A video from bballbreakdown.com made the rounds earlier in the week, even garnering a retweet from Houston Rockets general manager Daryl Morey. It professed to elucidate the viewer as to why Patrick Beverley is a “lockdown defender.”
The narrator zeroes in visually on Beverley’s tactics, underscoring a low and wide stance which he explains is a break from established fundamentals. The narrator surmises that its this unorthodox approach which is the catalyst for Beverley’s productivity.
But was Beverley actually a good defender? I remember him routinely getting destroyed, defending his man far too aggressively, and in result, getting buried under screens. I decided to dig into the numbers.
Much to my surprise, per ESPN.com’s real plus-minus stat, Beverley was #5 among point guards last season in defensive RPM, behind Chris Paul, Ricky Rubio, Kyle Lowry, and Ronnie Price. In 2014-2015, Beverley ranked 26th among point guards, at -.30, with Rubio finishing first, followed by Stephen Curry and Eric Bledsoe. And in 2013-2014, as far back as ESPN maintains these statistics, Beverley was 6th among point guards, with Bledsoe, Rubio, Jrue Holiday, and Chris Paul finishing at the head of the class. (Note: Ricky Rubio is really good at defense). So it looks like after a terrible year last season, where he was possibly still recovering from injury, Beverley bounced back to form this year. The irony there is that the Rockets were actually very good defensively in 2015, but abysmal last season.
Per basketballreference.com, with Beverley on-court this year, opponents had an offensive rating of 107.1. When he was off-court, that number was 109.4. In 2014-2015, with Beverley on-court, opponents had an offensive rating of 105.6. With Beverley off the court, that figure was 102.0. And in 2013-2014, when Beverley was on the court, opponents had an offensive rating of 105.4. With Beverley off the court, that figure was 106.7. This jibes with the RPM data over his struggles in 2015.
Digging in to NBA.com’s tracking tool, we can break things down a bit more, situationally. Overall, Beverley’s opponent’s usually shot 44%. Against Beverley, the opponent shot 43.8%. On 3’s, opponents usually shot 34.9%, but against Beverley, they shot 34.3%. On 2’s, they shot 48.2%, but against Beverley, they shot 49.5%. Here’s where things get strange: on shots less than 6 feet from the basket, Beverley’s opponents usually shot 60.6%; against Beverley, that jumped to 67.5%. On shots from less than 10 feet from the basket, Beverley’s opponents usually shot 54.4%, but that jumped to 59.4% against Beverley. The increase in percentage, however, might be a factor of the limited number of attempts per game we are dealing with.
We can really laser in on Beverley’s performance, through NBA.com’s play-type statistics. Last year, Beverley was isolated against on 52 possessions. His man shot 56.5% on these possessions, with an efficiency of 1.21 PPP, scoring 55.8% of the time. This was in the 3rd percentile among all NBA players. I can’t filter the data by position to see Beverley’s ranking against all NBA guards, and limiting the query to just other Rockets players presents the problem of situational differences: Jason Terry and Ty Lawson were typically facing backups, and other guys like Harden were facing opponents at a different position. So I’ll use Chris Paul for comparison, as he is universally regarded as an elite point guard defender. In 52 isolation possessions, Paul’s man shot 31.7%, with an efficiency of .67, with a score frequency of just 32.7%. This was good for the 85th percentile.
Beverley was posted up 50 times this year, with his man shooting 35.1%, with a 0.76 PPP, and a score frequency of 40%. This was in the 74th percentile. Paul, by comparison, was posted up 62 times, with his man shooting 43.5%, with a .92 PPP, scoring 46.8% of the time, landing him in the 38th percentile.
How about when their man was the ball-handler? In 302 possessions, Beverley’s man shot 40.2%, with a .67 PPP, and a score frequency of 32.1%. This was good for the 85th percentile. In the same situation, Paul faced 425 possessions, with his man shooting 41.4%, with a .78 PPP, and a score frequency of 37.6%. This was good for the 61st percentile.
And lastly, how about off screens? In 45 possessions, Beverley’s man shot 42.5%, with a .89 PPP, scoring 40% of the time. This was good for the 59th percentile. On the other hand, Paul found himself in 60 such situations, with his man shooting 53.2%, with a 1.32 PPP, scoring 56.7%. This was good for just the 6th percentile.
The caveat I always give in summation of weedsy evaluations like this: the aforementioned data is nowhere near comprehensive. This is the tip of the iceberg in comparison to what the Rockets and other teams have at their disposal. But what we have paints an interesting picture. The two figures that immediately stand out to me are the DRPM ranking and the productivity against isolation. The numbers confirmed the eye test in that 2015 was definitely a down year for Beverley. But they dispelled my recollection that Beverley was again ineffective this year. He was the 5th best defender among his peers this year.
On the other hand, while matching Paul’s effectiveness in all other categories, Beverley was absolutely obliterated in isolation this season. While my recollection is of him getting lost under screens, the problems in isolation seem to correlate well with selective bias: you’re more likely to retain the memory of someone getting scored on when they’re on an island, than when they just get beat backdoor. And this all seems to make sense against the broader scope of Beverley’s characteristics. Given his reputation and irritating tendencies, its likelier that the best offensive point guards are more determined to challenge and embarrass him and, looking at the data, they seem to be succeeding.
In conclusion, if we’re calling someone a “lockdown defender” for colloquial purposes involving their reputation, I’m not sure that’s an accurate description for Beverley who is in the 3rd percentile in isolation situations. He’s clearly not “locking” anyone down, in the macho connotation of the phrase. But for objective analytical purposes, he’s getting the job done overall for his team, if even in just less glamorous situations. To me, that’s really all that matters.