Basketball IQ: How would you grade this defensive play?

[BBAL]

I see what you're saying but I don't know. In that case, why shouldn't all players just guard the guy with the ball. Clearly that is the wrong thing to do...

Everyone can't be the help. There are certain responsibilities and defensive assignments. You don't leave shooters wide open for three. In this scenario Amar'e was responsible for slowing down the penetration. If what you are saying is the way to go about it, then at the very beginning of the play, Melo, Amar'e, Fields and Novak shuold have all rushed to stand under the basketb, disregarding their responsibility of guarding their men. That's not how it works. Certain guys shuold be there to help (mostly from the weak side), and certain guys need to stick to their men who are in dangerous positions (strong side corner,). Either way it is made moot by Amare failing to slow Parker at all, because it was impossible for Fields to help, since he was on the ball side he needed to stay close to his man (at least at first, as the play was developing). Melo is a different story becasue he could have hedged more toward the paint since his man was on weak side, far away from the ball and thus a much more dangerous and difficult pass to make. He should have been in a better position to help, with a foot in the paint. Fields has to basically deny his man the pass, or close to it because that's an easy pass for Parker to make off the pick and roll.

It's like option A is the driving layup, which Amar'e was supposed to stop, option b is a pullup jumper, option c might be Manu in the corner if Fields helps too much, option d would be Duncan on the pop, and option e and F are the weakside shooters.

Oh I'm with you, and no I am at no point suggesting everyone should stay under the basket. I edited my post while you were typing so point blank, if you had to choose...

And yes Amare is definitely responsible. The question then becomes, how do you grade it. Do you credit him with:

a) a "blow by", which is a separate stat, and put the "points allowed" on the final two defenders
b) all the "points allowed" and a "blow by"
c) a "blow "by and a portion of the points allowed"

?

Which one would you pick? I see it as separate events, and believe should be graded as such.

Again, I realize it's not 100% accurate but at the same time we can't possibly account for every scenario if we were to grade a full game. I'm trying to find a good balance between accuracy and practicality that could be applied over multiple defensive plays.

Naturally in this play the simplest mistakes to detect is:

-Amare allowing his guy to get by.
-We allowed 2 points and have to figure out who gets them.

Even the current NBA offensive tracking system isn't 100% accurate. You have to make compromises.

[cgmw]

^
^

Agree wit basketboule's Novak-for-Chandler comment. If Chandler had been in the game and hypothetically blocked that shot, then Amare's defensive "grade" doesn't take a hit. I see where you're going with the grade stuff, and I appreciate the creativity, but grading individual performance within a group scheme (on a blame-only scale) is suspect.

The flip-side to this very old problem is that individually excellent defenders don't have a true metric to gauge their effectiveness in comparison to other great defenders. I believe they spend a lot of time an energy trying to solve this problem up at MIT at the Slaon Conference.

Having a bastardized metric that attempts to assign punitive blame for bad individual defense seems like an exercise valuable mainly for its healing qualities among an angry subset of frustrated fans.

I understand your concerns and you make valid points, but I disagree that it is as difficult as you say, and there's a worse system already in place.

For example, imagine that a player, Player A, successfully defends his man, in a one to man situation for 20 seconds and allows no penetration. The attacker then eventually gives up on the penetration, and the defender thus FORCES a pass. Another one of our players, player B then jumps in the players face FORCING the attacker to take a poor shot at the basket. The attacker misses, and the ball takes a lucky bounce and ends up in one of our players' hands, player C.

Currently only player C is credited with a rebound. Nobody else is credited with anything else. Player A and B get no credit for their defense even though all 3 players directly contributed to a turnover, and 1 possession won.

The +/- system is very flawed for individual contribution the simple fact that any player that gets substituted in, could play very bad, and may just simply benefit from the other 4 players going on a run, or substitutions by the other team who then play worse. Thus they get credited with + impact, when said player may have actually had a negative impact. That's why it's mainly use to determine combinations, not to track individual performances.

Some of the concerns you have would eventually even themselves out across time, plays and players.

It wouldn't be an exact science but over time, you would be able to take away two things:

- point of entry by the attacker. Eventually if you kept tracked of points allowed/forced passes, you would be able to detect which defender(s) the opponent prefers to attack
- which defender is playing the most active defense forcing the most passes, or defers. Which one of our defenders is most active.

You could compare it to an average, or average per possession.

While I'm not sold on the grade system, I'm totally sold on the value of this conversation. People have made some great points, very insightful about the defensive lapses. Enjoying the thread.

[BBAL]

I agree, at first I was disappointed but then members started pouring in and there's a lot of great information.

One thing is clear though, there NEEDS to be a defensive grading system in the NBA.

It doesn't matter if its 100% accurate as long as it's used to grade everyone. Eventually you would figure out the flaws of the system, and keep that in mind.

For example the issue with "Points Per Game" is that not everyone plays the same amount of minutes. Still a useful stat though but nowadays "Points Per Minute" is garnering more and more credibility.

Likewise defense grading needs help. Blocks and Steals alone just isn't enough. Forcing Missed Shots, is big and should be tracked. One on One defense that leads to a Forced Pass and the Shot Clock going down, is a big defensive play, and should be tracked.

No tracking system is perfect, but I applied a bare-bones version to the 76ers game and came away amazed at the results. It turned out it was easy to determine who was the most ACTIVE defenders and it wasn't even close. Some guys were getting so many more Forced Pass points, Assist turnovers and giving up much few Points Allowed compared to others. I wanted to get some additional input and plan on going back and re-grading either that game, or another game, and posting the results.

Chandler for example was high both in Points Allowed AND Activity as well as Assist turnovers because he was usually the last line of defense, thus would get scored on a lot, but also equally participated in the most missed shots, forced passes etc.

Other guys .....got very few of either but would easily give up Points Allowed.

And every one of your responses have been very helpful.

[cgmw]

^
All that's well and good. I'm more interested in knowing what it'll take for the Knicks to improve defensively.

My latest theory is that D'Antoni's emphasis on the importance of corner threes on offense makes Knick defenders hesitate more than other teams to help defend by rotating off of a man stationed around the perimeter, especially the corners.

[Purdman]

I guess i would credit Amar'e with an egregious blow by. Total fail of his responsibilities. Melo gets a minor demerit. Like I said 90-99% Amar'e, and 1-10% Melo. So 1.95 points Amare .05 melo... something like that

[airchibundo507]

I initially gave Amare the full two points, and I stand by that.

However, I don't think Fields should have helped deny penetration because he doesn't have the athleticism of Dwyane Wade (and few do). If he collapses into the middle, that's usually a WIDE OPEN three-pointer. If
Fields is lucky enough to recover in time to stop the jumpshot, that's a pump fake and easy drive to create another easy scoring opportunity. He's guarding Manu frickin Ginobili people. Give Manu the ball against a compromised defense and he will create a great shot the majority of the time.

If you really wanted to get technical, both Melo and Novak should get 0.25 apiece. They weren't watching the ball. Ideally, Melo should have helped contest the lay-up, Novak should have been in position to guard his man and Melo's man and commit to one of the two depending on if a pass is made.

But again, this happened so swiftly. The ball didn't stop AT ALL from when it was in-bounded on the side of the court. Parker just cut up the defense like a knife through butter.

[BBAL]

^
All that's well and good. I'm more interested in knowing what it'll take for the Knicks to improve defensively.

My latest theory is that D'Antoni's emphasis on the importance of corner threes on offense makes Knick defenders hesitate more than other teams to help defend by rotating off of a man stationed around the perimeter, especially the corners.

Well that's the idea. Just using a bare bones system I was at least able to determine that against the Bucks... minute per minute it turned out the most inefficient players were Harrelson, JR and Novak which was surprising even though other guys gave up more points. Bucks concentrated most of their attack on these 3 players every time they were on the court.

So you could use information like this to figure some of that out.

However I only tracked points allowed, and what they did bad. Which while somewhat useful, it left out what they did GOOD. Forcing missed shots, winning turnovers, etc....So now I wanna refine it a bit.

[albert]

The worst part about this happening every game is that Amare effectively screens Lin or whoever from being able to get back to his own man or from letting the switching rotations begin to work. It used to happen a lot with Al Harrington and Duhon.

Of course there's also that whole thing about Amare never getting into a defensive stance because he doesn't think the play has started yet.

Lol at BBIQ.

#TRADESTAT

[BBAL]

I initially gave Amare the full two points, and I stand by that.

However, I don't think Fields should have helped deny penetration because he doesn't have the athleticism of Dwyane Wade (and few do). If he collapses into the middle, that's usually a WIDE OPEN three-pointer. If
Fields is lucky enough to recover in time to stop the jumpshot, that's a pump fake and easy drive to create another easy scoring opportunity. He's guarding Manu frickin Ginobili people. Give Manu the ball against a compromised defense and he will create a great shot the majority of the time.

If you really wanted to get technical, both Melo and Novak should get 0.25 apiece. They weren't watching the ball. Ideally, Melo should have helped contest the lay-up, Novak should have been in position to guard his man and Melo's man and commit to one of the two depending on if a pass is made.

But again, this happened so swiftly. The ball didn't stop AT ALL from when it was in-bounded on the side of the court. Parker just cut up the defense like a knife through butter.

Good points. But you also need to consider that, had Parker missed his lay-up, then just as Melo/Fields got credited with "Points Allowed" because they got scored on, on the plus side, if he missed and they recovored, they would get credited with "Turnover Assists".

The would have both contributed to gaining a possession for us.

And you can't use subjective things such as the fact a player should get off the hook because of their athleticism. That's actually part of the beauty of it because it allows you to find out which defensive players....are good.

For example if a 7 footer dunked on Lin, some would argue that there is no way he could have defended that. And that's true, but for that particular play he would still be the player that allowed the points. Irrelevant of the fact he can't in reality defend it.

Over the course of 100, 1000, 10,0000 ,etc plays....you would be able to start telling apart good defenders/bad defenders because that would happen rarely. You would rarely have Lin defending a 7 footer in that situation.

It would be inaccurate to grade one or two plays. But the more you used it, the closer you get to the truth.

[BBAL]

I guess i would credit Amar'e with an egregious blow by. Total fail of his responsibilities. Melo gets a minor demerit. Like I said 90-99% Amar'e, and 1-10% Melo. So 1.95 points Amare .05 melo... something like that

I just wanna stay away from anything that involves adjustable scoring each play. You could never realistically track anything if you decided to be subjective about things and tried to determine the exact value point for each player.

In other words going by your logic, if Melo was under the basket, then he might get 0.45 and Amare would get 1.55...

Try doing that for a full 48 minutes for EACH player. That's a nightmare and would take me a month to grade one game. Not only would it be inaccurate, but in that situation you really would NEED to know the type of defensive system the coach has in place.

That's far too much detail and I'm trying to keep it objective and more standard. I'd prefer to keep it to 1 point or half a point being the lowest possible value assigned to a player.

In the event you have more than 2 players responsible then you would either dish out 3 defensive points allowed for a 2 point play, or simply chose the 2 most responsible players and give them each 1 point.

PS: And although you are 100% correct in that the quality of the opponent matters and ALL NBA stats SHOULD be adjusted for that, that's an entirely different area and right now, nobody does it. Eventually you would be able to have position vs position. So for example Centers would probably end up allowing the most points but also contributing to most blocked shots.

[airchibundo507]

I initially gave Amare the full two points, and I stand by that.

However, I don't think Fields should have helped deny penetration because he doesn't have the athleticism of Dwyane Wade (and few do). If he collapses into the middle, that's usually a WIDE OPEN three-pointer. If
Fields is lucky enough to recover in time to stop the jumpshot, that's a pump fake and easy drive to create another easy scoring opportunity. He's guarding Manu frickin Ginobili people. Give Manu the ball against a compromised defense and he will create a great shot the majority of the time.

If you really wanted to get technical, both Melo and Novak should get 0.25 apiece. They weren't watching the ball. Ideally, Melo should have helped contest the lay-up, Novak should have been in position to guard his man and Melo's man and commit to one of the two depending on if a pass is made.

But again, this happened so swiftly. The ball didn't stop AT ALL from when it was in-bounded on the side of the court. Parker just cut up the defense like a knife through butter.

Good points. But you also need to consider that, had Parker missed his lay-up, then just as Melo/Fields got credited with "Points Allowed" because they got scored on, on the plus side, if he missed and they recovored, they would get credited with "Turnover Assists".

The would have both contributed to gaining a possession for us.

And you can't use subjective things such as the fact a player should get off the hook because of their athleticism. That's actually part of the beauty of it because it allows you to find out which defensive players....are good.

For example if a 7 footer dunked on Lin, some would argue that there is no way he could have defended that. And that's true, but for that particular play he would still be the player that allowed the points. Irrelevant of the fact he can't in reality defend it.

Over the course of 100, 1000, 10,0000 ,etc plays....you would be able to start telling apart good defenders/bad defenders because that would happen rarely. You would rarely have Lin defending a 7 footer in that situation.

It would be inaccurate to grade one or two plays. But the more you used it, the closer you get to the truth.

If Fields helps, the clip ends in Manu draining a wide open three-pointer.

And I completely disagree with blaming Lin in that situation. You'd look instead at what the hell caused that match-up, whether a defensive lapse or an incompetent coach's philosophy. And with that type of a mismatch, there would have to be help defense. That should garner most of the blame in a center scoring on Lin.

Of course you want to look at how skilled the center is at scoring, where he gets the ball and how he scores, too. If he puts the ball on the floor, Lin should be pesky and make him uncomfortable by trying to poke the ball out. But it's game over if he catches the ball under the basket.

[BBAL]

I initially gave Amare the full two points, and I stand by that.

However, I don't think Fields should have helped deny penetration because he doesn't have the athleticism of Dwyane Wade (and few do). If he collapses into the middle, that's usually a WIDE OPEN three-pointer. If
Fields is lucky enough to recover in time to stop the jumpshot, that's a pump fake and easy drive to create another easy scoring opportunity. He's guarding Manu frickin Ginobili people. Give Manu the ball against a compromised defense and he will create a great shot the majority of the time.

If you really wanted to get technical, both Melo and Novak should get 0.25 apiece. They weren't watching the ball. Ideally, Melo should have helped contest the lay-up, Novak should have been in position to guard his man and Melo's man and commit to one of the two depending on if a pass is made.

But again, this happened so swiftly. The ball didn't stop AT ALL from when it was in-bounded on the side of the court. Parker just cut up the defense like a knife through butter.

Good points. But you also need to consider that, had Parker missed his lay-up, then just as Melo/Fields got credited with "Points Allowed" because they got scored on, on the plus side, if he missed and they recovored, they would get credited with "Turnover Assists".

The would have both contributed to gaining a possession for us.

And you can't use subjective things such as the fact a player should get off the hook because of their athleticism. That's actually part of the beauty of it because it allows you to find out which defensive players....are good.

For example if a 7 footer dunked on Lin, some would argue that there is no way he could have defended that. And that's true, but for that particular play he would still be the player that allowed the points. Irrelevant of the fact he can't in reality defend it.

Over the course of 100, 1000, 10,0000 ,etc plays....you would be able to start telling apart good defenders/bad defenders because that would happen rarely. You would rarely have Lin defending a 7 footer in that situation.

It would be inaccurate to grade one or two plays. But the more you used it, the closer you get to the truth.

If Fields helps, the clip ends in Manu draining a wide open three-pointer.

And I completely disagree with blaming Lin in that situation. You'd look instead at what the hell caused that match-up, whether a defensive lapse or an incompetent coach's philosophy. And with that type of a mismatch, there would have to be help defense. That should garner most of the blame in a center scoring on Lin.

Of course you want to look at how skilled the center is at scoring, where he gets the ball and how he scores, too. If he puts the ball on the floor, Lin should be pesky and make him uncomfortable by trying to poke the ball out. But it's game over if he catches the ball under the basket.

You cannot do that though. You cannot grade it based on what could have happened. Again, we would never know, and that would trigger a NEW, SEPARATE defensive play. If it happened.

I agree with pretty much everything else you said.

[Purdman]

I don't understand your point about if Lin gets dunked on by a 7 footer, he would get the 2 points allocated to him, even though he couldn't do anything about it.

Wouldn't the point of your system be trying to judge who the good and bad defenders are. So in that case, you shouldn't discredit Lin for being put in a bad matchup by either misguided switches, or necessary switches.

What would you do, in your made up system, if for example Kobe is being guarded by Shump, dribbles til the shot clock is at 4, backs down a few feet and hits a perfectly contested fade away 3 pointer. 3 points against Shumpert hardly seems fair, as this isn't his fault per se, it's just oh well what can you do perfect defense gets beat by better offense a lot of the time.

If the metric is trying to judge who is to blame for points scored, who didn't do their job, etc. you may have to factor in that frequently everybody does their job, is in the right place, helps correctly, contests the shot strong, and the guy still hits a very tough shot. This should not be a discredit to the defense.

[Purdman]

"You cannot do that though. You cannot grade it based on what could have happened. Again, we would never know, and that would trigger a NEW, SEPARATE defensive play. If it happened."

Not sure what you mean by this either. Defense, and help defense in general is basically weighing the odds. What are the odds of if I go help, it will stop them from scoring, vs. what are the odds that if I help, my man will get the pass and score.

If Fields was guarding someone who was not good at 3's, then he would be more to blame for the score in question, because the risk vs. reward for him helping would be more rewarding and less risky. But he is guarding Manu, who is a great shooter, so he correctly decides not to help, because the risk (a wide open 3 for Manu) is greater than the reward (possibly, not likley, stopping Parker)

[BBAL]

"You cannot do that though. You cannot grade it based on what could have happened. Again, we would never know, and that would trigger a NEW, SEPARATE defensive play. If it happened."

Not sure what you mean by this either. Defense, and help defense in general is basically weighing the odds. What are the odds of if I go help, it will stop them from scoring, vs. what are the odds that if I help, my man will get the pass and score.

If Fields was guarding someone who was not good at 3's, then he would be more to blame for the score in question, because the risk vs. reward for him helping would be more rewarding and less risky. But he is guarding Manu, who is a great shooter, so he correctly decides not to help, because the risk (a wide open 3 for Manu) is greater than the reward (possibly, not likley, stopping Parker)

Ok to answer both of your points, when you are talking about doing things like that you are talking about me having to be VERY subjective and VERY good at grading defenses. It would require that you take into consideration defensive schemes, switching, etc....

It's far too complex to keep track of.

And a counter argument to your Lin example about him not being paired up correctly would be the same for offense. If an offensive player that isn't a good 3 point shooter is put in a bad situation where he has to take the shot because the clock runs out, should we NOT count his missed shot? Of course not.

Again, I'm trying to strive a balance between objectivity which allows as little input from me as possible, practicality and simplicity. Something anyone could understand.

For example if I WERE to grade the first play in this thread I would use a "Blown Assignment" Grading Method.

I would grade as follows.

Amare: 1 point allowed 1 Lane Drive Allowed 0.5 Screen Defense Allowed.
Fields: 0.5 points allowed
Melo: 0.5 points allowed
Lin: 0.5 Screen Defense Allowed.

This to me is the most accurate way of grading it.

But on other plays it gets very muddy. Taking into consideration what you are saying as well as a play like this for example really complicates things.


Clearly like in the first example I would give JR the same blame I give Amare. 1 point blame, 1 lane drive allowed.

But then who exactly gets the remaining half of it?

Notice Melo, who gets out of the way of his defender and makes a really bad move....

Do you credit Amare and Fields with half a point each?
Do you credit Amare, Fields, and Melo with 1/3rd of a point?
Or do you credit Melo, with the remaining half a point because had he stepped in front of his defender he would have been in the BEST position to take on the attacking player.

Other than JR, Melo makes the biggest mistake, right? Amare, is the least to blame. All of a sudden it becomes even more complex, which is why I'm trying to keep it as simple as possible and not have to get into questions such as this. Using a "last defender method" just keeps objectivity at the forefront, which is very important, and a big part of the equation. It keeps my subjectivity to a minimum.

If we were to grade subjectively and thoroughly like you want, then YOU and I would probably come up with completely different results. I'm trying to devise a system where if we both used it, we would come up with nearly identical numbers as long as we stayed within the frame of the system. It may not be the BEST and MOST accurate, bit it would be the most objective and consistent. And that goes for everyone else that would use it.

Again, considering the fact you would be looking at the FULL 48 minutes you would have a LOT of split points, assists, etc...and over the course of the game, accuracy would go up even though a simpler system would not be entirely accurate.

[illRoso]

I'm looking forward to your results. This is a major undertaking and very interesting.

[BBAL]

I plan on it, and I have narrowed it down to 2 types of grading systems which I will let you guys decide on...and also you get to pick the game.

I want to be as objective and accurate as possible and I cannot do it without you guys critiquing my system

[Knicksfan20]

Interesting responses and some are quite enlightening. I am surprised because most people seem to choose A(perhaps due to perception against Amare?). Then again, I 'm not sure if most really understood the direction I am going with this. I personally would have gone with any option other than A.

But I want to make sure so...

a)Amare allowed 2 points and should take full responsibility for the points allowed. Because the offensive screen was set properly, Lin was blocked, and Amare became Parker's defender. Defensive roles switched. Melo and Fields could not prevent Parker from scoring. No one else bears any responsibility because once he failed to pick up Parker, nobody else could prevent the score from happening.

So most people here believe nobody else could have prevented a score from happening once Amare failed to defend Parker?

So the "points allowed" stat sheet would look like this?

Novak: 0
Lin: 0
Melo 0
Fields: 0
Amare: 2 Points Allowed, 1 "blow by"

This is how most would record it? Ask yourself this: Does Fields and Melo bear no responsibility on this play just because Amare failed?

PS: I was hoping for answers more along the lines of just A, B, C, ...etc..and I was hoping the critique would go towards the explanations/methods given rather than Amare/play in general. But I still appreciate your input.

No disrespect this being your thread and all...BUt i think you are overanlyzing a simple defensive mishap. If you think that is anybody elses fault other then Amare's then you are wrong. He probably didnt call out the pick, and he totally did nothing to slow down parker or change his direction on the pick. There is a reason why he is known as a horrible pick and roll defender and why teams attack him on a daily basis.

That was all on Amare...Yes others could have done something to stop PARKER from scoring...but in order to do that they would have had to leave their man wide open.