Harrison Barnes as a defender

#1
I recently read an article about Harrison Barnes having the Kings' worst defensive stat (with the exception of Kyle Guy). I don't remember what the stat was or what the source was. But the way Barnes is disrespected around here, I'm thinking that some of you will know what this is about and whether it's a legit issue or a bad interpretation.

For me, Barnes passes the eye test as a defender. But I'm not into advanced stats. So I recognize that he may be a weak link in some way that my eye test overlooks. Clearly he offers little in terms of shot blocking and rebounding. Could it be that this is what biases the stat (whatever the stat is)? Especially since a lot of his minutes are coming at the 4...

Seems like we have a lot invested in Barnes. So if he is a true weakness on D, then what does he bring us? (besides being a great citizen)
 
#2
A lot of the advanced stats measuring defense are still viewed with skepticism even by analytics devotees. Some of them (such as basketball references' DBPM) are based on counting stats (blocks, steals, rebounds) and this obviously paints an incomplete picture of what defense is. More interesting are the measures of opponent shooting percentage when player tracking technology identifies the player in question as the closest defender but even there you're missing the important contextual information about whether they were rotating off of another defensive assignment or navigating a pick. I suspect that the most useful quantitative data on defense is proprietary at this point so unless you're in an NBA front office or have the spare time to tabulate your own database, that is information we don't have access to as sports fans.

As a kind of short-hand and lacking an easy to use alternative, my go to move for measuring relative defense impact has been to look at the advanced stats page on basketball-reference and sort the players on the team by DRTG. My reasoning here is that within the context of the team, DRTG is telling you which players are generating more statistically measurable defensive possessions. There are some enormous caveats to this approach: the stat is skewed in favor of frontcourt players and becomes more useful the longer a player is on the floor. So you have to immediately throw out bench bigs who play limited minutes because they're invariably going to end up at the top of the list unless they're truly awful. You also have to throw out players who only get minutes in garbage time as those stats aren't really reflective of competitive basketball. And lastly, trying to compare DRTG numbers of players on different teams is asking for trouble as pace and scheme are going to skew the results differently so you're usually comparing apples and oranges.

So when you do all that, Barnes was indeed worst on the team this past season. This was a weird season with some sample size issues so I also looked at the previous season and he was also tied for last on the team the season before. Is that telling us anything useful? Like you I feel like Barnes passes the eye test as a competent defender. He doesn't make plays that generate turnovers very often but he's rotating where's he's supposed to be and moving his feet. I wouldn't expect him to measure out as the worst defender on the team, but the stats told the same story about his career before he got to Sacramento. On paper he's a player who is at best neutral and skews toward negative value on defense. To fully understand why I think you would need to study the player movement data to see how he's reacting relative to the other 9 players on the court.

My own opinion about "good" defense in the NBA is that it's not about scheme (though you can use that to hide a poor defender) and it's not about effort and athleticism (though they do help raise the baseline). Both are common go-tos for analysts but without getting into a long explanation of why, I instead think defense primarily comes down to individual players and how well they understand what the offense is doing and how quickly they react to changes. Some players are exceptionally gifted athletically but they put little or no effort into studying their opponents so their physical gifts are wasted on defense. Other players are intelligent enough to understand the nuts and bolts of opposing offenses but find themselves a step or two behind the play when they try to process it in real-time on the court. I've put a great deal of effort into analyzing players for defensive aptitude (mostly unsuccessfully) but the one thing I'm convinced of is that defense is a skill all on it's own and just like you can't teach a player to dunk like Kobe Bryant or see the floor like Steve Nash I don't think you can coach up everyone to make them good defenders in the NBA. The willingness must be there but there also needs to be a heightened level of situational awareness which is something you either have or you don't. Barnes to me is in the category of players who understand the assignments but react too slowly to generate stops consistently. I think his effort level and commitment are above average but the results are not and that's not really his fault but as a GM you're interested in results. One of the biggest ways the team could improve it's overall defensive performance is to bring in a couple of wing players who can cover multiple positions and consistently generate positive results on that end of the floor. And that probably means moving on from Barnes.
 
#3
It’s not that advanced stats point to Barnes being a bad defender necessarily. Team sports are tough to isolate 1 person’s impact. It’s that they show his teams being much more effective when he plays the 4. We essentially go from being a negative team (give up more than we score) to a positive team (score more than we give up).

I’m not a big fan - too mechanical for my liking. But HB is definitely not a problem. If we have players at the 3 and 5 that can board and make plays, HB would look pretty good at the 4.
 
#4
A lot of the advanced stats measuring defense are still viewed with skepticism even by analytics devotees. Some of them (such as basketball references' DBPM) are based on counting stats (blocks, steals, rebounds) and this obviously paints an incomplete picture of what defense is. More interesting are the measures of opponent shooting percentage when player tracking technology identifies the player in question as the closest defender but even there you're missing the important contextual information about whether they were rotating off of another defensive assignment or navigating a pick. I suspect that the most useful quantitative data on defense is proprietary at this point so unless you're in an NBA front office or have the spare time to tabulate your own database, that is information we don't have access to as sports fans.

As a kind of short-hand and lacking an easy to use alternative, my go to move for measuring relative defense impact has been to look at the advanced stats page on basketball-reference and sort the players on the team by DRTG. My reasoning here is that within the context of the team, DRTG is telling you which players are generating more statistically measurable defensive possessions. There are some enormous caveats to this approach: the stat is skewed in favor of frontcourt players and becomes more useful the longer a player is on the floor. So you have to immediately throw out bench bigs who play limited minutes because they're invariably going to end up at the top of the list unless they're truly awful. You also have to throw out players who only get minutes in garbage time as those stats aren't really reflective of competitive basketball. And lastly, trying to compare DRTG numbers of players on different teams is asking for trouble as pace and scheme are going to skew the results differently so you're usually comparing apples and oranges.

So when you do all that, Barnes was indeed worst on the team this past season. This was a weird season with some sample size issues so I also looked at the previous season and he was also tied for last on the team the season before. Is that telling us anything useful? Like you I feel like Barnes passes the eye test as a competent defender. He doesn't make plays that generate turnovers very often but he's rotating where's he's supposed to be and moving his feet. I wouldn't expect him to measure out as the worst defender on the team, but the stats told the same story about his career before he got to Sacramento. On paper he's a player who is at best neutral and skews toward negative value on defense. To fully understand why I think you would need to study the player movement data to see how he's reacting relative to the other 9 players on the court.

My own opinion about "good" defense in the NBA is that it's not about scheme (though you can use that to hide a poor defender) and it's not about effort and athleticism (though they do help raise the baseline). Both are common go-tos for analysts but without getting into a long explanation of why, I instead think defense primarily comes down to individual players and how well they understand what the offense is doing and how quickly they react to changes. Some players are exceptionally gifted athletically but they put little or no effort into studying their opponents so their physical gifts are wasted on defense. Other players are intelligent enough to understand the nuts and bolts of opposing offenses but find themselves a step or two behind the play when they try to process it in real-time on the court. I've put a great deal of effort into analyzing players for defensive aptitude (mostly unsuccessfully) but the one thing I'm convinced of is that defense is a skill all on it's own and just like you can't teach a player to dunk like Kobe Bryant or see the floor like Steve Nash I don't think you can coach up everyone to make them good defenders in the NBA. The willingness must be there but there also needs to be a heightened level of situational awareness which is something you either have or you don't. Barnes to me is in the category of players who understand the assignments but react too slowly to generate stops consistently. I think his effort level and commitment are above average but the results are not and that's not really his fault but as a GM you're interested in results. One of the biggest ways the team could improve it's overall defensive performance is to bring in a couple of wing players who can cover multiple positions and consistently generate positive results on that end of the floor. And that probably means moving on from Barnes.
I really appreciate you taking the time to share your insights on this.
I've reached the point where I think the Kings MUST improve defensively at two of three positions (the 2, 3 or 4) via draft and via trade. Unless we're doing a full rebuild. But even then, if D isn't our first priority, we end up with salary cap consumers like Bogi and Buddy. Nice when their shots are falling, but otherwise we're always at risk of giving up a 50 point quarter and looking like fools.
 
#5
It’s not that advanced stats point to Barnes being a bad defender necessarily. Team sports are tough to isolate 1 person’s impact. It’s that they show his teams being much more effective when he plays the 4. We essentially go from being a negative team (give up more than we score) to a positive team (score more than we give up).

I’m not a big fan - too mechanical for my liking. But HB is definitely not a problem. If we have players at the 3 and 5 that can board and make plays, HB would look pretty good at the 4.
And yet Barnes lack of rebounding and shot blocking would seem to be a bigger issue when he plays the 4.
 
#6
Yeah defensive metrics are difficult to gauge. You kind of just have to average them all out + use the eye test.

For instance, Barnes is 4th on the team in D RPM. Up with the likes of CoJo, Baze, and Holmes.

But D PIPM, he's the 3rd worst on the team of the guys who played regular minutes.

Barnes is a guy who gets an abnormally low amount of steals as well as a low amount of blocks for a guy his size. On the flip side, he stays in front of his man decently well and played decent team defense. He's no stand out, but he's no Hield either.
 
#9
D Fox being terrible on defense leaps off that page.

I think he has a ton of potential - but has not done a good job of maintaining fundamentals. If he quick playing risky D, I think we'd be a lot better at the point of attack.
 
#10
Did a bit of an assessment on various advanced defensive impact stats to see where Barnes ranked on our team last year:

View attachment 10222
Pretty telling why we fell off a cliff when Holmes got hurt and why things got better once Len and Bazemore came into the rotation. 6 of our most important/most used rotation players (Buddy, Bogdan, Bjelica, Barnes, Fox, Giles) grading out rather horribly showed out in the eye test too.
 
#11
Holmes making the defense better with his quick switching out to stop penetration and back in to his man was his first attribute I noticed. I did not notice Swipa being that bad on D. Wonder whats up with that?
 
Last edited:
#12
I did not notice Swipe being that bad on D. Wonder whats up with that?
One reason Joseph is such a good defender is that he really fights through screens every time and he doesnt easily lose his man in pick n roll situations, or at least he gets in a position to deny the easy pass to the screener. Fox doesnt really do that consistently, he way too often dies on the screen and it creates easy two on one situations to the opposing team.

One reason Divincenzo is a good defender is his ability and willingess to rotate correctly all the time. Fox easily has the athletisism to do that but hasnt shown it.

As an iso defender against smaller guards he is very good. His bigger role on offense might have led him to spare energy on defense on situations like fighting through screens and constantly rotating and helping others. Also playing for a bad team for his whole nba career doesnt help. Hard to say how much its lack of effort (sparing energy), lack of defensive iq or lack of film studying to truly trying to be a great defender. Lack of effort is the easiest to correct. Lack of ambition as a defender could also be fixed but lack of iq is very hard to fix.

The context of these metrics are also important. Fox suffers a little in these advanced defensive metrics since his backup is a great defender. A guy like Divincenzo performes great in said metrics since he constantly played with great defensive lineups while he never had the role of guarding the best opposing players. I would strongly believe that in a winning enviroment where defense is truly appreciated, Fox would perform a lot better on that end of the court. The question is that how much his bad defensive metrics are because of his backup, his lack of effort, lack of taking pride of being a great (team) defender or lack of iq.
 
#14
You can't take anything stat wise into account other than 5 man units with all the injuries and how terrible Walton was last year as a coach. He misused players and relied heavily on team and help defense.

Here is a list of the best 5 man defensive rating units with a 10 and 10 qualifier. You can see that sometimes things make sense for a reason. You can see there are quite a few names that appear more frequently than others in these lineups. It's no coincidence.

https://www.nba.com/stats/lineups/a...&sort=DEF_RATING&dir=-1&CF=GP*GE*10:MIN*GE*10
 
#17
Barnes covers the best wing every time he’s on the court, unless playing with Baze (maybe). Not sure if the analytics control for that. I also know that, generally, horrible defense at the 1 & 2 flows downhill fast. Our guards are really bad defensively. That forces our guys behind them to show or rotate, leading to buckets for the guys they are checking.