Should the "guard defense isnt very impactful" narrative here be reconsidered or finished?

[capfan33]

That seems a little disjointed.

.

Yea I was just addressing defense and while I generally agree with what you wrote about KD and his limitations, that's an inditement against him as an overall player, not on his defense specifically.

[ShotCreator]

I agree with both.

It's for sure the most valuable it's ever been, but still seems to lag behind traditional elite rim protection (by +/- data).

"Guard defense does not matter" has always felt like a bit of a cop-out and a simplification, but not a terrible one because I feel it's true that most star guards (the subjects of most of our comparisons) have historically not been that impactful either way (positively or negatively), but just by chance, not due to limitations inherent to the position. Older guys like Kidd have absolutely moved the needle there, a lot.

In guys that sell out, go high motor with the physical on-ball D, with good hands, and play that god level guard two guys at once help defensive style, it really doesn’t even lag behind.

GP2, Tony Allen, Roberson, Iguodala, Kawhi, and Paul George have hit some insanely high highs defensively. Pretty much disrupted every type of play on the court. Just mostly doing it floor bound.

[giordunk]

The fact that Rudy Gobert can get played out of games says a lot about this. There would never be a situation where a team is like okay we need a stop, let's take our best perimeter defender out of the game.

Of course, yea over the course of the thousands of possessions, Rudy Gobert brings more than Marcus Smart, but that doesn't mean a good perimeter defender isn't important.

[Ryoga Hibiki]

The fact that Rudy Gobert can get played out of games says a lot about this. There would never be a situation where a team is like okay we need a stop, let's take our best perimeter defender out of the game.

Of course, yea over the course of the thousands of possessions, Rudy Gobert brings more than Marcus Smart, but that doesn't mean a good perimeter defender isn't important.

Gobert doesn't get played out, though.
And a top perimeter defender can be made just as ineffective witha screen or with mutiple point of attacks.
Of course very different it would be to have a strong perimetere DEFENSE, made of multiple defenders.
But that defense would also help Gobert's impact.

___
Sent from my Nokia 3210 using RealGM mobile app

[DraymondGold]

I agree with a lot of what's been said before. Is guard defense unimportant? Of course not. But that still doesn't mean that individual guard defense is as important as big man defense.

As I see it, there's two reasons.
1. Even Today, the most valuable shot is at the rim (though 3 point shots have clearly become the 2nd most valuable).
Evidence: If we look at points per field goal attempt in the 2022 regular season, we find the average team's points per field goal attempt from 3 is 1.06 points per attempt, while the average team's points per field goal attempt from less than 10 feet is 1.17 points per attempt (source: https://www.nba.com/stats/teams/shots-general/?Season=2021-22&SeasonType=Regular%20Season&sort=TEAM_NAME&dir=1&GeneralRange=Less%20Than%2010%20ft). If we look at shots closer to the rim, the points per attempt would go even higher.

Takeaway: Big men who provide rim protection still protect the most valuable spot on the floor.

2. Even today, there are more attempts at the rim than any other spot on the court.

One way to think of this is defensive volume, or defensive usage. Because there's more shots at the rim than any other single spot on the court, big man have a higher defensive volume than perimeter defenders.

Per thinking basketball's study of the 2011 season, centers have the highest average defensive usage at 22.4%, almost twice as much as the average point guard defensive usage at 12.6%. (source 1: https://elgee35.wordpress.com/2011/02/06/defensive-usage-ii/, source 2: https://elgee35.wordpress.com/2011/01/22/defensive-usage/).

If we look at the top 50 Players in the NBA this season in number of field goal attempts defended per game (minimum 40 games played), we find 24/50 are big men (either true centers or center-forward hybrids), while only 14 are guards (either true guards or guard-forward hybrids). (Source: https://www.nba.com/stats/players/defense-dash-overall/?sort=D_FGA&dir=1&Season=2021-22&SeasonType=Regular%20Season&CF=GP*G*40).

A simple way to think about this is that there's only one rim, while the 3 point line is ~84 feet long (if I remember correctly?). It's a lot easier for one player to defend 1 rim -- it's a lot harder for one player to defend the full 84 foot-long perimeter of the 3 point line.

Simply put, big men defenders have a far greater defensive volume on average. When you combine the volume advantage with Point 1 Above (rim attempts are the most efficient / valuable shot in basketball), it's clear that individual big men defenders are more impactful than individual perimeter players.

...

A few qualifiers:
1. Individual big men defenders are more valuable than individual perimeter defenders on average, but that doesn't necessarily mean there aren't perimeter defender outliers who have the impact of a big man. I'd love to see how the most valuable perimeter defender stacks up!
2. Individual perimeter defense is more valuable now than ever before, and it looks like that value's only going up. I guess we'll see how high it peaks.
3. Because of switching defenses and 3 point shooting, being a poor perimeter defender is a greater liability than ever before. This becomes even more important in the playoffs, when teams hunt defensive weaknesses. This can put a dent in the resilience of individual rim protectors if they're liabilities on the perimeter. Note: the same can be true for perimeter defenders who are poor rim protectors. Players like Giannis, Embiid, and Jokic will feast on smaller defenders. Overall, versatility is at an all time premium.

An interesting followup question might be: is overall team perimeter defense more valuable than overall team rim protection? This might be a way to balance the Defensive Volume/Usage advantage that individual big men have over perimeter defenders. In other words, are all 2–3 perimeter defenders on a team more valuable than all 1–2 rim protectors? This is sort of like comparing the defensive strategies of the Bulls (who prioritize perimeter defense) vs the defensive strategies of the Bucks (who prioritize rim protection). I'm not sure which is right, but it's certainly a lot closer!

[Warriors Analyst]

My sense is that guard defense is hard to quantify. Peak Klay, for example, was great at shaving seconds off of the shot clock and forcing teams to abandon their initial actions. How quantifiable is that? I’m not sure. But he never was much of a rebounder, which is quantifiable and highly valued by the metric stats as I understand them. The flip side of that is that there’s never been a primarily off-ball shooter ala Klay, Redick, Korver, Ray, Joe Harris, or Rip Hamilton who did all that much work on the glass and that has a lot to do with role.

[DraymondGold]

My sense is that guard defense is hard to quantify. Peak Klay, for example, was great at shaving seconds off of the shot clock and forcing teams to abandon their initial actions. How quantifiable is that? I’m not sure. But he never was much of a rebounder, which is quantifiable and highly valued by the metric stats as I understand them. The flip side of that is that there’s never been a primarily off-ball shooter ala Klay, Redick, Korver, Ray, Joe Harris, or Rip Hamilton who did all that much work on the glass and that has a lot to do with role.

That's really interesting! There's definitely a lot to defense that wouldn't show up in traditional defense "Volume" stats like Ben Taylor's Defensive Usage or the NBA's Defended Field Goal Attempts (both in my previous posts). Things like forcing teams to abandon initial actions, shaving time off the shot clock (like you said), deterring rim attempts (Gobert comes to mind), deterring perimeter isolation attempts (Kawhi comes to mind), pick and roll penetration defense using positioning... all these things might not come up in traditional defensive stats, but they definitely have value.

I would have thought that this kind of stuff would still come up in plus minus data, but the defensive portion of plus minus data is notoriously noisy and hard to isolate. In the case of Peak Klay, he often ranks poorly in defensive plus minus stats. I wonder how much of that comes from the flaws of the plus minus stats, vs how much comes from Klay losing value by not rebounding, having worse off-ball defense, etc.

Perhaps in the future, we might use tracking data for better defensive metrics? Seems like it would be hard to do, and I agree that defensive role is important, but this ultimately this seems like the best route forward for accurately quantifying the value of specific defensive actions/roles.

[Warriors Analyst]

My sense is that guard defense is hard to quantify. Peak Klay, for example, was great at shaving seconds off of the shot clock and forcing teams to abandon their initial actions. How quantifiable is that? I’m not sure. But he never was much of a rebounder, which is quantifiable and highly valued by the metric stats as I understand them. The flip side of that is that there’s never been a primarily off-ball shooter ala Klay, Redick, Korver, Ray, Joe Harris, or Rip Hamilton who did all that much work on the glass and that has a lot to do with role.

That's really interesting! There's definitely a lot to defense that wouldn't show up in traditional defense "Volume" stats like Ben Taylor's Defensive Usage or the NBA's Defended Field Goal Attempts (both in my previous posts). Things like forcing teams to abandon initial actions, shaving time off the shot clock (like you said), deterring rim attempts (Gobert comes to mind), deterring perimeter isolation attempts (Kawhi comes to mind), pick and roll penetration defense using positioning... all these things might not come up in traditional defensive stats, but they definitely have value.

I would have thought that this kind of stuff would still come up in plus minus data, but the defensive portion of plus minus data is notoriously noisy and hard to isolate. In the case of Peak Klay, he often ranks poorly in defensive plus minus stats. I wonder how much of that comes from the flaws of the plus minus stats, vs how much comes from Klay losing value by not rebounding, having worse off-ball defense, etc.

Perhaps in the future, we might use tracking data for better defensive metrics? Seems like it would be hard to do, and I agree that defensive role is important, but this ultimately this seems like the best route forward for accurately quantifying the value of specific defensive actions/roles.

Gonna bat signal FNQ here from the Warriors board. He's got plenty of info about the new-age tracking data that could add a lot to this thread and answer some of your questions.

[FNQ]

My sense is that guard defense is hard to quantify. Peak Klay, for example, was great at shaving seconds off of the shot clock and forcing teams to abandon their initial actions. How quantifiable is that? I’m not sure. But he never was much of a rebounder, which is quantifiable and highly valued by the metric stats as I understand them. The flip side of that is that there’s never been a primarily off-ball shooter ala Klay, Redick, Korver, Ray, Joe Harris, or Rip Hamilton who did all that much work on the glass and that has a lot to do with role.

That's really interesting! There's definitely a lot to defense that wouldn't show up in traditional defense "Volume" stats like Ben Taylor's Defensive Usage or the NBA's Defended Field Goal Attempts (both in my previous posts). Things like forcing teams to abandon initial actions, shaving time off the shot clock (like you said), deterring rim attempts (Gobert comes to mind), deterring perimeter isolation attempts (Kawhi comes to mind), pick and roll penetration defense using positioning... all these things might not come up in traditional defensive stats, but they definitely have value.

I would have thought that this kind of stuff would still come up in plus minus data, but the defensive portion of plus minus data is notoriously noisy and hard to isolate. In the case of Peak Klay, he often ranks poorly in defensive plus minus stats. I wonder how much of that comes from the flaws of the plus minus stats, vs how much comes from Klay losing value by not rebounding, having worse off-ball defense, etc.

Perhaps in the future, we might use tracking data for better defensive metrics? Seems like it would be hard to do, and I agree that defensive role is important, but this ultimately this seems like the best route forward for accurately quantifying the value of specific defensive actions/roles.

Gonna bat signal FNQ here from the Warriors board. He's got plenty of info about the new-age tracking data that could add a lot to this thread and answer some of your questions.

We do use tracking metrics for better data! The problem is that it kinda stretches the definition of metric, because when people think of that in sports, they think of an automated process that spits out a number that defines effectiveness. The team I work with uses advanced data - think Second Spectrum - and then builds off of that. Like WA was pointing out earlier, one issue we've always had with metrics is they define a possession with the conclusion of a play, which can give false credit/blame. And while this isn't airtight, we've used 'opportunities' as a way to measure defensive effectiveness, to much better results.

viewtopic.php?f=15&t=2194287&start=380#p99160038

That's from game 1 on Luka Doncic. That's far from a complete look though, a complete look would also include how many screens were successful (a switch, or a pass/drive immediately after) as well as evaluating the screened player and the defender who's man set the screens (Curry was used to hedge on Luka only some in game1, but a ton in games 2-4, a target spot for the Warriors). We also evaluate whether or not the offensive player's actions resulted in a quality assist opportunity - which is more refined than 'potential assists' because a drive could start a flurry of ball movement that creates an open 3, but the catalyst was the driver.

On the flip side, as WA said, what if a defender stops a dribble drive, and forces a kick out? Metrics don't pick that up.. so a hypothetical: Wiggins is on Luka, no screen. Luka tries to iso. Wiggins cuts him off, traps him. Luka passes the ball to Brunson to jack up a 3 with Curry very loosely on him, 3 seconds on the shot clock. Brunson rushes a 3, missed. Metrics, even raw data tracking, award everything to Curry. His man took the shot, he was the closest defender, he gets credit. Wiggins gets nothing.

But you're absolutely right about defensive metrics - most are too noisy to be really worth the effort. Sure, they are generally accurate in the same way other metrics are, but defense needs context way, way more than offense does to really get to the closest answer.

[LAL1947]

Guard defense isn't very impactful, of course, how could it be? The Spurs defense was great all by itself because of Timmy Duncan's incredible defense. Their guards and wings (some of whom were employed purely as defensive stoppers) had naught to do with it.

I've voted for the middle option btw.

[DraymondGold]

We do use tracking metrics for better data! The problem is that it kinda stretches the definition of metric, because when people think of that in sports, they think of an automated process that spits out a number that defines effectiveness. The team I work with uses advanced data - think Second Spectrum - and then builds off of that. Like WA was pointing out earlier, one issue we've always had with metrics is they define a possession with the conclusion of a play, which can give false credit/blame. And while this isn't airtight, we've used 'opportunities' as a way to measure defensive effectiveness, to much better results.

Thanks for the info FNQ! Super fascinating stuff. And you're right that I would have thought it was an automated process, but perhaps by adding human analysis we're better able to provide context. The 3 types of passes was interesting. In your link, is a kickout pass included in the cycle passes, or are they always separate?

I'll definitely keep an eye out for more of your stuff. Thanks

Guard defense isn't very impactful, of course, how could it be? The Spurs defense was great all by itself because of Timmy Duncan's incredible defense. Their guards and wings (some of whom were employed purely as defensive stoppers) had naught to do with it.

I've voted for the middle option btw.

Hi LAL! Not sure if I missed anything, but I didn't see anyone mentioning Tim Duncan in this thread, or going as far to say that defensive dynasties should have all credit to the big man, with no credit going to guards and wings.

As far as I can tell, most people are just saying big men provide more defense on average than guards, but perimeter defense is still not worth nothing. I bet most would agree that great defensive teams require multiple positive defensive players -- as Gobert and the Jazz have shown plenty of times, no big man can do it completely alone (besides maybe peak Bill Russell haha ).

To put some hypothetical numbers to it, a great big man might be +3.5 on defense today, while a great perimeter plater might be +2. 3.5 is more than 2, but you'll need both if you want to have a great team defense above +5.

Do you disagree? If so, I'd love to hear your argument! My main points in favor of big men being more valuable on average were: 1. field goal attempts at there rim are still more valuable than at the 3 point line, and 2. big men on average defend far more shots, having a greater defensive "volume / usage".

[uberhikari]

There's a big difference between important and impactful. Guard defense is very important but it's just not that impactful overall. Or, at least, it's not anywhere near as impactful as elite rim/paint protection.

[HeartBreakKid]

Guard defense isn't very impactful, of course, how could it be? The Spurs defense was great all by itself because of Timmy Duncan's incredible defense. Their guards and wings (some of whom were employed purely as defensive stoppers) had naught to do with it.

I've voted for the middle option btw.

This is a little sad. It's 2022 and you're making passive aggressive posts about Tim Duncan very obviously because of some comparisons to Kobe Bryant lol.

Also a wing and a guard aren't the same thing...Tony Parker is not a defensive stopper.

[GSP]

Bucks went from top 5 defense w/ Jrue to bottom 10 w/ Dame................Giannis and Brook are one of best defensive frontcourts in Nba yet their defense is still ass..........

[OhayoKD]

Bucks went from top 5 defense w/ Jrue to bottom 10 w/ Dame................Giannis and Brook are one of best defensive frontcourts in Nba yet their defense is still ass..........

They went from +2 to -2 which, taken at face value, is a pretty big delta. It is not however the same as the 7 point drop off we saw between 2020 and 2021 when the Bucks added Jrue Holiday but Giannis decided, wisely, to conserve his defensive focus for the playoffs where, thanks to Giannis lineups, the Bucks turned back into an all-time historic playoff defense which they would continue to be until 2023 where Giannis, like in 2020, was not able to cover ground like he generally does due to injury, and, like 2020, the Bucks defense unceremoniously collapsed.

PS: I suspect the Bucks defense is not going to look -2 in April, May, or June

The bigs have picked up more perimiter skills to keep pace but it's very obviously still a big-ruled league: Draymond, AD, Giannis, Embid, Gobert, and JJJ are all by far the primary paint-protectors on their respective teams and are pretty clearly above the Jrues and Smarts and Carusos in terms of influence.

[OhayoKD]

KD might be an average defender for a forward, but almost everyone will say he's more valuable than elite guards on defense because of his size. Well, this is a 5v5 league and KD needs a small guard to handle the rock since he lacks the handles, mobility and creativity.

This weakness and dependence on a smaller guard on offense has now turned into a weakness on defense. You can't tell me a guy is better because he's bigger when that same guy needs to be on the floor with smaller people in order to be successful.

This just seems to me like you want to double-count KD's offensive weaknesses as defensive ones. Lebron, Giannis, and Jokic are bigger players than Durant but can all also handle the rock theoretically allowing for bigger defenders on the floor. Do you want to now give all of them additional defensive credit?

And in that case, why not give offensive credit to the likes of Draymond or Jackson for allowing Steph or Ja on the floor?

Relative to position is interesting if you do things relative to position for everyone and everything. However in practice, all it's really used to do is set a lower bar to make smaller players who aren't as good look better than they are.

Unless you are willing to curve down Curry's offense because he is a guard, you should not be curving up his defense because he "allows" better defenders to cover for him.

[oaktownwarriors87]

KD might be an average defender for a forward, but almost everyone will say he's more valuable than elite guards on defense because of his size. Well, this is a 5v5 league and KD needs a small guard to handle the rock since he lacks the handles, mobility and creativity.

This weakness and dependence on a smaller guard on offense has now turned into a weakness on defense. You can't tell me a guy is better because he's bigger when that same guy needs to be on the floor with smaller people in order to be successful.

This just seems to me like you want to double-count KD's offensive weaknesses as defensive ones. Lebron, Giannis, and Jokic are bigger players than Durant but can all also handle the rock theoretically allowing for bigger defenders on the floor. Do you want to now give all of them additional defensive credit?

And in that case, why not give offensive credit to the likes of Draymond or Jackson for allowing Steph or Ja on the floor?

Relative to position is interesting if you do things relative to position for everyone and everything. However in practice, all it's really used to do is set a lower bar to make smaller players who aren't as good look better than they are.

Unless you are willing to curve down Curry's offense because he is a guard, you should not be curving up his defense because he "allows" better defenders to cover for him.

Yes, larger players that do not need to play with PG's get additional credit... as long as it's put into practice. Pippin/Jordan didn't need a true PG and it allowed them to play bigger on the perimeter. Someone like LeBron should have taken a similar rout, but instead he has pretty much always played with a small backcourt and larger/slower front courts.

Also, Curry is in no way dependent on Green offensively. It's the reverse. Curry and Thompson working off the ball allows Green to be a play maker. I've been a major advocate for the Warriors moving away from Green because the Warriors have to run Curry into the ground just to make Green look useful on offense. The year they won the last ring Curry had the same net rating when you replaced Green with Otto Porter Jr. The defense suffered but the offense was much better.

[rrravenred]

Think a lot of it came as a reaction to the big dick-swinging narratives around wings which started with Jordan and continued with Kobe, arguing that being a lockdown defender one on one somehow meant you were a bigger alpha, etc, etc (the downgrading of D-Rob also arguably falls within that trope. He was beaten one on one by Hakeem therefore...)

I think a better understanding of team defence developed parallel with the emergence of pace and space and the diminishment of iso-ball, and the counters to that formulated by coaches, some of which even filtered down to the casual fan level. It's still absolutely a truism that bigs can cover a bigger footprint and create more "negative space" for an opposition team than any wing (Wemby being the new prototype for that),

None of this is to discount the value of preventing a heliocentric wing from utilising handles, screens, switches and handoffs to get to their sweet spots and doing terrible things, but I think there's an acceptance that defending is a whole-of-team responsibility rather than a one on one perimeter defender.

[Heej]

Lmao I made a post about something similar nearly a decade ago. Interesting to see the idea catching on. I'd expand this post to include what OP is talking about now being a point of attack defender and being able to distort teams' pet actions.

viewtopic.php?f=64&t=1306395#p38767947

[HeartBreakKid]

The fact that Rudy Gobert can get played out of games says a lot about this. There would never be a situation where a team is like okay we need a stop, let's take our best perimeter defender out of the game.

Of course, yea over the course of the thousands of possessions, Rudy Gobert brings more than Marcus Smart, but that doesn't mean a good perimeter defender isn't important.

Tony Allen has been taken out of many games and he is probably the best defensive guard ever.

There are many great perimeter defenders who are stuck on the bench throughout NBA history. Actually most defensive one way perimeter players play little minutes.

Small forwards are more protected by it because they are more valuable defenders and less important on offense (which is the same logic we eventually use to prop up centers defense), but guards? Yeah...a lot of great defensive guards are taken out of games. Why do you think Gary Payton II plays so little minutes?