Is Utah’s Late-Season Defense Sustainable?

August 18th, 2015 | by Ben Dowsett
Photo by Jason Miller/Getty Images

Photo by Jason Miller/Getty Images

By now, even the most casual Jazz fan knows how incredible the team’s defense was over the latter months of the ’14-15 season. A group that was a train wreck the previous season and even early on last year turned that notion on its head for much of the 2015 calendar, going from bad to decent, good to great, and finally to briefly historically ridiculous in a span of a few short months.

Simply put, no group undergoes this sort of vast turnaround without major tangible improvements. The Jazz made some personnel tweaks, saw some individual developments in-season, and most importantly found a firm grasp of Quin Snyder’s new scheme that allowed them to play as a more cohesive unit. They began to view their lockdown defense as a point of pride.

Is their performance sustainable in the long run, though? It’s clear the Jazz have one of the top defensive foundations in the league, but it’s fair to question whether a group not even a year removed from being the worst defense in the league can transform permanently into one of the best of the last 15 years – which is exactly what the Jazz were. Utah’s 94.8 points allowed per-100-possessions post-All-Star break is a figure matched or exceeded over a full year only three individual times since 2000. All three were during the 2003-04 season, a year remembered for several historically great defenses and also the final campaign before defensive rule changes opened up the game. One could argue the Jazz’s post-ASB figure last season hasn’t been matched over a full year in the NBA as we know it.

This alone would seem to make regression a likelihood, and other bits of detail point in the same direction.

For instance, while it seems strange to call a unit as dominant as Utah’s “lucky” in any sense, they were exactly that in one significant area over the back end of the season and, coincidentally, were profoundly unlucky during the season’s first few months. Their figures for open jump-shooting – an element almost entirely outside the control of a given defense1 – showcase the swing.

For the ’14-15 season, the NBA average for all open jumpers2 beyond 10 feet from the hoop was 38.7 percent, according to But before the All-Star break, Jazz opponents shot 41.7 percent on such shots – highest in the NBA for that time period. Filtering for open 3-pointers only, the Jazz allowed 39.6 percent pre-ASB, 2nd-highest in the league and noticeably higher than a league average of 36.2 percent for the year. By this metric, the Jazz were almost certainly the “unluckiest” team in the league in this area before the break. A quick illustration here, along with their post-ASB figures:3

Jazz percentage allowed on open shots pre- and post-ASB.

Jazz percentage allowed on open shots pre- and post-ASB.

A quick comparison of the two Jazz sections reveals a stark discrepancy. Again, open shooting is an element over which defenses have little to no control, and Utah flipped from the league’s unluckiest team to one of its luckiest after the All-Star break in this regard. They did so despite allowing open shots with roughly the exact same frequency both before and after the break.

The difference may appear to be only a few percentage points, but it’s important to understand how wide a gap this actually is. The Jazz allowed 19.7 PPG on open 3s alone pre-ASB, a number that dropped all the way to 16.9 PPG after the break despite the rate of attempts remaining virtually identical. That’s a difference of nearly three points a night, all from an area the Jazz have basically no control over. This gap alone would have taken, say, Houston from 14th in the league to 22nd for the full year as far as per-game scoring allowed – and Utah’s figure would grow even larger if all open jumpers, not just 3s, were included.

In a nutshell: Utah’s defense was very unlucky on open shots against before the break, but similarly lucky during their well-known resurgence following it. Their level of effectiveness at preventing such shots underwent basically no change.

There are a couple other details that could also signal a regression. Personnel is one, particularly on the perimeter – Dante Exum’s injury removes Utah’s best defensive option at the point, and assumed health from Alec Burks and Rodney Hood seems likely to push Elijah Millsap, easily the team’s best perimeter defender, into a more complementary role. Millsap may not even see the 18 minutes per night he averaged after the break with other bodies in front of him on the depth chart, and Utah’s D was significantly worse during their strong run while he sat4. The team obviously projects to improve overall as a result, but their cumulative defensive figure should suffer slightly.

Opposing adjustments should have an effect as well, though the degree here is obviously unknown. When Rudy Gobert entered the starting lineup last season and in part initiated Utah’s seismic shift, he did so as a relative unknown for most teams. The same goes for Exum, Millsap and Rodney Hood to varying degrees. There isn’t nearly as much time mid-season for teams to log big practices and adjust schemes, especially for just a single opponent (one firmly out of the playoff picture, at that).

A full offseason will change things, as will what’s by now a well-known status as up-and-comers that should see Utah circled in darker ink on schedules in October. Gobert is a known entity at this point, one of the top big defenders in the league; teams had already begun adjusting to his presence as last season wore on, and will put even more into discovering Gobert’s kryptonite this year. The Jazz may find themselves tested by longer periods of small ball to combat Gobert and Favors’ dominance in the frontcourt. It’s a good bet the versatility we’ve heard discussed often this offseason will be tested early.

None of this is to suggest Utah’s defense will be anything but exemplary. The flip side of these regression arguments is the amount of cushion their near-legendary defense afforded them – they were so far ahead of the pack that even heavy backlash should still put the Jazz squarely in the running for the league’s top overall unit. More time and familiarity with Snyder’s system can’t hurt, and full-time starter status for Gobert is still a major net positive even if teams are more prepared for him.

Even if (when) they suffer expected regression, it could be in the name of increased efficiency on the other end of the ball. The Jazz could look to push in transition more often, which if done correctly figures to goose their early offense but may have some residual effects. Spreading most of Millsap’s minutes to Burks and Hood might loosen things a tad defensively, but the cumulative benefit should far outweigh the sacrifice.

No one is sleeping on the Jazz this year, nor would they want anyone to. They’re out to prove last year’s strong finish was no fluke, and expect them to do so even if simply dominating the current league, and not every defense in recent history, is the barometer for success.

Ben Dowsett

Ben Dowsett

Ben Dowsett is a life-long Jazz fan and current in-depth analyst based in Salt Lake City. He also writes for Basketball Insiders and BBallBreakdown, and can be heard on SCH Radio on ESPN 700 weekly. He can be found on Twitter at @Ben_Dowsett.
Ben Dowsett
Ben Dowsett

Latest posts by Ben Dowsett (see all)

  • 2015-16 Jazz SWOT Analysis
    April 22nd, 2016

    2015-16 Jazz SWOT Analysis

    SWOT analysis (Strengths, Weaknesses, Opportunities, Threats) is a popular planning method whereby an organization assesses its...Read More


  1. Mewko says:

    I don’t think the Jazz will sustain the #1 defense all 82 games, especially without Exum.

    The Jazz basically have two paths to choose from this year:

    1. Think long term, and sacrifice a few wins to try to incorporate Quin’s offensive schemes. Play Bryce Cotton because of the pace factor.

    2. Go all-in to replace Exum’s defensive value, and be the unbalanced team with super defense, mediocre offense. Reward minutes to who defends point guard the best. (Raul Neto, Elijah Millsap).

    I’m curious to see which one they choose, because it could mean playoffs or no playoffs.

  2. Sam says:

    Rushed release? Pretty tough to track, I would imagine. But anyone who has played against a tough defense knows that when you finally get an “open” look, you might be inclined to rush the open look before it closes.

    • Ben Dowsett says:

      It’s a good thought, and may well play a small role, but at the NBA level and with such talent on the floor, I don’t think any explanation like that, or even an aggregation of smaller factors like that, could really come close to explaining what’s honestly a pretty massive gap overnight. There really is very little evidence that truly open shooting is influenced by anything more than long-term variance, at least not enough evidence to truly discern anything different.

  3. IDJazzman says:

    I think the defense for the Jazz at the end of last season wasn’t as good as the rating suggests, but still tops. Some of the teams, like Denver and the Lakers just didn’t try to win, where as Quin had his players giving their all, at all times. A credit to Quin. With that said, the Jazz do have some really good defensive players. Rudy would be a top 3 rim protector in the NBA, which then enabled Favors to move from a bottom tier Center position, rim protector, to probably a top 3-4 PF rim protector. IMO, because of the agility and speed of both of these guys, small ball won’t work against the Jazz’s starting lineup, they recover to fast even against smaller players.
    I suppose if they wanted to be a top defensive team, play Millsap along with Hayward and Hood. Let Hayward and Hood bring the ball up. I don’t think there isn’t a PG in the league that Millsap could’t shut down. Don’t think Quin will do that, but just a thought. I’d say Burks, but we already know Burks doesn’t do well being a general on the floor, so Burks comes in as the 6th man.

  4. Kimball says:

    I disagree with the “luck” argument as a significant factor. Instead, I think the shift in open shot percentage had to do with the overall quality of the opponent and the replacement of Kanter with Gobert. Good teams get good shots and knock them down. Bad teams miss shots they should make and take more bad shots in general. Not all open jumpers beyond 10 feet are equal even if they can be lumped together statistically. As an example, compare a drawn up play to a knock down shooter on the Spurs/Rockets/Warriors vs. Rudy Gay/Kobe/(insert favorite chucker here) creating space on an iso for an “open” 18-footer.

    Last year our schedule was abnormally difficult early on, which, combined with Kanter, partially explains our horrible early season numbers on D. However, the schedule was abnormally soft post-ASB. So, in addition to the Kanter/Gobert factor, we were playing weaker competition on average that simply did not shoot well in general at a time our D was really surging. These factors combined for a historically great post-ASB on D. I think it was a coincidental and predictable, but not lucky or random.

    Overall, I agree that our post-ASB numbers will not be matched. However, I still fully expect our D to be at least top 5 and likely #1, though not at 94 ppg.

    • Ben Dowsett says:

      The evidence available to us just doesn’t really support what you’re saying, unfortunately – at least nowhere close to the degree that would allow it to fully explain such a massive discrepancy pre- and post-ASB. You could surely be correct that quality of opponent plays a small role (it’s something I considered also, of course), but this doesn’t really approach covering the entire gap when you aggregate over such a large sample on both sides. There were over 500 shots in the sample both pre- and post-ASB (over 1,000 pre), and these are NBA players shooting with no hand in their face.

      The reason 4 feet was chosen is the research done by guys like Seth indicating that it’s a distance at which non-track-able things (like, say, a guy with really long arms) are removed from the equation and a noticeable league-wide gap is observed in shooting percentage. So while I understand your point and agree there are surely some smaller tangential factors at play that may explain small parts of the discrepancy, again, they really come nowhere close to rationalizing the entire gap, which I should re-emphasize is fairly massive.

Leave a Reply

Your email address will not be published. Required fields are marked *