career, which consisted of more than 15,000 minutes, Thomas has been well below average. Which Thomas has shown up for the Clippers? Unfortunately, the 15,000 minute Thomas currently plays for the Clippers. After 28 games his Wins Produced per 48 minutes is in the negative range. So he’s not helping.

Corey Maggette, though, has apparently recovered from his injury plagued 2005-06 campaign. He is playing 26 minutes per game and is posting a WP48 that is above average. To some extent, the return of Maggette offsets the addition of Thomas.

Still, the Clippers are quite a bit worse. One might suspect that the problem is with Elton Brand or Sam Cassell. Brand had one of his better seasons (although he was a bit better in 2001-02) and one might expect that Brand has simply returned to his very good, yet not outstanding, level of productivity. Last year Brand’s WP48 was 0.274 and this year it’s only 0.218. So Brand is indeed offering a bit less.

Sam Cassell is also offering less. Cassell is now 67 years old (or at least seems to be that old). Last year he was above average, but at his advanced age one suspected that a decline was inevitable. Cassell had a WP48 of 0.128 (average is 0.100) last year and this season he is only offering 0.070. It’s important to note that although we see declines in the productivity of Brand and Cassell, this is not the primary source of the problem. Even if Brand and Cassell played as well as they did last year this team would still be on pace to win less than half their games.

Okay, if it’s not the addition of Thomas, the decline of Brand, or the aging of Cassell, what is the problem? If you look in the middle of the Clippers line-up you see Chris Kaman. Last year Kaman had a WP48 of 0.160. Close to nine of the wins the Clippers posted in their record-setting season were attributed to Kaman’s productivity. This year Kaman has missed five games. And when he has played he has posted a WP48 of 0.0004. Yes, Kaman has offered virtually nothing.

If we look at the individual stats, we see problems everywhere with Kaman’s performance. Last year Kaman scored 1.05 points per field goal attempt [(PTS-FTM)/FGA], which is above average. This year his shooting efficiency stands at 0.85, which is well below average. Kaman is also rebounding quite a bit less and has committed more turnovers. In sum, Kaman is quite a bit worse. I am not following the Clippers as close as I could (although I live in California so I guess I should), but I assume that Kaman is not completely healthy this season. Whatever the reason, though, it’s the decline in Kaman’s performance that is the one key difference between this year and last year.

So that is my explanation for the Clippers. Although I am sure my prose has been quite entertaining (okay, a little bit entertaining?), some people might prefer to just see the data. What I have been doing is putting links in my posts that take the reader to a table with each team’s Wins Produced. Judging by the comments, though, I think I am being way to subtle. At least, it doesn’t look like people know that you can see the Wins Produced for the teams I have thus far analyzed if you can find the link in the text.

Later today I will put together a post with all the tables I have constructed for each team I have analyzed in 2006-07. For the Clippers, though, you can look right HERE to see the Wins Produced for each player on the team.

As the table indicates, Shaun Livingston is still not quite the star people envision. His WP48 is just a shade below average. He is still young, so he could become better in the future. And maybe if that happens, the Clippers won’t need Kaman to return to form. Of course, if  that doesn’t happen, and Kaman keeps playing poorly, then this team will continue to flounder and Mike Dunleavy – the head coach – will probably join Mike Fratello in broadcasting.

- DJ

Categories: Basketball Stories

Is Vince Carter the Problem in New Jersey?

December 28, 2006 · 13 Comments

The Nets won 49 games last year and fully expected to dominate a very weak Atlantic Division. But after 29 games the Nets have only won 11 contests and find themselves in third place in the weakest division in the history of professional team sports. Adrian Wojnarowski of Yahoo! Sports argues that the problem is the attitude and play of Vince Carter. If we look at the team’s Wins Produced this year, though, we see a different answer.

Last year this team consisted of three stars – Carter, Jason Kidd, and Richard Jefferson – and very little else. This year Kidd is still amazing. In 2005-06 he posted a Wins Produced per 48 minutes of 0.382. This year his WP48 stands at 0.370. Carter is not quite that good, but still above average. Last year his WP48 was 0.190, this year it has dropped a bit to 0.179. Again, that’s still quite good.

The third member of the trio is Richard Jefferson, who Wojnarowski indicates has been injured. Last year he played small forward for the Nets and had a WP48 of 0.244. This year he is spending time at both forward spots and his WP48 is now in the negative range. This is a dramatic change in player performance. If we look at the individual stats we see a player who is rebounding less and shooting less efficiently.

If Jefferson was playing as well as he did last year this team would be on pace to win about 49 games. In other words, they would be about where they were last year. Given that this team did little to improve itself in the off-season, this is what we would expect. Jefferson, though, has offered considerably less this season and consequently the team’s fortunes have declined.

So here is the situation facing the decision-makers for the Nets. If Jefferson can become healthy and return to form, then the Nets can expect their team to improve in the near future. If not, then this team needs to make changes. Given the talents of Kidd and Carter this team should be contending. But when no one else on the roster helps, the talent of these two stars is just wasted.

I would note that it’s possible that the Nets can get additional production from rookies Josh Boone or Hassan Adams. Both have played well in extremely limited minutes. If these players are not the answer, it’s likely the Nets will need to make some changes to this roster.

And it better do so soon. Kidd and Carter are not getting any younger. It would be too bad if age robs these players of their talents before the Nets took the necessary steps to once again field a contender.

- DJ

Quick note to those making requests for analysis of specific teams. Requests have been made for the Wizards, Timberwolves, and Clippers. I hope to get to the Wizards tomorrow. The Timberwolves and Clippers are two teams I will look at this weekend or early next week.

Categories: Basketball Stories

Take the Grizzlies Job!!

December 28, 2006 · 1 Comment

An NBA coach is hired to be fired, a lesson Mike Fratello just learned from his good friend Jerry West. The Memphis Grizzlies were the first NBA team (and probably not the last) to fire its coach this season. Typically taking over a struggling team in mid-season is an iffy proposition. Bad teams tend to fire their coaches. But assuming the head coach position on a bad team can tend to reduce the attractiveness of your coaching resume. So at first glance, taking over for Fratello might look like a bad career move.

The Grizzlies job, though, needs a second glance. A glance at this team’s roster suggests that any coach that takes this job will probably look pretty smart.

After 30 games the Grizzlies have only won 6 games. But per game the Grizzlies are only being outscored 98.7 to 94.5, or by about four points per contest. Teams that are outscored by this amount typically finish the season with about 30 wins, or win about 10 games after 30 contests. In other words, it looks like part of the problem in Memphis has been simple bad luck. And since we expect luck to even out over time, the Grizzlies should start winning a bit more often as the season progresses.

If we look at the Wins Produced for this team, we see further reason for optimism. Pau Gasol produced 14.1 wins and posted a Wins Produced per 48 minutes [WP48] of 0.217 last year. This past summer he was hurt and consequently missed the first 22 games of the season. His WP48 thus far has only been 0.151, suggesting that he has not quite worked himself into mid-season form. As the season progresses he should play better, and again, this will lead to more victories.

Additionally, young players like Alexander Johnson, Lawrence Roberts, and Kyle Lowry have played very well in limited minutes. Roberts and Lowry have been hurt, so a new coach will have to wait for each to become healthy. But when that happens, a new coach can simply see more victories by putting these players on the floor (assuming that their per-minute productivity remains high with more minutes).

Even with all that, though, this team is still guaranteed a lottery pick in next year’s draft. A 0.200 winning percentage at this point is simply too big of a hurdle to overcome. But that is also good news for a new coach, since next year’s draft looks to be quite deep.

All that being said this team still has some problems. Eddie Jones and Jake Tsakalidis have been hurt this season and each has not played very well. Hakim Warrick has been productive for a small forward, but unfortunately has been miscast in the role of power forward. And Stromile Swift has again been a disappointment (which is the usual story when we look at his career).

If Jones and Tsakalidis can return from injury and start resembling the players we saw last year, Warrick can spend a bit more time at small forward, and the productive young players can be given more playing time, then this team can expect to see more wins. And any coach who over-sees such improvement can quickly see his star in the coaching fraternity rise.

So if you are offered the Grizzlies coaching job, take it. The job does require some travel and there is no possibility of earning tenure. Still, it does pay quite well and when you are inevitably fired, a job in the sports media (if you are reasonably good-looking) or on another coaching staff (if you are not) will be arranged.

- DJ

Categories: Basketball Stories

Taking Requests — The Dallas Mavericks

December 28, 2006 · 4 Comments

The Mavericks after 28 games.

Okapi asked yesterday for an analysis of the Mavericks this year. After looking at the data there are few surprises about this team. As expected, the Mavericks are one of the better teams in 2006-07. Last year this team was led by Dirk Nowitzki, with help primarily coming from Erick Dampier and Josh Howard. In 2006-07 the team is again led by Nowitzki, with even more help coming from Dampier and Howard.

All three of the leading players on the Mavericks have improved. This is good, since the team’s off-season moves are not paying off yet. Last summer the team lost both Marquis Daniels and Adrian Griffin, two players who were above average in 2005-06.

Speaking of Griffin, for his career he has posted a Wins Produced per 48 minutes [WP48] of 0.192. So Griffin being above average is not a surprise. For those who have read The Wages of Wins Journal since last June you might recall a post on the NBA Finals – “Thoughts After Game Four – The Dallas Mavericks Story” – extolling the virtues of Griffin. In essence, Griffin is one of those non-scorers frequently under-rated by NBA observers. Nevertheless, he does create wins for teams that bother to employ and play him.

This year the Mavericks are not employing Griffin. To replace him and Daniels the team added Greg Buckner, Austin Croshere, and Devean George. So far, none of these players that Dallas hired last summer are playing as well as they did last year. This has not been a problem thus far since the team is receiving so much from their big three. However, if the big three return to what they did last year, Dallas might have some problems.

One last comment on the Mavericks: Jerry Stackhouse, as I noted in the playoffs last year, is not a very productive player. Yes, Stackhouse can score. But he scores inefficiently. In essence – he is the anti-Griffin (which sounds like something out of a bad fairy tale).

After 28 games Stackhouse had taken 47 shots from three point range and only made twelve. That is a 26% shooting percentage from outside the arc. From two point range he is only shooting 45%, which is also below average. Stackhouse is the classic scorer who only gets points because his team let’s him take a large number of shots. And since he does little else besides taking shots, he doesn’t contribute much to wins. This story, which we have seen throughout his career, is the same story the data tells us this year.

As the season progresses we can expect the Mavericks – if the big three stay healthy – to remain among the league’s elite. Dallas may even win a title this year. But Stackhouse will not be the reason this team is successful. At least, that is how I read the data.

– DJ

Categories: Basketball Stories

Week Sixteen Quarterback Rankings

December 28, 2006 · 3 Comments

The Week Sixteen Quarterback Rankings

The Overall Rankings

The Averageness of Young and Leinart

ESPN the Magazine has listed the 2006 Rose Bowl as the top story of this past year. The game featured two quarterbacks – Vince Young and Matt Leinart — taken in the top 10 of the 2006 NFL draft. Young and Leinart are not just linked by this game. Both signal callers took over struggling teams from veteran quarterbacks soon after the 2006 NFL season began. And each quarterback proceeded to post very similar numbers.

After 15 games Young has participated in 422 plays, accumulated 2,386 yards with his arm and legs, and been charged with 13 turnovers. Leinart – whose season has now ended after being hurt last Sunday – participated in 420 plays, accumulated 2,438 yards with his arm and legs, and committed 14 turnovers.

Although the results are similar, how each player plays is a bit different. Young has rushed for 523 yards, a mark second in the league among NFL quarterbacks. Leinart only rushed for 49 yards, so he has accumulated more yards passing.

Because Young does more work with his legs, the NFL’s quarterback rating measure indicates that he is quite a bit below average as a quarterback. An average signal caller posts a rating that is close to 80. At this point in the season, though, Young’s rating is only 69.7. The NFL’s metric also rates Leinart as below average. His final rating was only 74.0, which is again below average.

As we note in The Wages of Wins, the NFL’s quarterback rating is not only immensely complicated, but also only considers what a quarterback does with his arm. Yards rushing, sacks, and fumbles are not part of the equation. QB Score includes actions with both the arms and legs, and when we consider rushing yards, sacks, and fumbles in our evaluation of Leinart and Young a different picture emerges. Both players have a QB Score per play of 1.1, which is right at the historical average we see from 1995 to 2005.

So Young and Leinart are just average. A few weeks ago Alan Schwartz wrote a Keeping Score column for the New York Times extolling the virtues of being average. In this column – which focused on baseball – was the following quote from Brian Cashman:

”Is it attractive? Oh yeah,” said Brian Cashman, the Yankees’ general manager. ”You have to have a great deal of talent to be an average major leaguer. We all like to have a roster of above-average major league players, but that’s not realistic. You’ll have a few above-average players, and you try to sprinkle the rest of the roster with as many average players as you possibly can. There’s value in their performance.”

Ned Colletti, Cashman’s counterpart with the Dodgers, echoed this sentiment: ”They may be average, but they look above average to me,” said Ned Colletti, the Dodgers’ general manager. ”They’re definitely closer to positive than negative. You try to make average your minimum, because average is pretty darned good.”

It is important to remember that average in professional sports is difficult. Half of the NFL’s quarterbacks this year have been below average. Andrew Walter has played 12 games for the Oakland Raiders this year and has not been average once. And the Detroit Lions have not had a quarterback finish the season above the average mark once this century.

Average may not be good. But it is also “not bad.” And sometimes “not bad” is pretty nice.

The Re-Birth of Jeff Garcia

Speaking of failed quarterbacks with the Lions, in 2005 the Lions employed two quarterbacks – Joey Harrington and Jeff Garcia. Harrington played in 12 games and posted a QB Score per play of 0.08 – which is a more than a bit below “not bad.” Garcia was not much better. In six games he posted a QB Score per play of 0.33.

Garcia, though, had an excuse. In the final pre-season game he broke his leg. Garcia did make it back into the Lions line-up by week seven of the 2005 campaign, but after two games once again returned to the bench to nurse his injury.

The Lions head coach, Steve Mariucci, had coached Garcia in San Francisco when Garcia was routinely going to the Pro Bowl. In the pre-season it looked like Mariucci was going to take the job from Harrington – a first round draft choice of general manager Matt Millen – and give it to Garcia. The broken leg, though, derailed this plan. Although the Lions started 4-5 – which is “not bad” relative to recent Lions history – two consecutive losses in November led Millen to fire Mariucci. In retrospect this is not looking like one of Millen’s smartest moves (insert joke here about the vast number of smart moves made by Millen). Since Mariucci departed the Lions have played 20 regular season games and only won three.

Yesterday at MLive.com, Tom Kowalski penned a column which argued events this year have vindicated Mariucci’s assessment of Garcia and Harrington. When Donovan McNabb went down it looked like the Philadelphia Eagles season was lost. But Garcia has stepped in a posted a QB Score per play of 2.22. If Garcia had enough attempts to qualify, he would currently rank 6th in the NFL.

Harrington does have enough attempts to qualify. His QB Score per play in Miami this year, though, is only 0.18. And if Cleo Lemon finishes what he starts this week, Harrington will finish the season among the five worst signal callers in the league.

If Jon Kitna, the Lions current quarterback, plays poorly against the Dallas Cowboys this week, he can join Harrington among the NFL’s very worst. Currently Kitna ranks 4th in yards gained by quarterbacks, but first in turnovers. That combination results in a QB Score per play well below average. In sum, Kitna joins the long list of Detroit signal callers that are far from “not bad.” Of course Kitna’s play has helped the Lions secure a very high pick in the 2007 NFL draft. With that pick the Lions could take another quarterback, whose career the Lions can proceed to destroy with that delightful combination of bad roster development and poor coaching.

Perhaps I should abandon my childhood team. I do live in California. Maybe I should start rooting for the Chargers. Of course that would be a new experience. How odd it would be to follow a team that has a chance to win a playoff game.

- DJ

QB Score has been discussed previously in the following posts:

Football Outsiders and QB Score

Consistent Inconsistency in Football

The Value of Player Statistics in the NFL

Simple Models of Player Performance

Categories: Football Stories

Taking Requests – The Golden State Warriors

December 27, 2006 · 10 Comments

Yesterday I noted how one can use offensive efficiency, defensive efficiency, and Wins Produced to examine how well NBA teams and players are performing this season. At the end of the post I said I would take requests. As I noted earlier today, apparently people are reading The Wages of Wins Journal and thought enough of my analysis to make such requests (which is kind of neat).

The first request came from Jeremy, who asked for more on the Golden State Warriors. Although my rule is one post per day, I don’t feel like doing much on my research today (or writing my syllabi for the next quarter). So here are some thoughts on the Warriors.

One paper I am currently working on is an examination of NBA coaching. The paper is co-authored with Mike Leeds (Temple) and Mike Mondello (Florida State) and gradually it’s nearing completion. Our tentative results thus far indicate that some coaches, although not all, appear to have a positive impact on player performance. One of these coaches is Don Nelson.

If we look at the Warriors in 2006-07 we see some evidence consistent with what our coaching research indicates. A number of Golden State players have improved. This list includes Andris Biedrins, Baron Davis, and Mickael Pietrus. Given what Golden State’s player did last year one might expect this team to win around 30 games. Yet after 30 games this year the team is on pace to win 42 contests. Is this the impact of Don Nelson?

Well, it might be. Cause and effect is difficult to pin down here. We do see Nelson coming on the scene and an up-tick in player performance. So there might be something to Nelson’s coaching.

Speaking of Nelson’s coaching methods, a few days ago Matt Buser – whose insights at Yahoo.com have proven quite useful for my fantasy basketball team – offered the following about Don Nelson’s use of a three-guard line-up:

Warriors coach Don Nelson had been using a three-guard lineup that included Baron Davis, Ellis, and Richardson as starters. But Nelson has made it clear that he won’t stick with it, saying “JR’s inability to rebound at the small forward kind of forces me to make that maybe a finishing team, not a team that I would play much.”

A few days ago I noted that Isiah Thomas’s extensive use of a three-guard line-up has cost the New York Knicks victories this season. This quote from Don Nelson illustrates why such a line-up might not help. Yes, a third scoring guard might help a team’s production of points (although it might not if the guards are inefficient scorers). But playing three guards typically hurts a team’s rebounding, and often that’s a trade-off that doesn’t help a team win games.

A final note on the Warriors: Although the team has received a boost for a number of players – including Matt Barnes who I failed to mention earlier – the team has received very little from Monta Ellis or Jason Richardson. Ellis can score, but he turns the ball over at a rate far above the average for his position. Richardson was quite productive last year, but this year has declined considerably. He has been hurt and the numbers suggest his injury has impacted his performance.

If Richardson can return to form and is able to take minutes from Ellis, this team could do even better the rest of the way. So things are looking up for Golden State. Again, is this Nelson? We don’t know for sure. But it doesn’t look like he’s hurting this team (and we are finding coaches who do).

- DJ

Categories: Basketball Stories

Reviewing the Signing of Big Ben

December 27, 2006 · 9 Comments

Yesterday I posted on how we can use offensive and defensive efficiency to evaluate a team and its players. In that post I said that I would begin offering evaluations of how teams have performed thus far in 2006-07. The first two teams I would like to analyze were involved in what I think was the biggest off-season move – Ben Wallace departing the Detroit Pistons for the Chicago Bulls. How did this move impact each franchise?

Big Ben Leaving the Motor City

Last year the Detroit Pistons finished with the best record in the NBA. The strength of the team appeared to be its defense. The Pistons only allowed 90.2 points per game, a mark that ranked 3rd in the league. Offensively the team ranked 19th, scoring only 96.8 points per contest.

Of course points scored and points surrendered, as I mentioned yesterday (and as John Hollinger and Dean Oliver have also noted), are misleading. To accurately assess a team’s offensive and defensive ability we have to consider efficiency, or points per possession. When we look at defensive efficiency we see that the Pistons in 2005-06 only allowed 1.00 points per possession, which ranked 5th in the league. Offensively Detroit scored 1.07 points per possession. Only Phoenix and Dallas were more efficient on offense. From this we see that if we compare Detroit’s offense and defense, the Pistons were actually a bit stronger on the offensive side of the game.

This year the Pistons are only allowing 92.5 points per game. This mark ranks 4th in the league and indicates that the loss of Ben Wallace did not dramatically impact this team’s defense. But if we look at defensive efficiency, we see that the Pistons this year are surrendering 1.04 points per possession, which ranks 16th in the NBA. Offensively the team is scoring 1.08 points per possession, which ranks 4th in the league. Consequently, it appears the loss of Big Ben had little impact on Detroit’s offense, but clearly reduced the effectiveness of its defense.

When we consider the Pistons offensive and defensive efficiency we see a team that should win about 50 games this season. And when we look at Wins Produced – the player evaluation metric we introduce in The Wages of Wins (which is based on a regression of wins on offensive and defensive efficiency) – we can see which players will produced these 50 wins.

The top returning player is Chauncey Billups, who produced about 16.3 wins last year and is on pace to offer 16.8 victories this season. The two prominent newcomers – Nazr Mohammed and Flip Murray – are also performing as expected. In terms of Wins Produced per 48 minutes (WP48), Mohammed is above average while Murray is far below.

Rasheed Wallace and Tayshaun Prince are each playing better in 2006-07, which might be representative of the Law of Diminishing Returns. As detailed in The Wages of Wins, when you add a productive player, one can expect other players to play somewhat worse. Likewise when a productive player departs, players should play somewhat better. One might expect in the absence of Big Ben, the remaining players might play better.

I would emphasize that it is very easy to make too much of this effect. Although the productivity of teammates has a statistical impact on a player’s performance, the size of the impact is somewhat small. In fact, when we look at Antonio McDyess, we can’t see any evidence that the departure of Big Ben is helping.

As we argue in The Wages of Wins, a number of factors impact player performance. In addition to the productivity of teammates one has to consider the player’s age, injury, coaching, and stability of a team’s roster. Above all, though, the most important determinant of what a player will do this year is what the player did last year. In other words, it does not appear that any amount of coaching or interaction with teammates will take a player like Flip Murray and turn him into a tremendously productive player.

Given what we did know about player performance in the past here is what we expected heading into this season:

  • The Pistons would not be as good without Ben Wallace
  • The Pistons would be led in wins by Chauncey Billups
  • Nazr Mohammed would be above average but not a replacement for Big Ben.
  • Flip Murray would offer very little help.

If we look at what this team had done after 26 games, all of these expectations have thus far been met.

Big Ben Goes to the Windy City

How has Ben Wallace impacted the Chicago Bulls? Last year the Bulls won 41 games. In terms of offensive and defensive efficiency one would expect Chicago to win about 54 games this year. Where will these wins come from?

Let’s begin with what we would have expected before the season started:

  • Ben Wallace would improve the team and lead it in Wins Produced.
  • After posting a WP48 of 0.200 in 2005-06, one would have expected Luol Deng to be the second most productive player.
  • Chris Duhon, Kirk Hinrich, and Andres Nocioni would be above average.
  • Ben Gordon would be far below average.

Thus far all of these expectations have been met. As noted HERE, the Bulls have been led by Wallace and Deng. Duhon, Hinrich, and Nocioni have also contributed. And Gordon continues to offer very little.

The Bulls would be even better, though, if it received more from P.J. Brown. For his career Brown has posted a 0.166 WP48. This year he has been far below average. P.J. was also a bit below average last year, so it could be the case that he is just getting old.

Brown is only 18 days younger than me, and I know my game has gone south in the last few years. At least, that is what I say now when I miss shots on the court. When I was younger I had other excuses for missing shots, such as the sun was in my eyes (excuse worked for me indoors and outdoors), my shoes are too tight, shoes are too loose, not sure my mother loves me, I am sure my mother loves me, etc…

Once again I digress. Beyond Brown’s decline I also would have expected more from Tyrus Thomas. Judging by what he did at LSU I would have thought Thomas would have been one of the better rookies on a per-minute basis. This has not happened so far, although his minutes have been quite limited.

If Brown and Thomas offer this team something, it could easily establish itself as the class of the Eastern Conference. If these players do not play better, though, the Bulls should still contend with the Pistons and Cavaliers for the best mark in the East and a trip to the NBA Finals.

Taking Requests

At the end of yesterday’s post I offered to take requests. Apparently people took this statement seriously. A-ro wanted more on the Pistons, which hopefully was satisfied by this post. Jeremy has asked for a review of the Golden State Warriors. Okapi wants to hear about the Dallas Mavericks. And Damo listed 12 different teams, which is a bit much. Perhaps I should have emphasized one request per person.

I was already planning on commenting on the Pistons and Bulls today, so A-ro’s request was satisfied first. But technically Jeremy made his request first, so the next team I will comment on will be the Golden State Warriors. Then I will look at the Mavs. And of Damo can narrow his requests to one team, I will take that team after the Mavs.

Of course, I still need to post the Week 16 Quarterback rankings. Hmmm…. is it just me or are these posts getting longer and more numerous? Perhaps my co-authors could finish their Christmas vacations and offer a bit help (hint, hint).

- DJ

Categories: Basketball Stories