Nylon Calculus: Exploring the Bucks’ defensive improvement

NEW YORK, NY - FEBRUARY 4: Demarre Carroll
NEW YORK, NY - FEBRUARY 4: Demarre Carroll /
facebooktwitterreddit

The Milwaukee Bucks have been among the hottest teams in the NBA since Jan. 22, when Jason Kidd was relieved of head coaching duties and succeeded by assistant Joe Prunty.

Entering tonight’s matchup against the Denver Nuggets, they’ve won nine of 11 games, averaging a +7 net rating that marks their highest point of the season to date. They’ve done it by shoring up an area that has held them down for some time.

While the Bucks have remained steady on the offensive end, their defense has made noticeable progress, as the following chart shows:

At the time of Kidd’s departure, the Bucks had allowed 113 points per possession over their previous 10 games — one of the worst defensive ratings during that stretch. They dipped below 110 immediately thereafter. Currently, their 10-game defensive rating stands at 101 and is matched only by the Utah Jazz atop the league.

This abrupt and sharp statistical improvement seems to suggest that the coaching change has already begun to pay dividends. But, of course, it’s much more complicated than that.

For one thing, as observed by Eric Nehm of ESPN Milwaukee, the Bucks have enjoyed a relatively favorable schedule, featuring games against the Phoenix Suns, Chicago Bulls, Miami Heat, Brooklyn Nets (twice), New York Knicks (twice) and Atlanta Hawks. These teams all have bottom-10 offensive ratings. Even the Orlando Magic, who had a fairly decent showing (111 offensive rating) against Milwaukee a week ago, typically score with an efficiency that approaches the bottom-10. The Philadelphia 76ers and Minnesota Timberwolves represent the only above-average offenses on the slate, and the latter put up a 115 offensive rating on Feb. 1.

Additionally, the Bucks have benefited from a downturn in opponent 3-point percentage:

If, as Justin Willard summarized here a few weeks ago, 3-point defense is “a noisy component that’s more akin to luck than team skill,” then the Bucks have been blessed with good fortune. Opponents have made just 28 percent of shots from downtown, the lowest rate in the NBA over the past 10 games. Indeed, it’s the lowest rate over any 10-game stretch for any team thus far this season. While the Bucks’ opponent 3-point percentage has been gradually declining since its worst period in early December, it’s highly unlikely that they’ll maintain their current level for the remainder of the year.

Nevertheless, there may still be some signal amid the noise. In a recent analysis, Dean Maniatt of Brew Hoop adjusted for opponent 3-point and free-throw percentages, then compared the Bucks’ expected and actual points allowed per 100 possessions over 10-game stretches. He found that, even after these “luck adjustments,” Milwaukee has appeared to outperform expectations. Obviously, further adjustments can be made (home-court advantage and back-to-back games come to mind), but it seems sensible enough to me that at least some tangible improvements would exist, given the rapid and pronounced reduction in defensive rating.

Where, though, does progress lie? As Zach Lowe, Kyle Wagner and many others have extensively discussed, Prunty has sought to “simplify” the Bucks’ defensive strategy, dialing back the aggressive blitzes that, in their worst incarnations, have counterproductively led to high-value shots. Most notably, they’ve begun to switch more frequently and keep their opponents in front. We don’t have publicly available data that directly shed light on schematic changes, but we can at least indirectly explore what’s happening through play-by-play logs and box scores.

Let’s start with shot locations. Overall, the Bucks have allowed a comparable rate of 3-point jumpers between the Kidd and Prunty eras: approximately a third of opponent field goal attempts. When these 3-pointers are taken, the Bucks are also generally situated in similar spots, at least as measured by the distance of their closest defender. This observation should be unsurprising. Most 3-pointers are “open,” and defender distance is a pretty crude measurement in the first place, offering no insight into angles and the placement of the rest of the defensive unit. Nonetheless, it’s worth a cursory check.

The type of 3-pointers that Milwaukee has allowed is where it gets a little interesting. Since opponent catch-and-shoot data are unavailable, we can use the frequency of 3-point shots without any preceding dribbles as a proxy. The Bucks have actually conceded these jumpers at a slightly higher rate after their coaching change. On the other hand, corner 3s have dipped a percentage point, while above-the-break 3s have inched upward by 1.6 percentage points. Altogether, the “difficulty” of opponent 3-point scoring opportunities has remained largely the same at a very surface level, and it certainly wouldn’t explain the sharp decline in opponent 3-point percentage.

How about shots in the restricted area? Well, if we examine 10-game rolling data, we can see a downward trend in the frequency of such field goal attempts near the rim.

But, as with opponent 3-pointers, restricted-area scoring opportunities have been fairly consistent through Milwaukee’s coaching change, with only a percentage-point difference between Kidd’s 45 games and Prunty’s 11 games at the helm. The field goal percentage on these shots has been practically identical, as well (roughly 63 percent).

While opponent shot profiles have held steady for the most part, the Bucks have seen encouraging signs in one key defensive area: the number of fouls that they commit and, by extension, the number of free throw attempts that they concede.

The Bucks had started to improve during Kidd’s final games and have pivoted toward the top half of the league under Prunty’s watch. Before the coaching change, they committed about 23 personal fouls and allowed 26 free throw attempts per 100 possessions, both of which were bottom-three marks. Since then, they have dropped to 19 personal fouls and 20 free throw attempts, with the latter result cracking the top-10.

Next: The 20 different emotions of Gregg Popovich

This reduction has been particularly notable given the recent success that opponents have experienced at the charity stripe. During Prunty’s tenure, they’ve made 83 percent of their free throws — about 6 percentage points higher than they did earlier in the year. But, more importantly, as the Bucks have limited their fouls, they’ve continued to force the same high number of turnovers (16 per 100 possessions). They’ve reaped the benefits of aggressive defense while lowering the costs.

We’ll see whether the Bucks can maintain their impressive play (even if at a less fluky level) as the competition stiffens and the playoffs draw near. Regardless, they seem to have brought a spiraling season under control.