Defense FG%?

[MixxAOR]

Belinelli cooked him tonight. Was it a bad defense? I dont think so

[DanH]

Tucker's DFG% by month:

Month | 2PT DFG% | 3PT DFG%
November: -10.1% | +4.4%
December: -1.7% | +0.3%
January: +5.2% | +14.8%
February (pre-deadline): +4.6% | -7.6%
Since all-star: -7.6% | +1.7%

Does that really read as a stable stat over small samples? It's basically a random number generator each month.

That's not cherry picking. It's the reality of the statistic.

[DanH]

http://fansided.com/2017/01/12/nylon...trics-actions/

Yeah, and even when I've looked at rim protection stats, I've been shocked by how much variance there is among the highest 50 or so players in defended attempts - I ran a regression from year to year, and if you wanted to predict the next season's rim protection FG% for a player, you'd be better off just picking a random number close to the league average than picking their current year's DFG%. Only the top 5-10 and worst 5-10 in the league show any stability (ie players who are so good or so bad that their skill or lack thereof can overwhelm the noise inherent in these stats).

[DanH]

There were real reasons for expecting that Tucker's defence might improve in the 15 games since joining Toronto.

• The chance to play with a playoff contender
• Contract ending in two months
• Impressing coach to earn playing time
• Impressing GM before potential pay day

Most importantly, watching Tucker, he appears to be playing with tremendous intensity. Does it not seem reasonable that those things might correlate with his improved DFG%.

I suspect you know that cherry picking an arbitrary 12 game period is not a legitimate argument.

It was literally the first month I looked at. If those 12 games are not a legitimate argument, why are your 15 games?

Those who watched him in Phoenix would suggest (and I've seen them suggest, when asked by Raps fans of late) that Tucker literally always plays with this intensity, even for the mess that Phoenix was this year.

[MixxAOR]

http://fansided.com/2017/01/12/nylon...trics-actions/

On 3-pointers, there is essentially no year-to-year correlation for either the Defended FG% stat or the Difference stat. This means that a defender doesn’t really have the ability to control their Defended FG%. This is not to say that on any one particular shot attempt a defender couldn’t make the shot attempt more difficult and force a miss. It simply means that as the sample size gets larger, more randomness seeps into this particular metric, making it essentially unusable as a measure of individual defense.

But what about 2-point shots? Are these shots mostly random as well? When is it appropriate to use these statistics as a measure of individual defense?

We can answer these questions by looking at the year-to-year correlations of the Defended FG% stats displayed on NBA.com. Because NBA.com has different categories like greater than 15 feet, less than 10 feet and less than 6 feet, I focused on specific ranges: 6-10 feet, 10-15 feet and 15 feet to the three-point line.

There are a few observations to make. First, there appears to be a decent year-to-year stability in the rim protection metrics. Interestingly, there appears to be an even stronger year-to-year correlation for the <6 feet Defended FG% stat. This indicates that players who are allowing a lower field goal percentage within six feet than the shooters normally make are consistent year-to-year. Basically, good rim protectors are likely to continue to be good rim protectors assuming no health or age regression.

Outside of six feet, everything appears to be mostly random — defenders don’t appear to have any control over their Defended FG%.

[BrieflySpeaking]

There were real reasons for expecting that Tucker's defence might improve in the 15 games since joining Toronto.

• The chance to play with a playoff contender
• Contract ending in two months
• Impressing coach to earn playing time
• Impressing GM before potential pay day

Most importantly, watching Tucker, he appears to be playing with tremendous intensity. Does it not seem reasonable that those things might correlate with his improved DFG%.

I suspect you know that cherry picking an arbitrary 12 game period is not a legitimate argument.

[DanH]

That's a huge jump, his numbers in 55 games with Phoenix vs 15 games with TO. But one has to ask if there was that kind of variance during his years in Phoenix. If not, then what could have caused the change. Certainly not a sudden change in skill level. I suspect it is system reliant. He is just in a system that he is particularly constructed to excel in.

Of course there was. For example, this season, in January, Tucker posted a +8% defensive FG% differential in 12 games played. That's as "bad" as he's been "good" here in Toronto. Was Tucker useless defensively in January? Would not the disappearance of every bit of his defensive effort and skill have scared Masai off from trading for him?

Noise noise noise, people. This stat is not stable over multiple seasons, nevermind ridiculously small sample sizes.

[Puffer]

That's a huge jump, his numbers in 55 games with Phoenix vs 15 games with TO. But one has to ask if there was that kind of variance during his years in Phoenix. If not, then what could have caused the change. Certainly not a sudden change in skill level. I suspect it is system reliant. He is just in a system that he is particularly constructed to excel in.

[DanH]

Dude, super small sample. If a stat cannot remotely predict itself in the future, then it is little to no value in evaluating skill, or effort, however you want to put it. Tucker has been fantastic. Amazing. And I'd say that no matter what his DFG% showed.

As I said, it describes what happened. But not why it happened. So, if Tucker plays great defence over the next 10 games, but his opponents just hit tough shots and his DFG% differential ends up on the wrong side of zero by luck, are you going to call him out for his poor defence? Because it is definitely possible. What he's doing here is not new - Phoenix fans absolutely love the guy, he was just as much a great defender there as he is here.

[BrieflySpeaking]

Hi Dan, Thanks for the response. If you say the year-to-year DFG% numbers are volatile, I will assume they are. That makes sense since effort effects defense so much and it is easier for a player to change his effort level than his skill level.

But, if you are saying that volatilty means the numbers are unreliable and therefore misleading, I do not agree. Perhaps the year-to-year changes are reflecting the fact that players' defense actually does change considerably from season to season and from situation to situation.

Let's take P.J. Tucker is as an example.

Season	TEAM	G	DFGM	DFGA	DFG%	FG%	DIFF%
2013-14	PHX	81	3.8	8.5	44.5	45.1	-0.6
2014-15	PHX	78	4.3	9.7	44.2	44.8	-0.6
2015-16	PHX	81	4.1	9.6	43.0	44.6	-1.6
2016-17	PHX	55	4.3	9.3	46.0	46.0	-0.0
2016-17	TOR	15	2.5	6.8	37.3	44.5	-7.2

In his years in Phoenix Tucker seemed to be a slightly above average DFG% defender. Don't see much noise in those numbers. So, what's going on in Toronto?

Has Tucker suddenly become a more highly skilled defender since moving north? No.

Are the Raptor's defensive schemes so brilliant that his game has transformed? I doubt that.

So is it just statistical volatility then? Are we saying Tucker hasn't really played great defense in Toronto (like the DFG% says he has?) This is measurement error? Noise?

I am pretty sure most observers agree that Tucker really has played inspired defense for the Raptors and the excellent DFG% reflects that. As you say the numbers can't tell us why he has been so effective. Perhaps he is excited to be on a contender for the first time in his career. Perhaps, it is because this is a contract season he is trying to prove something to Raptor management.

15 games is a very small sample... a span over which many traditional statistics utterly fail. Yet, I would say Tucker's DFG% actually tells us exactly what it seems to tell us -- he is making it difficult for opponent's to score.

[DanH]

Nice post. One thing to note - a lot of it is also scheme. The Raptors scheme gives up looks to perimeter players to protect the paint, and aims to force the ball out of the hands of guards when they do trap, rather than contest. So you see a skew to bigs having good defensive FG%'s. It's probably also worth splitting the numbers to 2's and 3's, as a guard giving up 40% shooting on threes when their opponent usually shoots 45% but on long two's is not actually a good thing, and what sort of shots the team gives up more often can be a scheme thing too.

As noted above, these numbers are highly volatile from year to year and are more noise than signal (ie they tell us what happened but not why or what quality of defence led to the result), but still interesting to look at patterns on the team.

[Rudy Bargnani]

Welcome to the forum and nice first post. Lots of effort.

I think a lot of stats have flaws this one included. Notably team defense and quality of opponent can throw this off.
I would expect Patterson to have strong numbers at this--he's starter quality who ends up guarding backups as he comes off the bench on a deep team. Doesn't mean the numbers would be the same if he guarded all stars all game.

[Cody73]

So if this was the only numbers people went off of for DPOTY, Patterson should win it?

[KHD]

I think DanH has discussed these individual numbers once before with the conclusion being that they appear to be hugely volatile from year to year and thus might not really be that illuminating.

in terms of team #s, we had a debate on this board once about rim protection, and it turns out that if you look at the allowed % in the paint between best and worst it's not nearly as striking as the differences in raw number of attempts allowed. At a point we were i believe last in the league in FGA and points given up in the paint, and if we'd brought our allowed attempts to league average we'd have been in the upper half in terms of points, whereas bringing our % allowed to league average would have much less effect.

[JawsGT]

Great 1st post! welcome to the forums. I wonder how switching onto other players factors into this? IF JV, for example, is running out to challenge a 3pt shooter, does he get penalized in your analysis if the shot is made? What about Patterson, who often ends up switching or having to guard smaller players? Or Bebe, who earlier in the season was blocking alot of shots made by opposing teams guards? Just some examples that just popped into my head there. Derozan often guards the opposing teams least productive wing, but it's usually a guy that can shoot 3s and it seems the Raps system is designed to protect the paint and you can see weak side guards leaking towards the middle. Makes it a little tougher to sufficiently close out on shooters. And, of course, 3pt shooters usually shoot a little lower percentage than players that shoot in the paint or thereabouts. I'm just surprised that Demar's number look better than Powells, Lowrys or Ibaka's. Does it take into account Ibaka's time in ORL and PJ's time in PHX?