Explaining the Disappointing Wizards

Each time someone links to the Wages of Wins Journal I am notified by Word Press. A few days ago I noticed a link to the following entry at Ted’s Take:

A Really Great NBA Blog

Check this one out.

That was the entire entry. 

Ted Leonsis and the Wizards

The Ted of Ted’s Take is Ted Leonsis.  Here is just a part of his lengthy bio (which one can read in its entirety at Ted’s Take).

Ted is also the founder, chairman and majority owner of Lincoln Holdings LLC, a sports and entertainment company that holds ownership rights in several Washington, DC entities including 100% of the NHL’s Washington Capitals and the WNBA’s Washington Mystics. Lincoln Holdings also owns approximately 44% of Washington Sports and Entertainment Limited Partnership (WSELP), which owns the NBA’s Washington Wizards, DC’s Verizon Center and the Baltimore-Washington Ticketmaster franchise.

So Ted Leonsis is owner of the Washington Capitals and a part-owner of the Washington Wizards.  Of these two, the Capitals are probably making Leonsis quite a bit happier these days.  And if he read what I said about the Wizards last summer, he is probably even more disappointed with the Wizards today.

Explaining the Disappointment

Last season the Washington Wizards were the worst team in the Eastern Conference.  This season the Wizards are 10-20 – and with a -3.6 efficiency differential (offensive efficiency minus defensive efficiency) – Washington is on pace to win about 32 games in 2009-10.  Despite this projected 13 game improvement, though, fans of this team are probably unhappy.

Again, such unhappiness stems from last summer’s evaluation of this team.  Last August, Chris Mannix of SportsIllustrated.cnn.com argued that the Washington Wizards were the 8^th best team in the NBA.  And I argued that this team could win between 45 and 50 games.  Clearly Mannix and I don’t look to be correct.  The Wizards don’t look like a top 10 team and a winning record in 2009-10 seems unlikely.

In evaluating this team’s problems, some might focus on the play of Gilbert Arenas.  Agent Zero is simply not the same player we saw in 2006-07 (the last season he was fully healthy).   A quick glance at his stats from Basketball-Reference (which reports his stats per-36 minutes) reveals that Arenas today – relative to what we saw in 2006-07 – has improved with respect to assists and is essentially the same with respect to shooting efficiency from the field, shot attempts from the field, rebounds, blocked shots, and personal fouls.   He has declined, though, with respect to steals, turnovers, free throw attempts, and free throw percentage. 

When we convert these numbers into Wins Produced per 48 minutes [WP48] we see that Arenas is posting a 0.069 mark this season.  Had his performance returned to what we saw in 2006-07, though, his WP48 would be 0.148.  Translating this into Wins Produced – and projecting across the entire 82 game season (as reported in Table One) – we see the decline in Arenas’  performance is costing the Wizards about five wins in 2009-10.

Table One: The Washington Wizards after 30 games in 2009-10

In other words,  even if Arenas was performing as well as he did before his recent health problems, the Wizards would still be below average.  So the declines in the performance of Agent Zero can’t explain the difference between where the Wizards are today and my evaluation of this team this past summer.

To see this difference we need to look at the second column in Table Two, or how many minutes each player has played.   Of the players listed, the top player in WP48 in 2008-09 was Mike Miller.  Thus far this season, though, Miller has only played nine games.  Had he been played the entire season – and maintained what he did last year in Minnesota – Miller would currently be on pace to produce 12.2 wins this season.  Such production would increase the Wizards season projection by 8.6 wins.  And coupled with Arenas returning to form, transforms the Wizards into a 45 win team.

The Wizards are currently 2.5 games out of the playoffs, and again, fans of this team might be a bit disappointed.  If Arenas and Miller were playing and producing, though, this team would currently be the 6^th seed in the Eastern Conference.  And this means fans of this team could expect to make the playoffs….where they would probably be eliminated in the first round.

Repeating What Was Said Three Years Ago

Okay, even if the players for the Wizards were healthy and performing as expected, this is still not a title contender.  A “good” team, yes.  But a championship parade is not to be expected.

To understand the bigger problem in Washington one has to return to what I said about this team in October of 2006.  Three years ago I argued that the Wizards had assembled a collection of above average talents.  But because the team lacked a superstar – defined as a player who produced well beyond the 0.200 WP48 mark (i.e. approaches the 0.300 level) – the Wizards ceiling was limited.

What I said back in 2006 appears to apply today.  The team still has a collection of above average players.  But again, really no superstar. Yes, Miller can post a mark above 0.200. But his career mark entering this season was only 0.163 and he has had trouble staying on the court in 2009-10.  So Miller can help, but by himself he is not going to transform this team into a title contender.

Again, what the Wizards need is a player – or better yet players — who can stay on the court and post a WP48 mark that approaches 0.300.  Until such players are acquired, one suspects fans of the Wizards — and their owner — will continue to be a bit disappointed. 

Let me close by note that looking back on what I said about the Wizards in 2006 led me to think about the history of this forum.  This blog began in April of 2006.  Across the past three years (actually, we are getting close to four), more than 1,000 entries have been posted (we hit the 1,000 mark with the post on the Phoenix Suns from December 14).  Each post tends to be about 1,000 words, so this means 1,000,000 words have probably been offered in this forum.  To put that in perspective, the Wages of Wins was about 120,000 words (and our next book is a bit shorter). This means about eight books have been posted in this forum (and that mark doesn’t even consider the more than 15,500 comments that have been offered). 

As the year and decade ends I want to once again thank everyone (and that includes both new readers like Ted Leonsis and others who have been coming here since 2006) who makes this forum a part of their day.  And although the thought of writing eight books makes me tired, as long as people keep stopping by it looks like this forum is going to be continuing into the next year and decade.   

- 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.

A Costanza Trade for Joe Dumars

Patrick Hayes at MLive.com (my favorites site for Detroit sports) points our attention to a trade suggestion from Bill Simmons:

I know that ESPN’s Bill Simmons columns are meant for entertainment and that his trade suggestions have no real basis in reality. ….His latest effort looks at teams that, because of the economy, need to make trades and shed salary. While he doesn’t include the Pistons in the “desperate to shed payroll” group, he does believe they need to make a move:

Tayshaun and DaJuan Summers (cap throw-in) to Boston for the Tony Allen/Brian Scalabrine/J.R. Giddens expirings plus Big Baby and $3 million. Imagine the Celts tossing out a defensive quintet of Pierce, Rondo, Prince (if healthy), KG and Perkins? Now that’s a championship quintet! Worth the risk, I say.

Hayes was not impressed by this suggestion.  Simmons, though, is a fan of the Celtics. And from Boston’s perspective, this is a pretty good deal. Unfortunately, I think Simmons is channeling his inner Costanza.

Defining the “Costanza Trade” 

For those who are not fans of Seinfeld (all two of you)… in 1996 Seinfeld had an episode where George Costanza is considered a candidate for the job of assistant general manager with the New York Yankees.  Such a job would give George input into possible trades (rather like the job Simmons imagines he should have in his columns). As George thinks about this job he imagines some trade scenarios.  Here is an example of George’s thinking:

I think I got it. How ’bout this? How ’bout this? We trade Jim Leyritz and Bernie Williams, for Barry Bonds, huh? Whadda ya think? That way you have Griffey and Bonds, in the same outfield! Now you got a team! Ha ha ha.

Essentially, George – as a Yankee employee and fan – supports trades where the Yankees clearly get the better end of the deal.  Certainly one can imagine the Yankees easily acquiring Bonds and Ken Griffey today. But in 1996, it’s hard to see how the trades Costanza envisions happening (and that’s why this is funny).

This is essentially the same approach offered by Simmons. The players in this proposed trade really haven’t played much this year (none of these players has played 200 minutes this year).  But here is what these players did last year [WP48 = Wins Produced per 48 minutes]:

Glen Davis: -2.5 Wins Produced, -0.074 WP48

Tony Allen: 1.6 Wins Produced, 0.088 WP48

Brian Scalabrine: -1.1 Wins Produced, -0.108 WP48

Tayshaun Prince: 7.8 Wins Produced, 0.122 WP48

An average NBA player posts a 0.100 WP48.  Of these four players (Summers is a rookie this year and Giddens only played eight minutes last year), only Prince is above par.  And Davis and Scalabrine are in the negative range. Again, what Simmons proposes is a “Costanza trade.” One can just hear Simmons as he writes his proposal: “I think the Celtics can acquire Tayshaun Prince and not really give up much at all.  Imagine the Celtics with Prince. Now you got a team. Ha ha ha.”

The Odd Moves of Joe Dumars

Clearly if such a trade were offered, Joe Dumars (the Pistons GM) should say no.  Then again, Dumars has said “yes” to some very suspicious moves lately.  In the draft the Pistons selected Austin Daye and Jonas Jerebko.  Both players are above average as rookies, so one can’t argue much with either selection. Sandwiched between these two picks, though, Joe Dumars selected a player named DaJuan.  When I heard “DaJuan” on draft day I fully expected to hear the word “Blair” next.  But much to my disappointment, the next word was “Summers”.  As the following table reveals, DaJuan Summers was the least productive player selected out of college in 2009.  And DeJuan Blair – selected two slots later by the San Antonio Spurs – was the most productive.

Table One: Ranking the Players Selected from College in the 2009 NBA Draft

Thus far this season the college numbers seem quite prophetic.  Summers has posted a -0.141 WP48.  Meanwhile, Blair is posting a 0.296 WP48 and has now moved into the starting line-up for the San Antonio Spurs. Yes, I know.  Blair had some injury concerns. But Blair has already produced 2.8 wins in his career.  Given what we saw from Summers in college and briefly in the NBA, there’s a real possibility that Summers will never produce 2.8 wins in his entire NBA career.

After missing on the NBA draft, Dumars then turned to the free agent market.  Blessed with an abundance of cap space, Dumars made a quick splash by signing both Ben Gordon and Charlie Villanueva.  As I noted when these signings were announced, the words “above average” are not often mentioned when it comes to these two players. Certainly both player sport career marks that are below par.  And as Table Two notes, neither player has been above average after 30 games in Detroit.

Table Two: The Detroit Pistons after 30 games in 2009-10

Gordon and Villanueva are not the only members of the “below average” club in Detroit.  Currently, only two players – Ben Wallace and Jerebko – are above average.  Consequently, we shouldn’t be surprised that this team is on pace to win only 30 games this year.

The poor performance of the Pistons in 2009-10 could also have been expected given what these players did last year.  As Table Two indicates, if the Pistons’ players maintained what they did in 2008-09 this season, Detroit would only be on pace to win 35 games. Yes, Detroit has suffered some injuries.  And if everyone was healthy this team would be better than a 30 win team. But even if Tayshaun Prince, Rip Hamilton, and Ben Gordon had been available the entire season the Pistons would still not be a serious contender in the Eastern Conference.

In sum, Detroit – the team I follow – is simply not very good.  And it’s not very good because Joe Dumars has recently been choosing below average players.

Given such behavior, maybe Dumars would agree to the Simmons proposal.  Again, Dumars didn’t know that the word “Blair” comes after “DeJuan.” So maybe he would agree to a classic “Costanza trade.” And Pistons fans could then start crying about Big Baby.

- 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.

Correlation, Causation, and Jamal Crawford

Ed Carson of Investors. com reported the following trend a few weeks ago (HT Freakonomics):

The best kept secret of the past 20 years has been this: When the Los Angeles Lakers won the NBA championship, the market would almost always fall that year. When the Lakers lost, the market would usually rise. The Laker Indicator only steered investors wrong in three years during the entire span, and not once from 1995-2007.

An investor who put down $1,000 into the Nasdaq at the start of 1987 and stayed fully invested through 2007 would have ended up with $7,604. But an investor who bought the Nasdaq in years the Lakers lost and stayed in cash when the Lakers won would have finished with $21,189. This strategy would have kept you in the market during the 1990s bull market, avoided the 2000-2002 bear and then got back in as the market uptrend resumed.

Such numbers speak for themselves. In fact, I can’t imagine anyone looking at Carson’s numbers and coming to another conclusion.

Okay, if you read a little further in the column you see that Carson disagrees with Carson’s numbers.  Here is Carson again: Correlation does not always mean causation. Psychological ‘secondary’ gauges may appear to work for a time, then suddenly stop. And it’s easy to look for excuses, a la the Lakers in 2008, for why your special indicator really still works.

Focusing on Jamal Crawford

The Carson story about the Lakers came to mind as I watch a bit of basketball on Christmas day.  While watching I heard (and I didn’t write this down so this may not be exact) Hubie Brown briefly discuss the contenders in the East. After noting Boston, Cleveland, and Orlando, Brown noted that people shouldn’t ignore the Atlanta Hawks.

So far I agreed with Brown.  The Hawks currently rank second in the NBA in efficiency differential (offensive efficiency minus defensive efficiency).  So any discussion of NBA contenders has to include Atlanta.

But after noting that Atlanta is a contender, Brown decided to offer an explanation for Atlanta’s surge.  And that explanation focused on Jamal Crawford.  Essentially Brown argued that the off-season addition of Crawford is the reason why Atlanta has improved. 

I should note that I am probably being unfair to Brown. Yes, I did hear him say this. But in the course of a game – where the talking heads have a job to do (i.e. keep talking) – one is bound to find something said that doesn’t make complete sense or that the person talking wouldn’t take back (or at least explain better).  That being said, though, I want to proceed as if Brown really meant what he said (i.e. Jamal Crawford is the reason why the Hawks have gotten better).  And I want to proceed in this fashion because Brown’s focus on Crawford illustrates a larger point about correlation and causation.

Explaining Atlanta’s Surge

Let me start by offering a defense for Brown’s comment.  It’s easy to see why someone might focus on Crawford.  Of the eight players who have logged at least 400 minutes for the Hawks this season, Crawford is the only one who was not with Atlanta last year.  Last season the Hawks won 47 games without Crawford.  This year the team’s efficiency differential suggests Atlanta will win more than 60 games.  Such evidence seems fairly self explanatory.  Atlanta only has one new player getting any minutes, so that one new player must be the reason why the team is much better.

Once again, though, correlation doesn’t prove causation.  Just because Atlanta has done better since it acquired Crawford, it doesn’t mean Crawford is the reason why the team has improved. 

Back in November I offered an explanation for why the Hawks have gotten better.  That explanation centered on the play of Josh Smith.  Now that Atlanta has played 29 games it seems like a good idea to re-check the numbers.

We can find Josh Smith’s box score numbers at Basketball-Reference.com.  A check of these numbers reveals – just as we saw in November – that Smith has improved with respect to shooting efficiency, rebounds, assists, steals, and turnovers.  His per game scoring is down, but with respect to the statistics that drive wins (both theoretically and empirically), Smith has improved tremendously.

We can see this improvement when we turn to Wins Produced.

Table One: The Atlanta Hawks after 29 games in 2009-10

Table One reports the Wins Produced for each player Atlanta has employed this year.  And it reports what we could have expected had each player maintained his performance from last season.  The numbers from last season indicate the Hawks should be on pace to win 46 games in 2009-10.  Again, that’s about what this team did in 2008-09.  So given the team’s roster moves – and what the players on this roster did last year — Atlanta shouldn’t be any better.

But Atlanta clearly is better.  When we look at how performance has changed, we see that the primary reason for this improvement is the play of Josh Smith.  Had Smith maintained his 2008-09 productivity level this season, he would be on pace to produce 6.2 wins. Instead, his improved box score numbers translate into more than ten additional wins.  In sum, both the empirical evidence found in the box score numbers – and how these numbers theoretically and empirically are linked to wins – indicates it’s not the addition of Crawford that has transformed the Hawks (Crawford – as I noted last April – has generally been a below average shooting guard in his career and he is once again below par in 2009-10).  The key is really the improved play of Josh Smith (Al Horford and Joe Johnson have also helped some).

Correlation Stories

Let’s say, though, you really wanted to stick with the Crawford story.  One could argue that somehow Crawford’s presence has led Smith to hit more shots, grab more rebounds and steals, and commit fewer turnovers. Such a story is tempting, especially if you begin your analysis with a correlation.

And this is a great example for why we tend not to begin our analysis with a data-mining search for correlations.  Once you find a correlation it’s too easy to start inventing theories.  A better approach – and the approach we teach our students – is that good empirical analysis begins with some sort of theoretical model, and then moves on to the data.  It’s only through theory that we can actually argue any causation at all. Or in other words, without a theory all we have is a correlation. 

The Lakers-stock market story and the Crawford-Hawks stories are good examples of correlations in search of a theory. Other examples in basketball can be seen whenever you see people argue that when player X is added to team Y, team Y does better (or worse).  Again, one can show that such correlations exist.  But without a theoretical structure, it’s often hard to believe that a causal relationship has actually been uncovered.  And without a causal relationship, you really don’t have much of a story.

- 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.

A Quick Note on the Trail Blazers Rash of Injuries

Yesterday I noted that the Rockets losing their top three scorers from 2008-09 did not appear to dramatically impact the team’s fortunes this year.  Yes, the Rockets are not quite as good as last year.  But the loss of Yao Ming, Tracy McGrady, and Ron Artest is only expected to cost the Rockets about seven wins (across the entire season). The Portland Trail Blazers are now experiencing a similar problem with injuries.  But for Portland, the effects might be more devastating. 

The following above average players have already missed – or will miss – extensive playing time this season (average Wins Produced per 48 minutes — or WP48 — is 0.100):

Nicolas Batum: 0.123 WP48 in 2008-09

Greg Oden: 0.154 WP48 in 2008-09

Rudy Fernandez: 0.167 WP48 in 2008-09

Joel Przybilla: 0.288 WP48 in 2008-09

Last season this quartet produced 26.6 wins for a team that won 54 games.  So losing these players will not help.

Of course, if other productive players can be inserted into the line-up – as the Rockets have demonstrated – the Blazers can still win games.  Unfortunately – as Table One demonstrates – Portland is running out of productive players.

Table One: The Portland Trail Blazers after 30 games in 2009-10

Across the first 30 games in 2009-10, the following players have posted WP48 marks that exceed the 0.100 mark of an average player:

Greg Oden [0.327 WP48]

Rudy Fernandez [0.196 WP48]

Joel Przybilla: [0.188 WP48]

Brandon Roy [0.159 WP48]

LaMarcus Aldridge [0.108 WP48]

As Basketball-Reference.com notes, Oden and Przybilla will probably miss the rest of the season.  And Fernandez is out for 4-6 weeks.  This leaves the Blazers with Roy and Aldridge.

Now if we look at last year’s performance, both Steve Blake and Andre Miller were above average.  Miller, though, is old (he was old last year, but now – as you can guess – he is even older).  And as I noted today, basketball is really a game for young people.  

All of this suggests the Blazers experience with injuries will be different from what we have seen in Houston.  This rash of injuries – coupled with a lack of productive players to take the spot of the departed players – means that Portland really has a problem.  So although Table One projects 50 wins for this team, the injury problem means this projection is going to be lower and lower as the season progresses.  And in the tough Western Conference, this might mean a trip to the lottery is a possibility for a team many felt (okay, maybe just me) thought could contend for in the Western Conference this season.

What is interesting about these two scenarios is that the injuries for Houston happened to scorers while the injuries for Portland are generally to players who don’t score.  Conventional basketball wisdom would suggest the loss of scorers will hurt a lot while the loss of non-scorers can be overcome (by other players “stepping up”).  Unfortunately for fans of Portland, though, I think the real story is the ability or inability of a team to replace productive players with other productive performers.  Whether the players lost are scorers or non-scorers really doesn’t matter much.

Okay, that is my second quick note today.  Now I need to go wrap presents (or something equally Christmassy). Hope everyone has a Happy Holidays.

- 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.

A Quick Note on Aging in the NBA

David Biderman of the Wall Street Journal has a short story today on the aging of NBA players.  The article notes that the peak age – or the point where player stop improving – is around 24 or 25 years of age.  Biderman’s source for this story is some economist from Southern Utah University.

The research Biderman sites will be discussed in Stumbling on Wins (which should be shipped to bookstores in February or early March).  Since everyone doesn’t have access to our book just yet (although we hope that changes for everyone – and we mean everyone – soon), let me offer a few observations.

Via a study of NBA players from 1977-78 to 2007-78 (a study discussed in more detail in the book), we found that an NBA player generally improves until he is in his mid-20s.  Performance after this point is not much different until a player reaches about 27 or 28 years of age.  After that point – and especially when a player passes the age of 30 – performance starts to decline more noticeably.

It’s important to note…

• we are reporting a tendency.  The peak at 24 or 25 will not be true for every player.  But when you look at the link between age and performance, controlling for a host of other factors, the general peak is in this range.
• the results were the same when we looked at NBA Efficiency.  So this result does not depend on looking at performance via Wins Produced.
• the key issue is not the specific point in the player’s 20s where the peak occurs, but rather that performance after age 30 has a noticeable drop-off.  In the player’s twenties the slope downward is quite gradual (and not something you would probably notice if you watched the player).  In other words, LeBron will still be LeBron – barring injury – for a few more years.

Let me also add that the drop-off after age 30 will not be the same for everyone.  For some players, performance declines considerably (as my post on Kareem and Shaq noted a few days ago).  However, John Stockton posted a WP48 of 0.262 at the age of 40 (Stockton’s best season, though, was at the age of 25).

And one last note…JC Bradbury had a very interesting article on this subject in the context of baseball. One issue Bradbury emphasized is that more athletic activities (like tennis, short distance running, and swimming) tend to see peak performances at a very young age.  In a sport like golf – and with respect to some aspects of baseball – peak performance occurs much later.  Basketball is a sport that relies tremendously on athletic ability, so we shouldn’t be surprised to see a peak in the mid-twenties (as opposed to a point closer to 30 years of age).

Again, we have more on this in our next book (which you can already  pre-order at Amazon.com).

-  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.

Lawson Regrets and the Amazing Tyreke Evans

From a statistical perspective – as Table One illustrates — Ty Lawson was the most productive point guard drafted out of college last summer. 

Table One: Ranking the Players Selected from College in the 2009 NBA Draft

Despite his college production, though, Lawson lasted until the 18^th pick in the draft.  Five point guards – or at least, players who can play point guard – were selected before Lawson went to the Denver Nuggets.  And now that were about one-third of the way through the 2009-10 season it seems like a good idea to look at the early returns on these point guards (well, it may not be a good idea but I am going to do it anyway).

The Early Returns

Before we look at the Wins Produced – and WP48 [Wins Produced per 48 minutes] – numbers, we should note that it’s very early in these player’s careers.  Young players do get better, so players who don’t look good today might look better later on (of course, players who don’t look good today might not look better later on). 

With this caveat in mind, let’s look at the numbers.  The following lists where each point guard was selected, how many minutes he has played (as of Monday night), his WP48 at the point guard position, and his projected wins.  The latter is calculated by simply assuming the proportion of minutes assigned each player at this point will be the same the rest of the year. Since this proportion will probably change, these projected wins should be thought of as rough estimates (and given trends in performance, perhaps very rough estimates).

4. Tyreke Evans [945 minutes, 0.126 WP48, 9.3 Projected Wins]

6. Jonny Flynn [820 minutes, -0.026 WP48, -1.3 Projected Wins]

7. Stephen Curry [841 minutes, 0.076 WP48, 4.7 Projected Wins]

10. Brandon Jennings [928 minutes, 0.101 WP48, 8.5 Projected Wins]

17. Jrue Holiday [343 minutes, -0.044 WP48, -0.9 Projected Wins]

18. Ty Lawson [576 minutes, 0.152 WP48, 5.3 Projected Wins]

An average rookie posts a WP48 mark of 0.042.  So Evans, Curry, Jennings, and Lawson are above average.  In fact, Evans, Jennings, and Lawson are above average when compared to all players (average WP48 is 0.100).  If we look back at Table One we will see that Evans, Flynn, and Holiday were below average in college while Curry and Lawson were above par.  Thus far, four of these five players are maintaining their position relative to an average player at his position.

So it looks like a few teams who passed on Lawson might begin to feel some regrets.  In terms of WP48, Lawson tops this entire sample.  Evans and Jennings, though, have been above average thus far.  So it’s likely the Kings (who took Evans) and the Bucks (who took Jennings) are looking not looking at envy at Lawson (as noted in a moment, this is especially true for Evans).  Even the Warriors (who took Curry), are probably are somewhat happy with their choice.

The Timberwolves and 76ers, though, would probably be better off with Lawson.  This was the story before the draft.  And it looks like this story can be told today.

Jennings and Evans go in Opposite Directions

Let me close by noting the play of Jennings and Evans.  A few weeks ago I noted that Jennings had posted a 0.181 WP48 after 15 games.  His mark today – when we factor in his time at shooting guard – is 0.111.  This means that Jennings has posted a 0.026 WP48 across the past 11 games. As we go forward it will be interesting to see which player is the “true” Jennings.  Will Jennings return to what we saw earlier in the season? Or are the comparisons to Allen Iverson accurate?

While Jennings has been getting worse, Evans has been getting much better.  After nine games, Evans had a mark of 0.058.  When you factor in his play at shooting guard, his current mark is 0.155.  This means Evans has posted a 0.205 mark across the past 17 games.  In sum, Evans looks like the early favorite for Rookie of the Year (Lawson is probably not going to play enough to seriously contend for this award).  At least, the amazing Tyreke Evans looks like the early favorite for ROY for guards (I haven’t looked at other positions yet). 

-  DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

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

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

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

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

The Rockets Surprise

Before the season started we learned that Yao Ming wasn’t going to play.  Furthermore, Ron Artest had departed for LA and the availability of Tracy McGrady was questionable.  These three players led the Rockets in points per game in 2008-09.  With so much scoring exiting the building, many NBA observers thought Houston was destined for the 2010 lottery.

In looking over Houston’s roster, though, it didn’t appear this team was quite as bad as people thought.  Entering this season the Rockets roster included the following players who were above average performers in 2008-09:  Trevor Ariza, Shane Battier, Chuck Hayes, Kyle Lowry, and Carl Landry.   Although these players were not scorers, their respective WP48 [Wins Produced per 48 minutes] numbers suggested the cupboard wasn’t bare in Houston.  In fact – although the loss of Ming and McGrady was going to hurt some (notice I left out Artest) – it seemed that the playoffs were still a possibility for the Rockets.

The Rockets Today

The Rockets have now played 27 games in 2009-10.  And if the season ended today, the Rockets would not have a seat at the NBA lottery.  So how was this possible?

Table One: The Houston Rockets after 27 Games in 2009-10

Table One reports the Wins Produced and WP48 numbers for the Rockets this season.  As one can see, the following players have so far been above average (average is 0.100): Luis Scola, Carl Landry, Kyle Lowry, Trevor Ariza, and Chuck Hayes.  Yes, five of the six players who were above average last season are still above average this season. 

Again, the loss of Ming and McGrady (the latter only recently returned) didn’t help.  But the Rockets are still on pace to win about 46 games this season (after winning 53 last season).  So apparently losing your top scorers is not necessarily a death sentence.

The performance of the Rockets this season demonstrates an aspect of basketball performance often noted in this forum.  Basketball players – relative to what we see in baseball and football – are very consistent over time.  For the most part, the productivity of the players the Rockets are employing this season is not much different from what we saw last year. Yes, the loss of the team’s primary scorers has forced other players to take shots.  In general, though, the increase in shot attempts hasn’t reduced each player’s effectiveness.

The Ariza Story

As a few commentators on the Miami Heat post from a few days ago noted, though, the exception is Trevor Ariza (how we went from a discussion of Dwyane Wade to Ariza, though, is a mystery to me). Last season Ariza took 14.1 field goal attempts per 48 minutes (FGA48) and his adjusted field goal percentage was 51.1%.  This season, Ariza’s FGA48 has increased to 20.4 and his adjusted field goal percentage has fallen to 45.8%.  And this has led some people to argue that Ariza is the classic example of how increasing shot attempts lowers efficiency.

In reading the comments I am somewhat convinced I cannot change everyone’s mind about the meaning behind Ariza’s numbers.  In fact, one commentator explicitly stated: “If a statistical study suggests otherwise, it must be missing something.”

Such a comment forces me to lower my expectations.  The following comments on Ariza’s declining field goal percentage, as I note, will not necessarily change minds (but they might give “true believers” something else to rationalize away).  Here are some things to think about when you consider Ariza’s drop in efficiency:

• Ariza’s career adjusted field goal percentage is 48.4%.  So his mark this year is not far from this career mark. 
• Ariza, though, has dropped off.  Looking at other players on Houston’s roster, though, reveals a different story.  Carl Landry is taking 5.6 additional field goals per 48 minutes and his shooting efficiency has only declined from 57.5% to 57.0%.  Chuck Hayes has seen his FGA48 rise by 4.9 and his shooting efficiency has improved by 7.7%. 
• The discussion of Ariza, Landry, and Hayes is purely anecdotal.  When you look at players from 1977-78 to 2007-08 (the sample includes over 5,000 season observations) you see that there is a link between a change in the number of shots he takes and his shooting efficiency.  But the impact is quite small.  Here is what we say in our next book:  “…imagine a player who takes 16.3 shots per 48 minutes and has an adjusted field goal percentage of 48.4% (these are the league average marks).  If that player increased his shots per 48 minutes to 25.3 (a two standard deviation increase), his adjusted field goal percentage would be expected to decline to 47.1%.”

Given all this, is the change we see in Ariza’s shooting efficiency simply due to the fact he is being asked to take more shots?  I don’t think the evidence leads to that conclusion.  We don’t see the same story when we look at the other players on the Rockets (who are also taking more shots).  And we don’t see such a strong link between shot attempts and shooting efficiency when we look at a sample of over 5,000 NBA players.   

All that being said, I don’t have a great story for why Ariza’s efficiency has declined.  I will note that although NBA players are very consistent from season to season, shooting efficiency is one aspect of a player’s performance that is the most volatile (as we note in the next book, it’s about as volatile as OPS and Slugging Percentage in baseball). 

The Bigger Story

Regardless of how you see the Ariza story, the primary observation remains.  Most players on the Rockets are playing about as well as we would expect given their past performance.  Yes, many of these players are playing more minutes.  And many are taking more shots.  But their overall effectiveness is roughly the same.

Now it’s important to remember that losing Ming and McGrady didn’t help.  And although the Rockets would be in the playoffs if the post-season started today, they are only barely in.  There’s still a chance Houston will be visiting the lottery.  That being said, what we have seen so far does suggest that losing your top scorers doesn’t necessarily kill an NBA team.  At least, that’s what we see when a roster has an abundance of productive non-scorers. 

- 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.

Boston or LA?

With 30% of the regular season complete, the Boston Celtics lead the Eastern Conference with a 10.3 efficiency differential (offensive efficiency minus defensive efficiency).  And the LA Lakers lead the Western Conference with an 8.3 mark.  Although it’s still early, such numbers suggest that for the 12^th time in NBA history the Celtics and Lakers will meet in the NBA Finals. 

Again, it’s early.  But if these numbers hold up people will be asking the following until next summer: Who is better, Boston or LA?

The Celtic Argument

Let’s start with a look at the Boston Celtics after 24 games.  Table One reports each player’s Wins Produced and WP48 [Wins Produced per 48 minutes] this season, as well as what Boston could have expected given what these players did last season.  As one can see, given what these players did last season Boston should be on pace to win 72 games this year.  After 24 games, though, the pace is only 66.4 victories.

Table One: The Boston Celtics after 24 games in 2009-10

When we look at the performance of the individual players we see that Rasheed Wallace, Ray Allen, and Eddie House have declined the most.  What do these players have in common?  All three are on the wrong side of 30 years of age.  And that illustrates the problem facing the Celtics. Six of the Celtics are more than 30 years of age and these players have played 60% of the team’s minutes. And of these six, only Paul Pierce is offering more this season (relative to last season).

Although basketball players may like to think they age like fine wine, the general pattern is that players age like milk.  So as the season progresses, the Celtics might slip some more.  At this point in time, though, the Celtics are a very good team.  And this is because of the performance of Rajon Rondo, Kevin Garnett, Pierce, Kendrick Perkins, Ray Allen, and Shelden Williams. These six players are on pace to produce nearly 68 wins this year.

There are two surprises on this list.  First, Shelden Williams has been very productive and should (but may not) make fans of the Celtics forget about Glenn Davis.  The other surprise is Rasheed Wallace, who is really not offering much at all. Once again, given his age this shouldn’t be a surprise. 

The Lakers Argument

Okay, the Celtics are good, but not quite as good as their past numbers suggest.  The story of the Lakers is the opposite.

Table Two: The LA Lakers after 24 games in 2009-10

As Table Two illustrates, what the Lakers did last season suggests that this team should only be on pace to win 52 games this year.  When we look at production this season, though, the Lakers are on pace to win 63 games. 

Both of these numbers are deflated because of the early season injury to Pau Gasol.  In projecting wins I am taking the easy way out.  Projections are simply what has happened across 24 games multiplied by 82/24.  So for Gasol, his projected minutes are only 1,596 (which essentially assumes Gasol will keep missing eleven out of each 24 game segment). Currently, though, Gasol is averaging 36 minutes per game.  At this pace, Gasol will play about 2,550 minutes this season.  Given what he did last season [WP48 of 0.256], projections based on last year’s number would increase by 5.5 wins (so the Lakers should have expected about 57 wins). 

After 13 games, though, Gasol has posted a 0.465 WP48.  Such a mark bests – by a wide margin – anything Gasol has done in his career.  And it also tops anything anyone did last season.  If Gasol can keep this up, the Lakers can expect to win about 72 games this season.

But can Gasol maintain this pace?  Again, he has never produced at this level before.  When we look at the individual numbers we see that Gasol is posting career highs with respect to free throw percentage and rebounds.  As noted in the past, rebounding tends to be very consistent across time.  So one wonders if Gasol will keep grabbing 16.9 rebounds per 48 minutes (his previous high was 13.1 in 2006-07). 

If this happens, Gasol and the Lakers will probably finish with a better record than the Celtics.  And the subsequent home-court advantage – assuming the playoffs hold to form – will give the Lakers an advantage in the NBA Finals. 

Then Again…

But if Gasol slips, then the Celtics will have the advantage.  Then again, if the aged Celtics keep aging, maybe the Lakers will be better.  Of course, Kobe is both aged and hurt.  So maybe Boston will still finish with the best record.

Okay, here is what we know. Right now, the season numbers favor the Celtics.  But there is evidence that the Lakers are better right now and could finish with better season numbers.

And all of this ignores the other teams in the NBA. I still believe it’s possible that the Cleveland Cavaliers can come back (although I recognize the distinct possibility that won’t happen).  And the Atlanta Hawks and Josh Smith continue to be amazing (9.0 efficiency differential).

In sum, it’s still early (have I said this?).  But if you are looking at the Celtics and Lakers, we can clearly see that the Celtics are better.  Or is it the Lakers?  Or… how ‘bout that Pau Gasol?

- 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.

Pointing Fingers at the Miami Heat

The Miami Heat began the season with a three game winning streak.  After losing a game, the Heat won another three in a row.  Since that 6-1 start, though, the team has only won six of seventeen games.  After one of their latest losses – a 28 point defeat at the hands of the Memphis Grizzlies – the Heat began searching for answers. (HT True Hoop)

“I didn’t say a word. I let the guys talk. Sometimes, as a leader, you have to listen,” (Dwyane) Wade said of a postgame powwow led by veterans Jermaine O’Neal and Udonis Haslem. “I won’t say what any guy said. Just know that there was communication back and forth.”

Haslem insisted that there was no finger-pointing or animosity, although there was plenty of blame to share for the disappointing play at home. Instead, Haslem’s message was on all-inclusive accountability.

Sunday was for soul-searching.

“It’s not just about one person or two people,” Haslem said. “It’s on all of us. We definitely have to dig deep and find out what type of team we are.

“We need to get that chip back on our shoulder we had earlier.”

The words of Haslem suggest that if the Heat simply change their attitude, or try just a bit harder, the Heat will become a winning team.  Furthermore, this is a team issue.  Blame cannot be assigned to any one player or person. The numbers, though, seem to tell a different story. 

Pointing at the Supporting Cast

Last year Miami was led by Dwyane Wade.  Of the team’s 43 wins, about 22 were linked to the productivity of Wade.  This means that without Wade, the Heat were not a very good team.

This year – as Table One illustrates – the Heat are once again led by Wade.  And after Wade, Miami – once again — doesn’t have much.  The Heat are currently on pace to win about 40 games this year (the team’s efficiency differential is -0.7). 

Table One: The Miami Heat after 24 games in 2009-10

Of these 40 wins, about 26 of these projected wins can be linked to players not nicknamed Flash.  About four of these wins can be linked to the play of Dorell Wright.  Last year, Wright only played 73 minutes.  So his production (which is quite similar per 48 minutes to what he did in 2007-08) is somewhat surprising.  In fact, Wright and Wade are the only players to surpass the WP48 [Wins Produced per 48 minutes] mark of 0.200 (twice the average mark).  And only Udonis Haslem and Quentin Richardson (of those who have played 100 minutes) surpass the average mark.  Yes, much of this roster is below average.

One player who comes close to average – although still falls short – is Jermaine O’Neal.  And like the play of Wright, this is also somewhat surprising. Last year O’Neal posted a -0.037 WP48.  This year he’s still below average, but his ability to approach the average mark – something he hasn’t done since 2006-07 — is worth about four additional wins to the Heat across an entire season (or about six more wins than this team could have expected O’Neal to produce). 

Without the play of Wright and O’Neal the Heat would not be as close to the 0.500 mark. So clearly one can point a finger at whoever assembled Wade’s supporting cast.  But one can also point the finger of accusation at Wade. 

Pointing at Flash

In 2008-09, Wade posted a WP48 [Wins Produced per 48 minutes] mark in excess of 0.300.  This season, Wade’s WP48 is 0.207.   In other words, last year Wade’s productivity went beyond what we see from Kobe Bryant [WP48 of 0.244 in 2008-09).  This season, Wade’s production lags behind Kobe.  And for the Heat to be a team that goes well beyond 41 wins (at least, more than one or two games beyond), Flash really has to do more than Kobe.

This year, though, that’s not happening.  And the big reason why Flash has dimmed is Wade’s shooting efficiency.  Last season Wade posted a 51.6% adjusted field goal percentage.  This season this mark has fallen to 44.8%.  If Wade could return to the level of efficiency seen last year – and the same happened with respect to assists (which are also a bit down) – the Heat would see the team’s projected wins rise by about nine.   

One should note, though, that even if Wade returned to form – and Wright and O’Neal kept doing what we have seen thus far –the Heat would probably fall short of 50 victories. And that mark is not quite what Miami envisioned when it started 6-1. 

The problem for Miami is that this start was a bit of an illusion.  As Dean Oliver noted in Basketball on Paper, there is a 90% chance that a team that ultimately wins 30% of its games – or about 25 games – will win three in a row at some point in an NBA season.  So when the Heat – a team that will probably win 50% of its games – started with a three game winning streak (and followed it with another three game winning streak after a loss); the expectations for the Heat shouldn’t have changed immediately.  At least, we shouldn’t be surprised that Miami has fallen short of the promise of these early streaks.

Pointing All the Fingers

At the end of the day, Miami doesn’t have much beyond Wade.  So a finger of blame needs to be pointed at the person (or persons) who assembled this roster.  And as noted, a second finger can be pointed at Wade himself. The Heat would not be one of the top teams in the league if Wade reverted to form.  But they could be a bigger threat to surpass the mark of an average team and perhaps come closer to 50 wins.

Unfortunately for Miami, “approaching 50 wins” is probably the ceiling for this team; and struggling to stay above .500 is closer to reality.  One suspects that this will not be enough for Miami to keep Wade when the season ends.  And if Wade does depart, we can expect the finger of blame to be frequently pointed next summer in Miami.

- 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.

How Much Has Phoenix Improved and Is Amare An All-Star?

If the NBA playoffs began today the Phoenix Suns would currently have the 4^th seed – and home-court advantage in the first round — in the Western Conference.  The Cleveland Cavaliers – in the Eastern Conference – also hold the 4^th seed.  Why is this comparison important?

Last year the Cavaliers were the best team in the NBA (at least, in the regular season) and the Phoenix Suns missed the NBA playoffs.  This past summer the Suns sent Shaquille O’Neal to the Cleveland Cavaliers.  And after one-quarter of the 2009-10 season, it appears that Shaq’s departure has made the Suns better and caused the Cavaliers to decline.

Shaq’s impact on the Cavs was discussed last week (and it ain’t a pretty picture).  Today we are going to examine the surprising Suns.

Are the Suns Rising?

The Suns have won 16 of their first 24 games.  Such a record, though, is somewhat deceiving.  When we turn to efficiency differential (offensive efficiency minus defensive efficiency) we see a mark (after Saturday’s games) of 2.85.  This mark ranks 7^th in the West, just ahead of the Utah Jazz (2.70) and the Houston Rockets (2.15).  So a focus on efficiency differential reveals that the Suns are not far from slipping into the 9^th spot (and out of the playoffs). 

A differential of 2.85 is consistent with a team that wins 48 games across an 82 game season.  Last season the Suns posted a differential of 1.95, a mark consistent with a 46 win team (the number of wins the team had last season).  So the Suns have not really improved much relative to last year.

Not much, though, is still something.  And something has happened without Shaq.  When we look at Wins Produced – reported in Table One — we can see where the improvement has come from. 

Table One: The Phoenix Suns after 24 games in 2009-10

Table One reports what each player has done for the Suns this season.  It also reports each player’s expected production, given what the player did last year.  A quick glance at the numbers reveals that the two players who have improved the most are Steve Nash and Channing Frye.

A different picture emerges, though, if we consider what Nash and Frye did two years ago.   If Nash and Frye were repeating what they did in 2007-08, each player’s production would be as follows:

Steve Nash [2oo7-08 production]: 0.276 WP48, 15.7 Projected Wins

Channing Frye [2007-08 production]: 0.043 WP48, 2.3 Projected Wins

Team Wins Produced [with Nash and Frye from 2007-08]: 46.1

Again, the team is currently on pace to win 48 games.  So the story in Phoenix is simply that Nash and Frye have reverted to what we saw in 2007-08. 

It’s important to emphasize.  Frye from two years ago was still below average (and a bit better than what we are seeing this year).  He just wasn’t as bad as he was in 2008-09.

All-Star Amare?

When we look at the leader in Wins Produced, we don’t see Channing Frye.  And we don’t see fellow big man Amare Stoudemire until we get past Nash, Jason Richardson, Grant Hill, and Jared Dudley.  Yet it’s Stoudemire  — with the help of David Spade – who is campaigning to start the 2010 All-Star game. 

Once upon a time it made sense for Stoudemire to think about the All-Star game.   In 2004-05, Stoudemire produced 12.9 wins with a 0.214 WP48 [Wins Produced per 48 minutes].  After missing much of 2005-06 with injury, Stoudemire came back to produced 12. 1 wins in 2005-06 [with a 0.217 WP48]. And then in 2007-08, he posted a 0.251 WP48 and produced 14.0 wins. 

Last year, though, Stoudemire’s WP48 dropped off considerably.  Many people blamed the addition of Shaq or perhaps a change in the team’s offensive philosophy.  This year, though, Shaq is in Cleveland. And the team’s offense is supposed to be a return to what we saw before Terry Porter became coach.  Despite these changes, though, Amare’s production is hardly at an All-Star level.  So what’s happened?

Table Two: Evaluating Amare

Table Two reports the box score statistics for Amare.  Relative to what we saw prior to last year, Amare is offering fewer rebounds and blocked shots.  These changes, though, don’t explain the entire gap.  Another issue is the number of shots Amare is taking.  More specifically, Amare is simply taking fewer shots than he did in the past.  And because he’s an efficient scorer, this reduction in shot attempts is reducing his overall production.

Unfortunately for Stoudemire, both Nash (59.6% adjusted field goal percentage) and Richardson (56.8% adjusted field goal percentage) are currently more efficient scorers.  So it’s not clear that giving more shots to Stoudemire (55.8% adjusted field goal percentage) is such a good idea.  Still, it does appear Stoudemire’s drop off in shot attempts – coupled with a decline in rebounds and blocked shots – explains why Stoudemire is no longer that productive.

So here’s what we’ve learned.  In spite of their impressive record, Phoenix is not dramatically improved over what we saw last year.  What improvement we have seen can be linked to Nash and Frye reverting to form.  And Stoudemire – who really wants to be an All-Star – is simply not as productive as he was in the past. 

Can any of this be linked to the departure of Shaq?  I really don’t think so.  Although Shaq bears some responsibility for what’s happened in Cleveland, the small improvement we see in Phoenix – and it is rather small – is really not about  Shaq departing.

- 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.