In 1977-78 the Washington Bullets won the NBA title. The next season, this team won 54 games and appeared in the NBA Finals. And then this franchise descended into the depths of the Association. Across the next 24 seasons, the Washington franchise never won more than 44 games and only finished with a winning record five times.
And then, Gilbert Arenas came to town. Although his first year was not a success, the Wizards have finished with a winning record each of the past three seasons. And each time, Agent Zero led this team in Wins Produced. In sum, as the following list reveals, Arenas is a very good player.
Season: Wins Produced, WP48
2006-07: 11.0, 0.180
2005-06: 11.1, 0.157
2004-05: 12.3, 0.181
2003-04: 2.2, 0.052
2002-03: 6.8, 0.114
2001-02: 3.0, 0.123
Career before 2007-08: 46.4, 0.142
Surviving the Loss of Agent Zero
This season Arenas has only appeared in eight games. And with a WP48 [Wins Produced per 48 minutes] of 0.017, Arenas played quite poorly.
Despite the poor play and eventual absence of Arenas, the Wizards – as noted a few days ago – posted a 13-10 mark after 23 games. With an efficiency differential of 2.4 (offensive efficiency minus defensive efficiency), this team was on pace to win 47 games. If this happened, this would be the best record posted by a Washington team since that NBA Finals experience in 1979.
So how could this happen? To find an answer, let’s go back to the approach I offered earlier in the season. Specifically, I am going to offer two projections. The first assumes the players on the team will perform as they did last year. The second will look at what will happen if the players on the team keep playing as they have this season. Each perspective is offered in Table One.
Table One: Projecting the Wizards after 23 games
If each player maintained what he did last year (except for the rookies), this team would be on pace to win 34 games. Instead, as noted, this team is on pace for 47 victories. When we look at the second projection – based on this year’s performance – we see that majority of this leap can be tied to the improved play of Caron Butler and Brendan Haywood.
Butler and Haywood Defy the Usage Story
To see how these two improved, consider Table Two.
Table Two: Haywood and Butler after 23 games in 2007-08
Relative to last year, Haywood has improved with respect to rebounds, blocked shots, and scoring. Surprisingly (at least to me), the change in scoring is not related to improvements in shooting efficiency. No, Haywood is simply taking more shots from the line and the field. And perhaps surprising to some, the increase in Haywood’s shot attempts has not caused his shooting efficiency to decline.
The same story is seen with respect to Butler. With Arenas out of the line-up, Butler is taking more shots. His shooting efficiency (except for Wednesday night), has increased dramatically. Although Butler has also cut down on his turnovers (despite having the ball more often), it’s his increase in shooting efficiency that’s primarily responsible for his increased productivity.
More on Usage
One of the stories people tell about basketball is that star players create their own shots. Consequently, star players should not be credited for just making shots, but they should also receive extra credit for just hoisting the ball in the direction of the basketball. I have been skeptical of this approach for two reasons:
1. There does not seem to be much of a relationship between shooting efficiency and shot attempts.
2. Even if there was such a relationship, the usage story is really not a part of how productive a player has been. It seems to me that this story tells us more about why a player is productive. And I have always been in favor of treating the how and why questions separately.
Back to Arenas
Returning to Arenas and the Wizards, do we see evidence that this team cannot get shots off without Agent Zero? Last year this team took 83.2 field goals per game. After 23 games, this team took 83.4 field goal attempts per game. So shot attempts have gone up slightly.
When we turn to adjusted field goal percentage, we see the Wizards had a mark of 49% last year. And this year, after 23 games, the Wizards mark is 49%. Yes, without Arenas the team is taking basically the same number of shots with the same level of efficiency.
This result leads me to ask two questions:
1. Does all this “prove” the usage story is bunk? No, this doesn’t “prove” anything. It’s interesting, though (at least, I think so).
2. Is there any evidence Agent Zero is missed? Yes, but it’s free throw attempts where we see an impact. Last year this team took 29.6 free throw attempts per game. This year the team is only taking 25.2 shots from the line each game.
This result highlights one of the under-appreciated aspects of a typical star’s game. Stars get to the free throw line. And the charity stripe is a very effective place to create points.
The Big Question
Although this team misses Agent Zero’s ability to get to the line, the team is still winning without him. Let’s assume Arenas does come back from his injury and can return to what we saw the past three seasons. Given this team’s ability to win without him, should Washington pay the major contract Arenas demands to keep his services?
If Arenas can return to what we saw in the past, and Butler and Haywood maintain their levels of efficiency, the Wizards could be a 50 win team in 2008-09. In other words, Arenas will definitely help. But even though he has been the best player on this team for three consecutive years, he’s not one of the very best – i.e. top ten – players in the league. At least, he’s not if you look at Wins Produced. This is a point I made in the following columns:
The Washington Wizards: The Anti-Nets
Star Power and the Washington Wizards
The Washington Wizards in 2005-06
The problem with signing Arenas is that his scoring (where he is a top ten player), on the current NBA’s market, costs more than the value of his wins. And so re-signing Arenas may not be worth it.
Of course, all that assumes Arenas can come back this season and demonstrate he is healthy. If he can’t, but somehow heals before 2008-09, perhaps Washington will not have to overpay to keep Agent Zero.
Then again… yes, you can go around and around on this. In the end, one thing is clear. It certainly looked like Arenas was due a maximum contract last summer. But with his injury, and the team’s success without him, that’s not as sure anymore.
– 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
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.
Wages of Wins Journal 
Panda Bear
December 20, 2007
This is a great post. I love the analyses like these when it previews or reviews a team that just lost or gained a player. It’s like a good out of sample case study in wins produced. Wins produced was prescient with this one I think (based on earlier posts on agent zero).
BasketballGuy
December 20, 2007
Another nice column.
Kevin Broom
December 20, 2007
The usage/efficiency question is an interesting one. I’m not sure this is going to be a great test case (although I’m sure folks will try to use it). The key issue in the usage/efficiency question is one of ability — can a player increase his usage while maintaining efficiency? However, the fact that Butler is doing it now does not necessarily mean he could have done it last season. There’s plenty to suggest that Butler has actually improved — that his skills have not remained static.
Specifically, Butler improved his workout regimen and came into the season lighter, stronger and more flexible — in the best shape of his life. He improved his shooting by working with the team’s new assistant coach and shooting instructor Dave Hopla. And, he improved his diet by cutting out stuff like high fructose corn syrup. Last season (and throughout his playing career) he drank a 2 liter Mountain Dew during each game (half before the game, half at halftime). This season he’s cut that out and his stamina is improved. In other words, the fact that his efficiency has improved with more usage THIS season does not prove that he could have done it at any time previously. The evidence suggests that there has been actual improvement in his physical capabilities and basketball skills.
The comparison of shots per game doesn’t really address thinking about shot creation. Teams are going to take shots because of the shot clock. All shots are not created equal, however. Some are more difficult, some are easier. I think one of the perceived benefits of high-usage “stars” is that they have the ability to attempt shots while well-defensed that have some chance of going in where teammates with lesser ability would have less of a chance of making a similar shot.
The thinking here is that when the shot clock is running down on a vigorously defended possession, SOME shot (even a shot that has a 1-in-10 chance of going in) is better than NO shot. Someone has to take the shot, and the ball usually ends up in the star’s hands in that situation. The issue is not (and never was) that removing a guy who puts up 20 FGA per game will reduce his team’s FGA by 20 per game (or some other number). It’s that removing his shot attempts then pushes the possessions (even the more difficult ones) onto other players and a drop in efficiency results.
I use the term “efficiency” intentionally, because the drop may not necessarily come in shooting percentages. Someone at APBRmetrics posted at some point that the effect may show up in increased turnovers. Dean Oliver posted his finding (though he did not describe his method in any detail) that for every 1% increase in possession usage, there’s a 0.6 drop in efficiency (using Dean’s offensive rating measure).
Very interesting post, as usual.
dberri
December 20, 2007
Kevin,
Excellent explanation of the usage hypothesis. Thanks for posting this.
Oren
December 20, 2007
“Relative to last year, Haywood has improved with respect to rebounds, blocked shots, and scoring. Surprisingly (at least to me), the change in scoring is not related to improvements in shooting efficiency. No, Haywood is simply taking more shots from the line and the field. And perhaps surprising to some, the increase in Haywood’s shot attempts has not caused his shooting efficiency to decline.”
I’m wondering if you can explain this statement. Haywood is shooting an impressive 15% better from the FT line this year then last year.
If my calculations are correct, Haywood’s WS Points from just shooting has increased 1.475 this year. .77(53%) are due to the fact that Haywood is shooting more efficiently from the FT line.
Pete23
December 20, 2007
Kevin Broom, great comment.
Jason
December 20, 2007
In 0607, 12% of the Wizards’ shots came with more than 21 seconds off the shot clock. They made 41% of these shots (effective FG%). This year, 16% of their shots come in the final 3 of the shot clock and ~50% of these are going in (again, effective FG%).
More of their shots are coming late this year, but more of them are going in. This is anecdotal, but as far as the Wiz are concerned this year, without Arenas, they’re taking more late shots, but also making more late shots. Butler appears to be responsible for much of this. More of his shots are coming late in a possession and he’s increased his percentage of makes on these substantially. Similar things are true for Haywood and Jamison. In Haywood’s case, the FG% increase is monumental.
Owen
December 20, 2007
Oren – I think what DB probably meant was that Haywood’s fg% is slightly lower than it was last year. You are correct though that his overall scoring efficiency is higher due to his ft%. His ts% is at 60.1%, 3% higher than last year.
Kevin – Excellent comment indeed. I noticed over at 82games is that Arenas took 44% of his shots in the last 10 second of the shot clock last year, with an efg 4% higher than the 27% he took in the 11-15 slot.I don’t know what that means, but it seems that Arenas did in fact do a lot of bailout shooting for the Wizards last year. But they seem to have been able to adapt in his absence this year.
Patrick Minton
December 20, 2007
Jason,
Very good point.
I think it’s also important to point out that shots that fall later (but not too late) in the shot clock are more likely to be “good” shots, because the team expended more effort to work the ball around (as opposed to just jacking up any old shot the first time a player can see the rim).
Of course, that’s just my intuitive feel for it. The Pheonix Suns offer a great counter-example…
Westy
December 20, 2007
Good comment, Kevin.
Using anecdotal evidence such as looking at what the Wizards are doing this year is very worthwhile, but there are variables and questions that remain unanswered.
For instance, if Butler is shooting at the same efficiency as last year despite all his offseason regimens, wouldn’t it seem likely he’d be shooting at a lower percentage if he were not in the best shape he’s ever been in?
Likewise, while I suspect many players would not be able to sustain high usage (and the better defense that would come with it), there are certainly players out there who may currently be underutilized. Butler could be an example of this. Maybe he IS better at creating last-second (in the shot clock) shots than Arenas, but never could show it previously. There are probably other players in the league who are fantastic at creating their own shots who aren’t yet using that ability if their team already has someone ‘designated’ for that role.
Finally, in looking at usage, there are two types of players who take a lot of shots (and these categories may overlap). One is the person who will jack up the first halfways-open shot they see. The other is the person who usually ends up taking the shot at the end of the shot clock. Both will likely tend to shoot more than their teammates. What the presumption that efficiency doesn’t change as usage increases says is that no matter who takes those types of shots, their efficiency will remain the same. That says that if each team designated 100% of their shots be taken by their highest efficiency shooter, they would maximize their offense, which doesn’t make sense.
Oren
December 20, 2007
Personally, I think that the Wizards are a good example to prove the validity of Win Score.
When Arenas went down, Daniels became the starter and took his minutes. Well guess what, Arenas didn’t have a much higher WP48 then Daniels. So, it makes sense that the Wizards wouldn’t do much worse with Daniels then with Arenas. And they didn’t, so one point for WS.
But now that Daniels is injured, Nick Young is starting. And Mason is the first guard off the bench. So, at this point, the Wiz should start to lose most of their games because their guards are all below average according to WS, and Nick Young has a negative WS48.
If they do lose most of their games with those two guys out, then this is a perfect example to show the validity of WS.
It’s worth noting that the Wiz lost last night because their guards sucked.
mrparker
December 20, 2007
KOBE BRYANT IS THE BEST PLAYER ON THE WIZARDS!
mrparker
December 20, 2007
all joking aside…why wouldn’t a team improve if there main shot taker improved his shooting percentage by 5%. Thats basically what happened here. Butler is more efficient and took Arenas’ shots. They had an actual point guard playing point guard who is distributing the ball well ADaniels had a wp48 above .2 at some point last season. The guy is a player.
dustin
December 20, 2007
Mrparker, the reason they wouldn’t improve is that if shot creation was valuable and arenas was able to do it and no one else on the team was. In this instance, shot creation does not seem to be a commodity
andrew
December 20, 2007
Actually it would appear shot making is exactly a commodity.
-Andrew
dustin
December 20, 2007
Let me rephrase, a rare commodity
Pete23
December 20, 2007
mrparker, Bryant doesn’t play on the Wizards. He still plays on the Lakers though is the subject of frequent trade rumours.
Kevin Broom
December 21, 2007
I’m puzzled about why the Wizards would show the validity of Wins Produced. First, because one event wouldn’t really prove the validity of ANY measure. Second, because I hardly think Wins Produced would be the only measure that could show why the Wizards have done okay without Arenas but are likely to suffer without both Arenas and Daniels. (Shoot, I think I’ve explained it reasonably well without reference to a single number.)
I know what PER says, but PER isn’t the only other measure out there. Most stats analysts (even of the amateur variety) use an array of stats in their analysis. In my own spreadsheet, I include a system similar to NBA Efficiency, a simplified version of PER, Win Score, Dan Rosenbaum’s Alternative Win Score, as well as a heap of measures of my own devising. (One of my favorites is taking a player’s per minute stats, adjusting for the pace and position, then resetting the average in each category to 10 (purely arbitrary number). It gives me a quick and handy reference for how a player ranks relative to average for those I’ve decided play at the same position.)
The validity of Wins Produced will be shown/not shown based on the accuracy of MANY predictions — not any one case.
Kevin Broom
December 21, 2007
As for relationships between usage and efficiency, logically there MUST be a relationship. Otherwise coaches would simply insist that the team’s best shooter take all the shots. Clearly there’s some benefit to having different players take shots, and to teammates working together. Unless the argument is that coaches simply misunderstand this aspect of the game.
I think there’s lots of variation — some guys will be unaffected by shooting more frequently, some might shoot better, some might shoot worse. What would be interesting would be see if there’s a way to predict which guys fall into which categories. This is a tough area to study, though. I have a few thoughts on how to do it. If I have time, I may run a small-scale test to see if it’s a worthwhile way to examine the issue.
doclinkin
January 10, 2008
Westy says:
“For instance, if Butler is shooting at the same efficiency as last year despite all his offseason regimens, wouldn’t it seem likely he’d be shooting at a lower percentage if he were not in the best shape he’s ever been in?”
Maybe. But not necessarily. Yes Arenas’ presence on the floor actually creates easier shots for his teammates because he requires more defensive focus from the opponent. His remarkable range opens up room underneath for players with Caron’s advanced skillset in the paint.
Caron has consistently ranked high in the league in converted And-1 chances. By observation this year he seems to be collecting fewer of those opportunities, whether due to the absence of an outside threat to unclog the paint or because his lighter frame absorbs impact less well.
But the greatest improvement in Caron’s game is his sudden proficiency in shooting the outside shot. He’s a far easier way to get your three points than the traditional method.
During a hot stretch early on he was actually leading the league in FT shooting %. (At least until teams learned to respect his range and guard him out there).
Also, the knock on the hybrid Princeton offense that the Wizards run is that it often requires a few iterations of running sets resets and counters before the lay-up lane or open shot is available. This would account for more late-clock shots for any/all Wizards players compared to when a quick ballhandler like Arenas is on court to push the ball upcourt and force the action.
Without Arenas’ bail-out talents, the team has been more faithful to the scripted offense. Previously Coach Jordan encouraged players to run the jailbreak upcourt and take the first open shot available, Pheonix style since with Gilbert (and Larry Hughes before him) the team had players suited to the fighter-jet speed attack style.
But Daniels is more of a pull-up set-up-the-offense type of point guard. The team has been mroe patient in running the sets and counter counters with Daniels on the floor.
The difference in pace may not show up in FGA’s since many of the quicker possessions resulted in Free Throw attempts not field goals.
Ho will this change when Arenas returns? I expect a drop in efficiency in the short term as the team has to rediscover it’s rhythm and re-integrate Gil. And naturally you’d expect a player returning from a essentially year off for injury to spend some time knocking the rust off and figuring out how much to trust his knee. And whether or not he can regain his remarkable quickness (or how quickly he can recover it) remains to be seen.
Still I expect his deadly range to return, which should only help the team in the long haul.