The NBA season is almost half over (Tuesday night is the official midpoint). Given the relative consistency we see in basketball (relative to football and baseball), teams that are currently at the bottom of the standings should probably be making plans for the lottery. In other words, the Washington Wizards – who currently reside in last place in the Eastern Conference – know this season is over.
Washington Looks at Next Season
Washington only won eight of their first 39 games. The team’s efficiency differential (offensive efficiency minus defensive efficiency) is -6.72, a mark that also ranks last in the conference. For this team to reach the 0.500 mark it would have to win 33 of its last 43 games. Such a pace would be consistent with an efficiency differential in excess of 8.0. In sum, the Wizards would have to improve by nearly 15.0 in efficiency differential to just win half their games in 2008-09. Even if Gilbert Arenas came back (and there are indications that will not happen), Washington is not going to improve this much.
Although the season is effectively over in Washington it doesn’t mean the 2008-09 campaign was entirely negative. Table One reports the Wins Produced and WP48 [Wins Produced per 48 minutes] of each player employed by the Wizards this season. As one can see, most of this team is below average.
Table One: The Washington Wizards after 39 games in 2008-09
Not surprisingly, Antawn Jamison and Caron Butler – the two stars on this team – are posting above average numbers. And joining these two – and this might be a surprise – is Dominic McGuire.
McGuire was taken in the second round by the Wizards in 2007. Last year – in only 695 minutes – McGuire was below average (not surprising for a rookie). This year he has already played 648 minutes and started 14 games. And although he is only averaging 3.0 points per game, his WP48 of 0.163 leads the team (primarily because he has improved with respect to rebounds and turnovers).
If McGuire can continue to play well – and Arenas and Brendan Haywood can both play and be productive next season – the Wizards will have a formidable starting five in 2009-10 (Arenas, Butler, McGuire, Jamison, and Haywood).
The remainder of this season, though, is not going to look good. At least, not for fans of Washington.
The Pacers Have Some Hope this Season
If we look at the team just above Washington in the standings, though, a different story could be told. Again, if a team is out of contention today it will probably be seated at the lottery in a few months. An exception to this story, though, might be the Indiana Pacers.
Across the first forty games the Pacers only won 15 games with a winning percentage that only ranks ahead of the Wizards in the East. But when we look at efficiency differential, the Pacers mark of -2.06 currently ranks 9th in the conference.
Indiana’s efficiency differential is consistent with a team that should have already won 18 games (or 36 games across the entire season). Such a mark is probably still not good enough to make the playoffs, but the Pacers would not be incorrect if they thought they were close.
When we look at Table Two we can see which players are responsible for giving Indiana some hope.
Table Two: The Indiana Pacers after 40 games in 2008-09
For many fans, the player who is leading the Pacers is Danny Granger. He certainly lead the team in scoring. But when we look at overall production, Granger is only slightly above average.
Again, as a scorer Granger is very good. But when we look at net possessions (a term coined by frequent WoW Journal and Knickerblogger commentator Owen Breck) Granger is below average. Per 48 minutes he posts the following numbers:
Rebounds: 6.7 (average 7.6)
Steals: 1.4 (average 1.6)
Turnovers: 3.5 [average 2.8]
All of Granger’s net possession [rebounds + steals - turnovers] marks are below average. Consequently, although he’s an above average player, he’s not the leading producer of wins on the Pacers.
If we look over this roster we see that Troy Murphy, Jeff Foster, and T.J. Ford all post WP48 marks beyond what we see from Granger. And if we look at last year’s numbers, we see that Mike Dunleavy – who just returned from injury – is also capable of putting up numbers beyond what we see from Granger.
After 152 minutes, though, Dunleavy hasn’t yet returned to the form we saw last year. If he does, though, it’s quite possible that the Pacers will make the teams ranked ahead of Indiana sweat a bit. Although it’s unlikely given the Pacers current record of 15-25 that Indiana can finish in the playoffs, there’s certainly hope heading into next season.
Next season Indiana – like the Wizards – could field a starting line-up that consists entirely of above average players (Ford, Dunleavy, Granger, Murphy, Foster). And that line-up – if on the court the entire season – could contend for the playoffs in 2009-10.
Scheduling Note
The only teams I have not yet devoted a post to in the first half of this season are the Dallas Mavericks and the New Orleans Hornets. On Tuesday I will comment on the former. The Hornets will be discussed next Thursday.
After I review the 2008 NFL season next Sunday, I will turn to reviewing the entire first half of the 2008-09 NBA season. The first of these posts – and there will be more than one on the NBA at the mid-season mark — will be offered on Tuesday, January 27.
- 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
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.
Foster’s decline is rebounding/turnover related, while Murphy has picked up the slack in the former.
Looking at their season stats, the Pacers could really use a big man with an outstanding offensive skill set who can provide consistent minutes. In other words, they miss having a healthy Jermaine O’Neal, even if he isn’t M2P material.
As far as the Wiz are concerned, I think Brandon Haywood, not Arenas, was the bigger loss. Per 36 minutes, he provided 13.7 points, 9.3 rebounds, and 2.1 blocks last season. While these numbers aren’t Dwight Howard-esque, they aren’t terrible, either. Losing him put a dent on the defensive end of the ball.
Also, Arenas is a shoot-first type of point guard. He did not facilitate ball movement as much as, say, Chris Paul or Jason Kidd in New Jersey, not to mention his lack of discretion in shot selection. (Aside: Nick Young has been scoring more while Arenas has been out)
Jermaine O’Neal was never a very productive player… and “outstanding offensive skill set” is a pretty huge leap for a center shooting .460 for his career.
Arenas and Haywood are worth somewhere near 15 wins healthy and Etan Thomas is worth another 5 when healthy. If the rest of the team performs well next season (Jamison and Butler recover last season’s form), the Wizards could be a 50 win team.
Granger’s WP48 would likely improve if he was judged against SGs rather than SFs. He plays like a SG. Pacers are fourth in the league in rebounds, so his SG-like rebound rate is hardly hurting them.
Daniel – Jermaine was ridiculously overrated in his prime, but he had a few productive years by Berri’s metric, and several more productive years by those of us who watch the game of basketball with our eyes and notice his defensive contributions. He was never worthy of MVP whispers, and he was never an elite player, but he was very good for a couple years, and decent for several more years. If he reduced his offensive game to putbacks and work in the post (instead of going to that nasty fadeaway or 17 foot J) he might’ve been a really special player in the league.
As a Wizards fan I hope Gilbert decides not to come back, unless we plan to trade him. This season is a lost cause and I don’t want to have the draft be a lost cause if we win more games. The only reason I would want Gilbert to come back would be so we could showcase him to other teams and trade him in the off season.
If I were Ernie Grunfeld I would try to trade Gilbert to the Timberwolves for Kevin Love and Mike Miller along with one of Minnys several first round pick (preferrably the highest selection but I’d take the second highest).
The contracts match up this season and that would improve the Wizards tremendously. Also noting the Wolves level of competence in the management office, I have reason to believe they’d want to pair a young star Al Jefferson with Gilbert Arenas. Especially with the misperception that Love has underachieved so far in his rookie season.
Then in the draft we could choose Griffin, Harden, Rubio, or Cole Aldrich. Man if I were the GM… One can dream.
Its amazing looking at the Pacers numbers that this team projected to be a playoff team, and the major reason why they aren’t is the decline in the play of….Damn Stephen Graham. Guy goes from 5 mins a game being as productive as D Wade, to 15 minutes a game and costing them wins! Is that the peril of a small sample size? Or an indication low usage players can appear more valuable in this system? Or are those both really the same explanation! So many questions….
Can’t wait to hear about the Hornets.
It looks to me like Stephen Graham’s stats last year were an anomaly. He’s been terrible every other year including this year. Small sample size + luck contributed.
In the Western Conference, the Pacers would likely be out of the playoffs by now.
In the east, they’re just 3 1/2 games back…and tied with Toronto (O’Neal’s current team.)
Actually, as far as big men are concerned, Roy Hibbert, if he can get more minutes, will be just what the Pacers need on offense. His field goal percentage is higher than any season percentage O’Neal has had in his career.
Here are my calculations of each team’s efficiency differential:
Orlando:
Offense: 109.4 pts/100 possessions
Defense: 98.4
Total: +11.0
Cleveland:
Offense: 111.2
Defense: 97.3
Total: +13.9
Boston:
Offense: 108.8
Defense: 98.2
Total: +10.6
LA Lakers:
Offense: 111.7
Defense: 103.0
Overall: +8.7
I do not use the same formula that Dave uses. However, it looks as if Cleveland has exhibited the best body of work of each of the championship contenders thus far, and that the Lakers’ defense leaves something to be desired.
However, LA is still a step above all the other West teams, and looks like a near-lock to reach their second straight NBA Finals.
Hey Dave!
I was wondering if you knew what was up with http://www.WinsProduced.com. It hasn’t been updated at all this year? It would be a very useful and interesting website, much like your own. If it is down for good, do you know any other location I can find up to date wins produced statistics?
I use the userscript for greasemonkey which works with espn to include WP48 in the boxscore. I doesn’t seem to be entirely accurate but I can’t figure out in the code where it’s going wrong. Thanks Dave!
Dave — thanks for the post on my poor, pathetic Wizards.
I’m having some trouble correlating my numbers for Andray Blatche with yours. I have a WS for him, at the same 881 minutes you note, of 10.0 — i.e. .3 below an average power forward at 10.3 (numbers for that come from one of your tables). I’m pretty sure of my numbers and formulas: the model is at http://www.tommandel.com/andrayblatche.xls
If you have him at .65 wp, where an average nba player is 1.0 (and perhaps an average power forward is above that 1.0 number), then either the small .3 difference I come up with is wrong or that small a difference translates somehow into a much larger difference in WP.
Where am I going wrong?
Tom,
Your numbers are right. In calculating WP I combine the averages of power forwards and centers. If you do the same for Win Score you will see that Blatche is further from the average than the strict comparison to power forward suggests.
Hmmm, no I don’t get those results, Dave.
Using the above WS (.208), and your fudge factors that get me to PAWS (.215 for a PF and .225 for a Center), and then your conversion to WP (1.04+(1.621*PAWS)), I get Andray at a WP of .076 (compared to a Center) or .093 (compared to a Power Forward).
Obviously, you are using a more sophisticated algorithm than that to generate WP directly, but I’m surprised at the difference. What am I missing?
(That is, yes, his or any player’s numbers will be lower if treated as a center than as a pf. But, I don’t get anywhere near .065, as you reference above)