The Washington Wizards Are Finally Embracing Math, But Do the Numbers Like Them Back?
By spacing the floor and shooting threes, the Washington Wizards plan to play modern, analytically influenced basketball. What do the numbers say about their chances this season?
Bill Streicher-USA TODAY Sports
The Washington Wizards enter the 2015-16 NBA season as a sort of in-between team: a mix of developing youth and declining veterans, built to win now but not exactly all in, grafted together by a front office attempting to maintain respectability while preserving ample salary cap space for next summer, the better to pursue Kevin Durant or another elite free agent.
A historically weak Eastern Conference and stop-gap acquisitions have put the team in the playoffs the past two seasons, but this year's roster is more likely to yield relatively disappointing results on the court. Still, the season could end up being a success—provided Washington ends up with Durant or another top talent.
After receiving heavy criticism for running an offense seemingly stuck in the 1980s—two lane-clogging big men on the floor at the same time; a inefficient diet of long two-point jump shots—Wizards coach Randy Wittman moved Paul Pierce to power forward and unleashed a small ball offense during last year's playoffs that nearly carried Washington to the Eastern Conference Finals.
Pierce is gone, but the Wizards' newfound commitment to "pace and space" remains. Predicated on playing fast, spreading the floor, and shooting more threes, Washington's new look looked good in the preseason: 27.8 percent of the team's field-goal attempts were from three-point range, up from just 20.3 percent last season, and the team's pace factor (possessions per 48 minutes) rose from 93.6 to 101.5.
While all this may feel like radical change to Wizards fans, this offense doesn't actually qualify as innovation; it's adoption of the new orthodoxy. During the preseason, the entire NBA played faster and shot more threes, and Washington was actually below average in three-point attempts.
Moreover, while "upping the pace" sounds good, it means little in a vacuum. Given that teams have roughly the same number of possessions in a game, what matters is relative efficiency: How does a team perform during its possessions compared to its opponent?
If playing faster doesn't boost Washington's efficiency differential, it won't help the team win. League history indicates that pace alone doesn't correlate with offensive or defensive efficiency, and my analysis of the Wizards specifically over the past few seasons suggests that they've been slightly better in slower-paced games.
Look, it's hard to argue that embracing contemporary offense isn't a good thing, especially considering how much Washington hamstrung itself with two-point jumpers the past few years. However, the Wizards may have forfeited some of their comparative advantage by waiting to enter basketball modernity. Better spacing and an emphasis on threes instead of long twos could improve the team's offensive efficiency, but it's coming at a time when many other teams are doing the same things—and designing new defensive strategies to counter them.
Over the years, I've ripped so many decisions made by Washington general manager Ernie Grunfeld and team owner Ted Leonsis that what follows could reflect a personal bias. I don't think it does, but be warned.
With that out of the way: I think the team's roster moves indicate a wrong-headed understanding of the nature of basketball. Specifically, the team continues to acquire players based on specific skill sets rather than their overall impact. This offseason, they signed swingman Alan Anderson and guard Gary Neal, ostensibly to add "shooting." In previous years, they've kept or acquired players to supply defense (Garrett Temple) or rebounding (Kris Humphries and DeJuan Blair) or ball-handling (Eric Maynor) or shooting (Martell Webster, Al Harrington).
Washington's front office seems to have a football mentality, where situational specialists are useful because they can be easily inserted or removed from the lineup to maximize strengths and hide weaknesses. Only in basketball, you have to take the whole player. A player helps a team to the extent that all of his contributions are greater than the all of his negatives. "Player X can shoot the three" doesn't mean much if Player X also commits turnovers, doesn't rebound, and plays poorly on defense.
In Pursuit of Kevin Durant
Of course, none of that matters if the team somehow wins the Durant sweepstakes. As you'll see in the individual player projections to follow, Washington needs a talent infusion, and they desperately need an elite producer if they're going to move from the NBA's muddled middle to title contention.
Placing a prime Durant on this year's Wizards would improve the team by nine or 10 wins. My concern with their short-term roster moves is that they could lead to a lower win total this season, and therefore hinder their pursuit of Durant or another elite free agent—in part because top players typically sign with winning teams; in part because disappointment this season could call into question Grunfeld's ability to construct a championship roster around Durant or someone else.
The below projections are based on a metric I developed called Player Production Average (PPA). In PPA, players are credited for things they do that help a team win, and debited for things that don't, each in proportion to what causes teams to win and lose. PPA is pace neutral, accounts for defense, and includes an adjustment based on the level of competition faced when a player is on the floor. In PPA, average is 100, higher is better, and replacement level is 45.
The projection system runs each player through my Statistical Doppelganger Machine, which uses an array of statistical categories to find players with similar production at a similar age. Then it looks at what those players went on to do, runs it through a fancy-schmancy algorithm, and—voila!—predicts what the Wizards roster will do this season. Those predictions aggregate into a projection of the team's record.
Let's look at Washington's presumptive starters:
John Wall: A phenomenal talent who's just 25 years old, Wall has managed to walk a seemingly impossible path of receiving well-deserved honors, being somewhat overrated, and simultaneously getting dissed.
In other words: Wall deserved being an All-Star, isn't an elite player, and probably deserves more consideration than he's gotten for Team USA.
Questions: Wall's defense, passing, and transition offense already are strengths. Will the team's new offense give him more room to operate? If so, he could get better shots and reduce his turnovers, both of which would make him even more effective.
What to Expect: While improvement would seem a given considering Wall's youth, his historical comparables peaked young (average age: 24.7). The list is a good one, including Deron Williams, Isiah Thomas, Andre Miller, Terrell Brandon, Chris Paul, and Kevin Johnson. While the full group didn't actually improve much from the point in their careers most similar to Wall's, they also didn't experience significant drop-offs for several more seasons.
Projected 2015-16 PPA: 151
Bradley Beal: Assertions that Wall and Beal comprise the NBA's second-best backcourt, behind Golden State's Stephen Curry and Klay Thompson, are premature, in no small part because Beal 1.) has been average and 2) hasn't improved much since entering the league. Still, Beal remains a fan and media favorite, and I believe he has elite potential, even though the numbers suggest otherwise.
Questions: During training camp, Beal publicly committed to replacing long two-point shots with threes or drives to the basket. Will he remain committed as the season grinds on? Will he be able to continue shooting a high percentage from three-point range as defenses adapt? If the answer to both questions is "yes," he should be a much more effective player.
What to Expect: Like Wall, improvement would seem almost inevitable since Beal is young, talented, and hardworking. Only it hasn't really happened yet. Beal's PPA progression has been flat: 92, 96, 99. Entering a contract year, he seems ripe to make a leap, yet his historical comparables didn't make huge jumps. Some of the names on the list are encouraging: Mike Miller, Jason Richardson, Glen Rice, Thaddeus Young, and—wait for it, Wizards fans—Ray Allen. Other names are not: Calbert Cheaney, O.J. Mayo, and Dennis Scott. Beal needs to show progress this season to merit the max or near-max contract he wants next summer, but unfortunately my numbers suggest his big jump wouldn't be until the 2016-17 season—with a peak PPA around 130—if it comes at all. I hope I'm wrong.
Projected 2015-16 PPA: 102
Marcin Gortat: The team's most productive player per minute last season, Gortat is substantially underrated by fans. He's efficient on offense, solid on defense, and holds his own on the boards. He's also 31 years old, which is entering the age range where players often experience significant drops in health and production.
Questions: Will normal aging effects be offset by a combination of Gortat's superb conditioning, low career minutes for his age, and the team's shift in offensive philosophy? (Gortat thrived in Orlando and Phoenix when surrounded by shooters and working with a deft pick-and-roll partner in Steve Nash.)
What to Expect: More of the same. Gortat posted the second-best PPA number of his career last season, and I anticipate him being much the same player in 2015-16, albeit with a modest dip in production based on a combination of age, regression to the mean, and random fluctuation. Players most similar to Gortat tend to experience performance declines at this point in their careers, so asking him to improve is probably asking too much.
Projected 2015-16 PPA: 153
Kris Humphries: Humphries has been a good jump shooter throughout his career, but he limited himself to long twos, which are the worst shots in the game. Via some regression analysis this summer, I estimated that a player who shoots long twos like Humphries would likely be a good three-point shooter—around 36-38 percent. In the preseason, Humphries seemed to be trying to prove me right, hitting 8-of-22, a very respectable .364 from beyond the arc.
Questions: Can Humphries add a three-point shot to his game without sacrificing his other contributions, particularly offensive rebounding? Can he get better defensively this late in his career?
What to Expect: So far, so good in the grand "Kris Humphries, stretch four" experiment. His shooting from the three has been adequate, and doesn't seem to be hurting other aspects of his game. Players similar to Humphries last season tended to age well; many improved at age 30, some significantly.
Projected 2015-16 PPA: 128
Otto Porter: There's so much excitement about Porter this season that I fear the former No. 3 overall draft pick will have a hard time living up to expectations. Porter had a good but not great postseason with a 125 PPA. It was a small sample size, though—just 331 minutes—and while his 2014-15 regular season wasn't terrible, it was below average.
Questions: Who is the "real" Otto Porter—someone like the above-average contributor from the playoffs and preseason, shooting with confidence and logging major minutes, or someone closer to the below-average guy from the 2014-15 regular season?
What to Expect: With the departure of Paul Pierce, Porter will take on an expanded role with the team. Historically, players most like Porter in 2014-2015 tend to improve modestly the following year before making a more substantial leap in the future. I anticipate an inconsistent season from Porter, with a trend toward improvement as the season progresses and he learns how to be a regular starter.
Projected 2015-16 PPA: 87
The Bench: Declining, oft-injured Nenê (projected PPA: 75) moves to the bench, where his defense, passing, and ability to hit jump shots—even though he shoots too many—will be a major upgrade over departed, woefully unproductive Kevin Seraphin. Ramon Sessions (projected PPA: 89) was solid last season as Wall's backup and figures to be about the same. New acquisition Jared Dudley (projected PPA: 91) is recovering from offseason back surgery, but can be a valuable contributor, locker room leader, and spot-minute stretch four when healthy. Alan Anderson (projected PPA: 42) has a reputation for defense and shooting, but my analysis indicates that he's only good at the latter. Drew Gooden (projected PPA: 59) is a relatively cheap, "break glass in case of emergency" big man whose productive three-point shooting is offset by horrific defense. Trading up to get Kelly Oubre (projected PPA: 78) in the 2015 draft could turn out to be a terrific move for the Wizards, as he rated as a top 10 prospect in my stat-based analysis; however, Oubre is only 19 and Washington has a suspect history of developing young players. Deep bench players Gary Neal, DeJuan Blair, Garrett Temple, and Martell Webster shouldn't play major roles, though Temple is a competent reserve, while Blair had an outstanding preseason and appears to be in much better shape than last season.
The Bottom Line
The Wizards hit the pause button on immediate efforts to improve their roster in order to make a run at Durant, and it shows in my projection. They've benefited the past couple seasons from playing in the East, which has been home to teams like Philadelphia and has somewhat masked Washington's overall mediocrity. A weak East could work in the Wizards' favor again, but I expect their record to more closely match their roster. Only four players (Gortat, Wall, Humphries, and Beal) project as above-average players this season—and while Gortat and Wall rate as likely All-Stars, no one looks to be the kind of elite producer (hi, KD!) necessary to make the Wizards a truly competitive team. Of the players who project below average, only Porter realistically could come through with above-average production instead—and he's not going to be elite this season, either.
Prediction: 42-40 and the sixth seed in the playoffs.