Houston Still Has a Problem

The writer’s strike has meant that we have had to learn to live with re-runs.  Consistent with the current state of television, the WoW Journal is also repeating its programming.  About a month ago I posted the following on the Rockets struggles this season.

The Houston Rockets Fail to Launch

This column made the following two arguments:

1. As I argued HERE and HERE, we should not have expected the Rockets to contend with the Best in the West in 2007-08. Still, Houston should not have become much worse.

2. The Rockets failure to launch is tied to a weakened supporting cast.

Although what I said a month ago is still basically true, I still want to return to the topic of this team’s troubles.  Hopefully this column can shed additional light on what’s happening in Houston.

Reviewing the Decline

In terms of efficiency differential – offensive efficiency minus defensive efficiency – the 2006-07 edition of the Houston Rockets was the best in franchise history.  Given that this was accomplished with Yao Ming missing 34 games, there was an expectation the Rockets would be even better in 2007-08. This optimism was also fueled by the trade of Juwan Howard for Mike James, the trade for Luis Scola, and the signing of Steve Francis.  The combination of all these moves led John Hollinger to forecast that the Rockets would be the NBA’s best team in 2007-08.

After 28 games, though, it’s quite clear the Rockets are not the NBA’s best team.  In fact, with a differential of 0.4, this team is not even counted among the NBA’s elite.  So what happened?

To see why this team has declined we turn to Table One. 

Table One: Projecting the Rockets after 28 games

Table One offers two projections of the Rockets in 2007-08.  The first looks at how many wins we could expect if each player plays as well as he did last year. The second estimates how many wins each player will produce if he keeps playing as well as he has this season. 

Had each of the Rockets maintained what we saw last year, this team would be on pace to win about 51 games.  Such a mark would still be lagging behind the Spurs, Lakers, Suns, Jazz, and Mavericks in the West.  But it clearly would have placed this team in the playoffs.  

With a differential of 0.4, though, this team is only on pace to win 42 games.  And this mark puts a playoff appearance in jeopardy.

Explaining the Decline

To see who is responsible for this decline, we turn to the performance of the individual players.  In studying the Mavericks we saw that this team’s decline was primarily linked to the play of Dirk Nowitzki.  When we looked at the Lakers, we saw this team’s improvement was primarily linked to the increased production from Andrew Bynum.

For the Rockets, the differences we see in team outcomes can’t be linked to just one player.  Instead, we see declines in the production of Shane Battier, Yao Ming, Tracy McGrady, Mike James, Steve Francis, Dikembe Mutombo, and Luther Head.

With so many players offering less, the explanation of what’s plaguing this team becomes more difficult to pin down.  My co-authors (a list that includes Martin Schmidt, Stacey Brook, Tony Krautmann, Mike Leeds, and Mike Mondello) and I have looked at various factors (in various papers) that we have found to have a statistical link to changes in player performance. I think many of these factors might be relevant to the Rockets this year.  So here’s a list of issues – in no particular order — that might explain why Houston has such problems.

1. Coaching change

Before the season started the team switched head coaches. Rick Adelman and Jeff Van Gundy have both been found to positively impact player performance.  But we have also found that just switching coaches causes per-minute performance of individuals to decline. Therefore, one possibility is that Houston’s players are simply having trouble adapting to a new coach.

2.  Roster stability

In addition to changing coaches, this team has also made some significant roster changes.  James, Scola, and Francis are all new faces.  We have found that keeping a roster together will boost the per-minute productivity of an individual player.  Likewise, new teammates can cause the productivity of individuals to decline.

3. Minutes Played

The more minutes a player plays, the better he performs on a per-minute basis.  In other words, controlling for what a player has done in the past, increasing his minutes tends to lead to more productivity per-minute.  Likewise, if you cut a player’s minutes, he tends to give you less on a per-minute basis.

Part of this might be coaches giving more minutes to players who have actually improved.  But we think that part of this might reflect an old adage (at least, I think this is an old adage) you hear in football.  “If you have two starting quarterbacks you really don’t have one.”  In other words, if you try and split playing time between two players you tend to get less from both.

In the off-season the Rockets added guards James and Francis to a team that already had Rafer Alston, Luther Head, Bonzi Wells, and McGrady.  With six players commanding minutes at the two guard positions, the team has to cut everyone’s minutes to get everyone some playing time.  Alston, James, Francis, and Head are all playing far fewer minutes than they did last year.  It’s perhaps not surprising that three of these players are offering less on a per-minute basis.

4. Age and Mutombo

One of the big reasons the Rockets were so good last year was the play of Dikembe Mutombo.  When Ming went down with an injury it was expected that this team would collapse.  Instead, as noted last year, Mutombo was able to produce wins in place of Ming.  

This year Mutombo’s minutes have been cut.  In addition, Mutombo is not offering much in the few minutes he is getting.  Part of this might be the minutes issue just noted.  Part of this might be age, which ultimately has to limit productivity. 

Regardless of the reason, Mutombo is not offering much this year.  And since he produced 9.0 wins last season, his absence has also harmed this team’s chances to contend.

5. McGrady and Battier

Both of these players are offering less.  In my previous post on the Rockets I offered some thoughts on Battier.  And I was planning another post on the Battier and Rudy Gay trade (this is not it).

McGrady, though, I have not talked about much in this forum.  I will note that across McGrady’s first eight seasons he produced 107 wins, for an average of 13.4 Wins Produced per year.  Across the past two seasons he only produced 15.6 wins.  Part of this is because injuries have cut into his minutes.  His per-minute performance, though, has also declined.  Yes, he is still above average.  But he’s not the dominant player we saw in Orlando.

One suspects that McGrady’s injuries have also taken a toll on his productivity. Perhaps a post detailing his decline might be a good idea in the future.

For now, let me end this post by noting that there are many possible explanations for why Houston is not playing well this year.  Changing coaches, changing rosters, changing minutes, age, and injury can all be cited as possible reasons for why so many players are offering less this year. 

Which one of these is most important? Hmmmm…I don’t know.  But I look forward to seeing in the comments section which issue people think matters most.  Hopefully the factor that gets the most votes is at least on my list.

– DJ

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.

Posted in: Basketball Stories