Analyzing Lebron's defensive activity via DFGAs

[Baski]

Also explain why ranking Birdman over LeBron has any relevance. Per 36 is not the way to evaluate whatever it is you're trying to evaluate.

[VanWest82]

I've posted reasons to not buy the tracking data. It doesn't tell you if the player contesting is doing it because their man shot the ball or if they were forced into a bad switch or if they're being actively attacked by an offense. Maybe offenses are avoiding Lebron in this instance. The inference in the OP from the shot contests is "Lebron isn't giving effort." Maybe it's really "defenses don't challenge Lebron and avoid him". Using a guy like RJ or Shump spending time on certain players also seems to dismiss Lebron's defense a good bit.

Ultimately using shot contests to try to extract defensive effort or ability seems pretty futile. There are about a million things that go into it that aren't in the output. Something like an APM at the least gives an idea of effect on team defense. Circling back to the OP:

Again, we know Lebron is a good defender based on eye test and the fact that he's generally had negative net Drtg on/offs over the years. But there seems to be a major disconnect between the observable data and common narrative surrounding Lebron's defensive impact, particularly in help defense. I don't doubt that he's doing things to deter good offense but the easiest way to do that is contest a shot and make them miss, and that's not something Lebron has done with any meaningful frequency.

As noted by Blackmill above, an easier way to do this is to not allow a shot, or to deter it, or again, to have a defense avoid you. Focus seems to be in the wrong spot.

Frankly the deterrent argument and the idea that teams are desperately avoiding Lebron seems pretty convenient to me. I'd have an easier time believing this if Lebron was more frequently guarding the opposing top options, and they were going away from them as a result of that. Not having their 3rd or 4th option shoot while Lebron is guarding them just sounds like good common sense, and not some grand strategy to avoid Lebron at all costs.

[RCM88x]

Now tell me how often the other teams score when LeBron is on the floor vs when he is off. Aka the only thing that matters. Why? Because the objective is to outscore your opponent.

Both this year and last year the Lakers have had a better Drtg with AD on the bench. I guess they should bench him even though he's clearly their best and most impactful defender.

No. I would say that they've simply let their opponents score less when he is on the bench. It's not predictive or explanatory, it's a "what it is stat". The Lakers hold opponents to fewer points with AD on the bench, and their opponents score more when he's playing. Defended FGAs tell litterally nothing because teams don't score based on their FGA. It's like grading defenses in the NFL based on how many passes they defend against.

[Homer38]

The defense of LBJ is not a problem since last year.Frank Vogel has said in the past that he was the QB for the lakers defense.The lakers finish 3rd in defense last year and they are first this year.The defense of LBJ is not a problem.

[bondom34]

I've posted reasons to not buy the tracking data. It doesn't tell you if the player contesting is doing it because their man shot the ball or if they were forced into a bad switch or if they're being actively attacked by an offense. Maybe offenses are avoiding Lebron in this instance. The inference in the OP from the shot contests is "Lebron isn't giving effort." Maybe it's really "defenses don't challenge Lebron and avoid him". Using a guy like RJ or Shump spending time on certain players also seems to dismiss Lebron's defense a good bit.

Ultimately using shot contests to try to extract defensive effort or ability seems pretty futile. There are about a million things that go into it that aren't in the output. Something like an APM at the least gives an idea of effect on team defense. Circling back to the OP:

Again, we know Lebron is a good defender based on eye test and the fact that he's generally had negative net Drtg on/offs over the years. But there seems to be a major disconnect between the observable data and common narrative surrounding Lebron's defensive impact, particularly in help defense. I don't doubt that he's doing things to deter good offense but the easiest way to do that is contest a shot and make them miss, and that's not something Lebron has done with any meaningful frequency.

As noted by Blackmill above, an easier way to do this is to not allow a shot, or to deter it, or again, to have a defense avoid you. Focus seems to be in the wrong spot.

Frankly the deterrent argument and the idea that teams are desperately avoiding Lebron seems pretty convenient to me. I'd have an easier time believing this if Lebron was more frequently guarding the opposing top options, and they were going away from them as a result of that. Not having their 3rd or 4th option shoot while Lebron is guarding them just sounds like good common sense, and not some grand strategy to avoid Lebron at all costs.

Frankly if we're already saying:

1. A defender doesn't actually effect the shooter's percentages very much on the perimeter.

2. Simply counting the number of shots a player defends doesn't accurately represent his defensive impact.

3. Defense is more off ball based than on ball based and

4. The best way to prevent a shot going in is to prevent a shot....

I'm not sure what the point is. Like, using basic counting and box score defensive production yields poor results. That's not something that's even far out to say, it's been a known for quite some time. And as someone who's been around here a while I think it's known I'm not a guy who goes out of his way to (or even often does) stick up for Lebron actively. This isn't even about that. It's just more or less a pretty open misuse of a statistic to try to make an argument.

Jordan was no higher than 5th on the Bulls in steals per 36 and lower in blocks per 36 on 2 of the 3 teams in the 2nd 3 peat. By this same reasoning one could try to concoct an arguement his defense wasn't really that good...but that'd be a pretty poor way to phrase it.

I'm not even in the thread to make a point on Lebron's defense, more a larger point that this is a pretty general misframing of a statistical case when we've got a lot of evidence to the contrary.

[PistolPeteJR]

OP, you’ve been given a multitude of reasons from multiple, objective posters, and you still want to hold to the fact that your research is the be-all.

I’m all for debating and challenging in a healthy, open-minded way, but at some point, when faced with all of this objective, substantial opposition, I encourage you to take a step back and try to reframe your perspective. You might find that you’ll receive what is being said differently.

[VanWest82]

Frankly if we're already saying:

1. A defender doesn't actually effect the shooter's percentages very much on the perimeter.

I'm not saying that. If defenders don't affect shooting percentages on the perimeter then what's the point of even contesting? Have to wonder why defenses don't just stop doing that if it doesn't matter.

2. Simply counting the number of shots a player defends doesn't accurately represent his defensive impact.

Never said this. I said it's good evidence that Lebron's defensive activity is low.

3. Defense is more off ball based than on ball based and

For Lebron it is. For other guys who are asked to defend the POA it's more on ball based, or even just guys who exert more effort in help defense.

4. The best way to prevent a shot going in is to prevent a shot....

I'm not sure what the point is. Like, using basic counting and box score defensive production yields poor results. That's not something that's even far out to say, it's been a known for quite some time.

Again, how can the results be poor if it's just a simple measurement? When they measure a shot against it's just a shot against. When they measure a deflection it's just a deflection. RAPM yields poor results. RPM yields poor results. Those stats are trying to do something incredibly ambitious and failing miserably at it. Using tracking data to look at the specific activity it's measuring is a perfectly fine use for it.

[bondom34]

I'm not saying that. If defenders don't affect shooting percentages on the perimeter then what's the point of even contesting?

See my first post in the thread. That's what I was talking about, we know that a player's defensive FG% against doesn't mean much. So I'm not sure what this is telling us either.

Never said this. I said it's good evidence that Lebron's defensive activity is low.

By which I don't know what is inferred. Simply running around aimlessly isn't accomplishing anything. Knowing positioning/where to be/when to be there, and timing it best is ideal defensively.

For Lebron it is. For other guys who are asked to defend the POA it's more on ball based.

I'd say in general it is. There's 1 ball on the court and 5 defenders. Think this was a Ben Taylor quote but in general what's overrated the most by fans seems to be on ball offense and on ball defense, because it creates highlights. But it's overall a relatively small portion of the game.

Again, how can the results be poor if it's just a simple measurement? When they measure a shot against it's just a shot against. When they measure a deflection it's just a deflection. RAPM yields poor results. RPM yields poor results. Those stats are trying to do something incredibly ambitious and failing miserably at it. Using tracking data to look at the specific activity it's measuring is a perfectly fine use for it.

Tracking yields pretty poor results, so do steals and blocks. I'm not sure what you're saying by "RAPM yields poor results" when it's in general shown pretty decent results. Far better than whatever tracking has managed, especially in measuring actual impact on games. Giannis was 3rd on the Bucks last year in steals/36 and in DFGA/36, he remained 1st in 3 year DRAPM and was clearly the best defender in the NBA and on his own team. Using the box score to try to measure defense is just not a way to go, this isn't even a stats type take overall, its been pretty commonly known for a while.

Just counting steals/blocks, and in this case tracking stuff isn't terribly informative in terms of measuring actual impact on a game. And again given that preventing a shot would actually make this look "worse" for a player but be an unambiguously better outcome for the defense, I don't see the point.

If a player defends a shot he's "active", if he prevents the opponent from shooting he's "not as active" but its absolutely a better outcome.

Edit: Here's a quick glance at a team level of the teams who defend the most FGA/game:



These teams defensive ranks:

25th
14th
12th
30th
2nd
9th
18th

Just a glance at the top 7 but they include 2 bottom 5 defenses, and the team who's defending the most shots per game is actually the worst defense. I'm just saying...I'm not sure this data is actually informing us of anything regarding anyone's defense.

[VanWest82]

Lebron's not preventing the opponent from shooting much of anything from the other side of the court. He's defending back cuts. That's just a red herring.

[bondom34]

Lebron's not preventing the opponent from shooting anything from the other side of the court on many possessions. That's just a red herring.

See my edit but I'll leave it there. I think enough other folks have chimed in, but would recommend further looks into other measures of defensive impact (3 year LA-RAPM has been what I've looked into lately, and generally seems pretty reasonable). I'd generally trust it as a decent guideline at least. I wouldn't touch using tracking data.

I think you know I'm not really someone who capes for Lebron (respect him a ton, he's amazing, but not someone I generally go out of my way to defend). Just don't care for this framing. Again I'll leave it to others.

[VanWest82]

Lebron's not preventing the opponent from shooting anything from the other side of the court on many possessions. That's just a red herring.

See my edit but I'll leave it there. I think enough other folks have chimed in, but would recommend further looks into other measures of defensive impact (3 year LA-RAPM has been what I've looked into lately, and generally seems pretty reasonable). FWIW, the top 10 from 2017-20:

Spoiler:

Giannis, Gobert, Brook Lopez, Draymond, Covington, Nurkic, Embiid, Zubac, Myles Turner, Rubio

I'd generally trust it as a decent guideline at least. I wouldn't touch using tracking data.

I think you know I'm not really someone who capes for Lebron (respect him a ton, he's amazing, but not someone I generally go out of my way to defend). Just don't care for this framing. Again I'll leave it to others.

I wouldn't use 3 year RAPM outside of a discussion about players over that specific rolling 3 year period, and only if the topic is very general in nature. Without considering game specific events you're admitting that you're just guessing.

You're trying to take something like DFGA and correlate with Drtg. That's completely missing the point. DFGA isn't trying to say anything that big, just who was actually finishing the possessions. Nothing more. It's meant to provide literal context to what's happening in the games, so we can better describe what's happening, like that Lebron's defending less shots.

DFGA per 36: Siakam 13.3, Giannis 12.5, Lebron 9.6.

Three guys with roughly the same off ball forward defensive role. Which guys would you say have greater defensive usage if that were a thing? Do you think usage isn't important? All that off ball stuff is important on offense too but ultimately it's about making the ball go in the hole (or preventing it). Lebron's teammates do more of the latter than Siakam's or Giannis's. The idea that guys are seeking out and going at Giannis and Siakam but not Lebron because he's a deterrent and they're not is humorous to say the least.

Anyway, I probably should've thought a little harder about the title. It would seem people all implicitly tried to correlate DFGA and defensive impact broadly which is not what I meant at all.
.

[bondom34]

Spoiler:

Lebron's not preventing the opponent from shooting anything from the other side of the court on many possessions. That's just a red herring.

See my edit but I'll leave it there. I think enough other folks have chimed in, but would recommend further looks into other measures of defensive impact (3 year LA-RAPM has been what I've looked into lately, and generally seems pretty reasonable). FWIW, the top 10 from 2017-20:

[spoiler]Giannis, Gobert, Brook Lopez, Draymond, Covington, Nurkic, Embiid, Zubac, Myles Turner, Rubio

I'd generally trust it as a decent guideline at least. I wouldn't touch using tracking data.

I think you know I'm not really someone who capes for Lebron (respect him a ton, he's amazing, but not someone I generally go out of my way to defend). Just don't care for this framing. Again I'll leave it to others.

I wouldn't use 3 year RAPM outside of a discussion about players over that specific rolling 3 year period, and only if the topic is very general in nature. Without considering game specific events you're admitting that you're just guessing.

You're trying to take something like DFGA and correlate with Drtg. That's completely missing the point. DFGA isn't trying to say anything that big, just who was actually finishing the possessions. Nothing more. It's meant to provide literal context to what's happening in the games, so we can better describe what's happening, like that Lebron's activity level on defense might be low.

DFGA per 36: Siakam 13.3, Giannis 12.5, Lebron 9.6.

Three guys with roughly the same off ball forward defensive role. Which guys would you say have greater defensive usage if that were a thing? Do you think usage isn't important? All that off ball stuff is important on offense too but ultimately it's about making the ball go in the hole (or preventing it). Lebron's teammates do more of the latter than Siakam's or Giannis's. The idea that guys are seeking out and going at Giannis and Siakam but not Lebron because he's a deterrent and they're not is humorous to say the least.

Anyway, I probably should've thought a little harder about the title. It would seem people all implicitly tried to correlate DFGA and defensive impact broadly which is not what I meant at all.
.

[/spoiler]

Edit Later too: To be clear, I don't think RAPM (or any measure) is perfect, just that it's better at measuring impact than something like defended FGA, which is what I was trying to get at before this went on so long. It has some flaws, and some hits and misses, but I don't see DFGA as measuring impact which I guess was the issue. IDK what it really measures overall, and even with flaws and some holes with certain player types, I'd side more toward a longer term APM if possible, and don't buy into the tracking data for this use much at least as of now. RAPM is far from perfect, and has some big flaws if not looked at with context and regard to role, and has some wonky-ness with younger players. I'm pretty open that I don't know everything, nor is any measure perfect, just think the framing was off, or as you said:

Anyway, I probably should've thought a little harder about the title. It would seem people all implicitly tried to correlate DFGA and defensive impact broadly which is not what I meant at all.
.

So like you said, the title may have been the biggest issue too.


1. Defensive usage is a thing measured, sort of:

https://apanalytics.shinyapps.io/knarsu3/

And of those 3...Lebron actually spent the most time on an opponents 1st option I believe.




2. Probably should have phrased some stuff differently myself. There is value in contesting shots I'd assume, but the value at an individual level and using it as a measure of anything seems futile is I guess a way to put it. And knowing that at a team level it's not correlating to defense, I don't know why it matters a ton. I probably should have stated some things differently as well, but if the ultimate concern isn't how they're impacting the game (and the tool above seems to measure what's trying to be inferred), I just guess I don't see the point.

If a shot being contest statistically seems largely variance based in terms of it going in or not (and again this is something I've tried to ask and gotten clarification from someone who worked in an NBA analytics office, not saying it has no value but we know the percentages are generally fairly variant) it just seems to not really be overly informative at least in terms of non-rim attempts IMO. Could just be people putting up a token hand in the air. This seems like a way to try to frame an argument against a player by using a metric that doesn't measure anything meaningful. And a "defended FG" can mean a lot of different things, which is further why I don't like the data.

Ultimately, the number of shots a player defends is highly dependent on so many variables outside the control of them and has such little meaningful impact on overall defense it's not terribly informative. To add, looking at 3 year RAPM is actually a measure of defensive impact. I just struggle to think that not using something measuring impact but looking to measure whatever this is is sort of looking at everything entirely backward. .

Edit: Changed some formatting...But I'll drop it there again, if you want to continue with others go for it. Don't really care to post further, and not even so much was posting to talk Lebron, but the use of the measure. Will let it for others, but wasn't really a Lebron specific thing to me.

Just realized I could have said "if we're not really concerned about overall impact, I'm not sure the point" lol, and the title was what threw that off, as this isn't a measure of any sort of impact really. But think that tool might be what is being searched for. Guess the title was my issue, and how the number is maybe being portrayed.

And to be clear, APM/RAPM isn't perfect either, but I'd trust it over this to measure impact. Nothing is a perfect metric, and it has its flaws too. Didn't want to say otherwise.

[freethedevil]

Now tell me how often the other teams score when LeBron is on the floor vs when he is off. Aka the only thing that matters. Why? Because the objective is to outscore your opponent.

Both this year and last year the Lakers have had a better Drtg with AD on the bench. I guess they should bench him even though he's clearly their best and most impactful defender.

No. I would say that they've simply let their opponents score less when he is on the bench. It's not predictive or explanatory, it's a "what it is stat". The Lakers hold opponents to fewer points with AD on the bench, and their opponents score more when he's playing. Defended FGAs tell litterally nothing because teams don't score based on their FGA. It's like grading defenses in the NFL based on how many passes they defend against.

This discussion si a great example why plus minus is vastly more effective when you....adjust for lineups.

[No-more-rings]

I always thought Lebron had more impact coming from his help defense than man to man though. I think his reputation as this man to man lock down defender gets overblown. People bring up Derrick Rose(2011), Tony Parker(2013), Steph Curry(16), but re-watching tape it's clear these were very much team efforts that Lebron perhaps led the way on. Then there's back in 2012 vs KD, but KD actually scored very efficiently he just couldn't do anything else at a passable level. Lebron also sort of gets superstar treatment on defense too, he gets away with more fouls than your average player which is fine but it bears mentioning when comparing him to defensive specialists.

[DQuinn1575]

I'm not saying that. If defenders don't affect shooting percentages on the perimeter then what's the point of even contesting?

See my first post in the thread. That's what I was talking about, we know that a player's defensive FG% against doesn't mean much. So I'm not sure what this is telling us either.

Never said this. I said it's good evidence that Lebron's defensive activity is low.

By which I don't know what is inferred. Simply running around aimlessly isn't accomplishing anything. Knowing positioning/where to be/when to be there, and timing it best is ideal defensively.

For Lebron it is. For other guys who are asked to defend the POA it's more on ball based.

I'd say in general it is. There's 1 ball on the court and 5 defenders. Think this was a Ben Taylor quote but in general what's overrated the most by fans seems to be on ball offense and on ball defense, because it creates highlights. But it's overall a relatively small portion of the game.

Again, how can the results be poor if it's just a simple measurement? When they measure a shot against it's just a shot against. When they measure a deflection it's just a deflection. RAPM yields poor results. RPM yields poor results. Those stats are trying to do something incredibly ambitious and failing miserably at it. Using tracking data to look at the specific activity it's measuring is a perfectly fine use for it.

Tracking yields pretty poor results, so do steals and blocks. I'm not sure what you're saying by "RAPM yields poor results" when it's in general shown pretty decent results. Far better than whatever tracking has managed, especially in measuring actual impact on games. Giannis was 3rd on the Bucks last year in steals/36 and in DFGA/36, he remained 1st in 3 year DRAPM and was clearly the best defender in the NBA and on his own team. Using the box score to try to measure defense is just not a way to go, this isn't even a stats type take overall, its been pretty commonly known for a while.

Just counting steals/blocks, and in this case tracking stuff isn't terribly informative in terms of measuring actual impact on a game. And again given that preventing a shot would actually make this look "worse" for a player but be an unambiguously better outcome for the defense, I don't see the point.

If a player defends a shot he's "active", if he prevents the opponent from shooting he's "not as active" but its absolutely a better outcome.

Edit: Here's a quick glance at a team level of the teams who defend the most FGA/game:



These teams defensive ranks:

25th
14th
12th
30th
2nd
9th
18th

Just a glance at the top 7 but they include 2 bottom 5 defenses, and the team who's defending the most shots per game is actually the worst defense. I'm just saying...I'm not sure this data is actually informing us of anything regarding anyone's defense.

I think I’m missing something as it looks like you just pulled def field goals attempts per game. That tells you very little about a defense. If you want to look to test the theory presented, you should look at total field goals defended divided by field goals against and see how that compares to defensive rating. If the correlation is high, the the OP has some valid questions. If the result is noise, then we would draw a no conclusion.
I’d be interested in someone doing this, to see if the numbers follow all of the comments.

[bondom34]

I'm not saying that. If defenders don't affect shooting percentages on the perimeter then what's the point of even contesting?

See my first post in the thread. That's what I was talking about, we know that a player's defensive FG% against doesn't mean much. So I'm not sure what this is telling us either.

Never said this. I said it's good evidence that Lebron's defensive activity is low.

By which I don't know what is inferred. Simply running around aimlessly isn't accomplishing anything. Knowing positioning/where to be/when to be there, and timing it best is ideal defensively.

For Lebron it is. For other guys who are asked to defend the POA it's more on ball based.

I'd say in general it is. There's 1 ball on the court and 5 defenders. Think this was a Ben Taylor quote but in general what's overrated the most by fans seems to be on ball offense and on ball defense, because it creates highlights. But it's overall a relatively small portion of the game.

Again, how can the results be poor if it's just a simple measurement? When they measure a shot against it's just a shot against. When they measure a deflection it's just a deflection. RAPM yields poor results. RPM yields poor results. Those stats are trying to do something incredibly ambitious and failing miserably at it. Using tracking data to look at the specific activity it's measuring is a perfectly fine use for it.

Tracking yields pretty poor results, so do steals and blocks. I'm not sure what you're saying by "RAPM yields poor results" when it's in general shown pretty decent results. Far better than whatever tracking has managed, especially in measuring actual impact on games. Giannis was 3rd on the Bucks last year in steals/36 and in DFGA/36, he remained 1st in 3 year DRAPM and was clearly the best defender in the NBA and on his own team. Using the box score to try to measure defense is just not a way to go, this isn't even a stats type take overall, its been pretty commonly known for a while.

Just counting steals/blocks, and in this case tracking stuff isn't terribly informative in terms of measuring actual impact on a game. And again given that preventing a shot would actually make this look "worse" for a player but be an unambiguously better outcome for the defense, I don't see the point.

If a player defends a shot he's "active", if he prevents the opponent from shooting he's "not as active" but its absolutely a better outcome.

Edit: Here's a quick glance at a team level of the teams who defend the most FGA/game:



These teams defensive ranks:

25th
14th
12th
30th
2nd
9th
18th

Just a glance at the top 7 but they include 2 bottom 5 defenses, and the team who's defending the most shots per game is actually the worst defense. I'm just saying...I'm not sure this data is actually informing us of anything regarding anyone's defense.

I think I’m missing something as it looks like you just pulled def field goals attempts per game. That tells you very little about a defense. If you want to look to test the theory presented, you should look at total field goals defended divided by field goals against and see how that compares to defensive rating. If the correlation is high, the the OP has some valid questions. If the result is noise, then we would draw a no conclusion.
I’d be interested in someone doing this, to see if the numbers follow all of the comments.

Nope, was pulling from the defense dashboard:

https://www.nba.com/stats/teams/defense-dash-overall/?sort=D_FGA&dir=1

The list of teams is similar, though not identical to a team's opponent FGA. As well, from Partnow's stuff it seems like the correlation of an individual's DFG% to anything is pretty random.

https://www.nba.com/stats/teams/opponent/?sort=OPP_FGA&dir=1

Bottom here is still kinda similar, but different values and ordered Nets, Thunder, Jazz, Bucks, Kings, Hawks, Spurs, Warriors (similar teams, different order). It's not really indicating anything at that level either.

[whitehops]

a big part of lebron's value as a defender (especially in his advanced age) is essentially playing the role of a big, just as a wing. normally because bigs are inside and the plays are unfolding in front of them, they're in the best position to call screens, coverages, switches, etc. lebron tries to play off-ball as much as possible for that reason (as well as conserving energy), and with his IQ it's a great role for him.

here's one of my favourite examples of this:

https://gfycat.com/malethatalbino

Gfycat Video - Click to Play

he recognizes the set, calls for a switch BEFORE the play happens to get better matchups, while it's happening calls the switches so the cavs have optimal matchups the entire play. derozan made the shot but it was literally the only thing available to the raps and derozan still had to take a tough floater contested by thompson and lebron.

lebron wouldn't have been able to call any of those things out if he started the play on derozan, or at least he wouldn't have been able to see what was happening as soon as he did when he was off ball.

[feyki]

Look at Lebron's 2016 Finals and Rox series last year, they're very impressive.

[VanWest82]

Siakam per 36 (20 & 21)

3 pointers: 7.1 DFGA, 32.3 DFG%; 6.6 DFGA, 40.8 DFG%
2 pointers: 7.4 DFGA, 48.5 DFG%; 6.6 DFGA, 51.0 DFG%
< 10 feet: 5.1 DFGA, 51.7 DFG%; 4.7 DFGA, 54.1 DFG%

Giannis per 36 (20 & 21)

3 pointers: 5.3 DFGA, 31.1 DFG%; 5.1 DFGA, 41.4 DFG%
2 pointers: 7.1 DFGA, 40.0 DFG%; 7.3 DFGA, 48.1 DFG%
< 10 feet: 5.0 DFGA, 40.7% DFG%; 4.5 DFGA, 56.0 DFG%

Lebron per 36 (20 & 21)

3 pointers: 3.3 DFGA, 31.9 DFG%; 4.0 DFGA, 27.6 DFG%
2 pointers: 4.9 DFGA, 53.4 DFG%; 5.7 DFGA, 49.7 DFG%
< 10 feet: 3.4 DFGA, 59.8 DFG%; 4.0 DFGA, 57.1 DFG%

I like using Siakam and Giannis because I've watched so many games for both over the last two years that I can say some things comfortably without needing much data support, like neither guy is the same player defensively this year. Tracking data matches that. Siakam is actually contesting less in addition to having poorer results. He's a step slow which is what I would've said before looking. Giannis is contesting about the same but he's not playing with the same F U edge defensively, and is similarly getting beat more. Both those guys have perhaps been a little unlucky vs. threes but the fact their DFG% increased across the board is telling. I bet if we got to see average distance away on those contests they're a little further out this year.

Lebron is the only one playing better defensively this year as he's had to do more with AD coming in out of shape and playing with the bad Achilles. His DFGA have not only gone up, but because he's playing with more energy on that end his success rate has gone up too, though he's probably been a little lucky on those threes.

The fact that in both years Siakam and Giannis challenged more 3 pointers and closer range shots than Lebron in either year highly suggests their level of defensive activity is greater. And Again, it's not that you can't have an impact being a weakside guy who doesn't get challenged as much but helps QB the defense even though he's not going out of his way to help challenge shots, but there's a limit on how much credit you can get when you're physically doing less. It's one of the reasons I would've voted AD last year for DPOY over Giannis. He did more / had a bigger role.





All this graph shows is Siakam, Giannis, and Lebron have essentially the same defensive role on their teams (Siakam and Giannis spend a little more time on the top 2 options than Lebron, but mostly all three guys are off ball help defenders). The fact they all have the same role but that Siakam and Giannis defend way more shots suggests they're way more active in help defense.

[VanWest82]

Speaking of AD and his drop off defensively due to injuries / being out of shape, here are his 20 & 21 splits (per 36):

3 pointers: 5.8 DFGA, 30.5 DFG%; 5.8 DFGA, 31.6 DFG%
2 pointers: 9.5 DFGA, 43.0 DFG% 10.3 DFGA, 52.9 DFG%
< 10 feet: 6.7 DFGA, 47.2 DFG%; 7.6 DFGA, 59.6 DFG%

His role is still the same and so he's been mostly defending the same shots, but the impact hasn't been there because he's not as explosive. Dismissing this stuff out of hand because it doesn't correlate well enough with Drtg is wrong. Also, we do know that opponent FG% at the team level is correlated with Drtg.

It would seem logical that for high level usage guys defensively, their volume and efficiency would correlate some with on court Drtgs assuming similar level play from their teammates (big assumption) (e.g. KAT = high volume, low efficiency, bad Drtg vs Gobert/Turner = high volume, high efficiency, good Drtg). Trying to draw lines between low volume guys is a tougher road. Yes they can help, and Lebron is perhaps the best version of this low usage defensive type, but the ceiling for defensive impact is undoubtedly less when you're involved in the play less.