Why are Anthony Edwards impact metrics so poor relative to his hype?

[Chessboxer]

I'll say this. The refs do NOT give Ant the whistle. It amazes me and that hurts his efficiency. Also, he works on his game in the regular season. He tried so many different shots like banks from 3 and left handed shooting. He really worked on his midrange and started out poorly. He'll get there but I give him credit for trying hard to improve. Now he's got to figure out the referees. Last nights game is a perfect example.

As far as the refs, I think its an age thing, hes only 22. You generally do not see guys his age getting the benefit of the whistle. Its either very young players, or older legacy superstars that dont get the calls they used to. But I expect that to change soon, maybe even next year as he continues to emerge as as superstar.

[Godymas]

what Caleb Martin did vs. Boston is a heater

what Ant is doing is Ant, he's elevated in the playoffs 3 years in a row now and he's only 22.

Yeah, but 16 games. So that's quite a noisy sample, and the numbers he's posted are all heavily contingent upon him shooting SIGNIFICANTLY better from the floor than he has over a much larger sample in the regular season. So some skepticism is fairly normal.

you seem to want to discredit Ant for whatever reason. I don't agree with you, there is no "we", it's just you and your opinion.

No, you just aren't listening because you've already decided, which is common enough. You immediately jumping to this idea of "discrediting" the player is just an indication that you aren't interested in reasonable discourse.

I am not trying to discredit Ant, I've only said it's worth being patient. His shooting numbers aren't likely to sustain at precisely this level because of 3/4s of a century of sample showcasing that, and basic probability. To what level they will regress and when, that's another story. If you actually read what I've written, that would be clearer. Maybe don't just instantly jump to a conclusion when someone isn't saying everything exactly as you want to hear it.

16 is enough in statistics, take a community college class

Strike: Trolling/Baiting

[tsherkin]

16 is enough in statistics, take a community college class

That... isn't actually true.

But, thank you for making clear the nature of this conversation and your participation. Have a good one.

[Dirk]

what Caleb Martin did vs. Boston is a heater

what Ant is doing is Ant, he's elevated in the playoffs 3 years in a row now and he's only 22.

Yeah, but 16 games. So that's quite a noisy sample, and the numbers he's posted are all heavily contingent upon him shooting SIGNIFICANTLY better from the floor than he has over a much larger sample in the regular season. So some skepticism is fairly normal.

you seem to want to discredit Ant for whatever reason. I don't agree with you, there is no "we", it's just you and your opinion.

No, you just aren't listening because you've already decided, which is common enough. You immediately jumping to this idea of "discrediting" the player is just an indication that you aren't interested in reasonable discourse.

I am not trying to discredit Ant, I've only said it's worth being patient. His shooting numbers aren't likely to sustain at precisely this level because of 3/4s of a century of sample showcasing that, and basic probability. To what level they will regress and when, that's another story. If you actually read what I've written, that would be clearer. Maybe don't just instantly jump to a conclusion when someone isn't saying everything exactly as you want to hear it.

16 is enough in statistics, take a community college class

If someone has the trouble of articulating things back to you, the least you could do was show some respect. You don't have to agree, but you can be cordial.

[Godymas]

16 is enough in statistics, take a community college class

That... isn't actually true.

But, thank you for making clear the nature of this conversation and your participation. Have a good one.

you don’t explain why something isn’t true

https://www.statisticshowto.com/large-enough-sample-condition/

good bye forever

[Godymas]

Yeah, but 16 games. So that's quite a noisy sample, and the numbers he's posted are all heavily contingent upon him shooting SIGNIFICANTLY better from the floor than he has over a much larger sample in the regular season. So some skepticism is fairly normal.



No, you just aren't listening because you've already decided, which is common enough. You immediately jumping to this idea of "discrediting" the player is just an indication that you aren't interested in reasonable discourse.

I am not trying to discredit Ant, I've only said it's worth being patient. His shooting numbers aren't likely to sustain at precisely this level because of 3/4s of a century of sample showcasing that, and basic probability. To what level they will regress and when, that's another story. If you actually read what I've written, that would be clearer. Maybe don't just instantly jump to a conclusion when someone isn't saying everything exactly as you want to hear it.

16 is enough in statistics, take a community college class

If someone has the trouble of articulating things back to you, the least you could do was show some respect. You don't have to agree, but you can be cordial.

https://www.statisticshowto.com/large-enough-sample-condition/

their fundamental knowledge is flawed and they are trying to strong arm their way to win an argument, to me that does not warrant meaningful respect and that is my respectful opinion on the interaction.

[tsherkin]

16 is enough in statistics, take a community college class

That... isn't actually true.

But, thank you for making clear the nature of this conversation and your participation. Have a good one.

you don’t explain why something isn’t true

good bye forever

I'm not going to teach you basic math, friend. Perhaps take your own advice? The burden here is on you to explain why 16 games is a meaningful sample, especially in the context of visibly-unsustainable shooting percentages with the weight of nearly a century of basketball behind them (even factoring in modern trends).

16 games is an inherently noisy sample. It's the same line you hear every year at the beginning of the season when someone is championing something happening over the first 10 or 20 games of the season, before that eventually evens out. Same same with a single playoff run. It's something to be viewed with at least SOME skepticism until and unless it is repeated. And especially so when there are some eye-popping numbers so discordant with RS production (and league/personal averages).

In any case, take care.

[Godymas]

That... isn't actually true.

But, thank you for making clear the nature of this conversation and your participation. Have a good one.

you don’t explain why something isn’t true

good bye forever

I'm not going to teach you basic math, friend. Perhaps take your own advice? The burden here is on you to explain why 16 games is a meaningful sample, especially in the context of visibly-unsustainable shooting percentages with the weight of nearly a century of basketball behind them (even factoring in modern trends).

16 games is an inherently noisy sample. It's the same line you hear every year at the beginning of the season when someone is championing something happening over the first 10 or 20 games of the season, before that eventually evens out. Same same with a single playoff run. It's something to be viewed with at least SOME skepticism until and unless it is repeated. And especially so when there are some eye-popping numbers so discordant with RS production (and league/personal averages).

In any case, take care.

it appears that you did not see the article i shared with you in the post you quoted, therefore let me give it to you again https://www.statisticshowto.com/large-enough-sample-condition/

[cupcakesnake]

you don’t explain why something isn’t true

good bye forever

I'm not going to teach you basic math, friend. Perhaps take your own advice? The burden here is on you to explain why 16 games is a meaningful sample, especially in the context of visibly-unsustainable shooting percentages with the weight of nearly a century of basketball behind them (even factoring in modern trends).

16 games is an inherently noisy sample. It's the same line you hear every year at the beginning of the season when someone is championing something happening over the first 10 or 20 games of the season, before that eventually evens out. Same same with a single playoff run. It's something to be viewed with at least SOME skepticism until and unless it is repeated. And especially so when there are some eye-popping numbers so discordant with RS production (and league/personal averages).

In any case, take care.

it appears that you did not see the article i shared with you in the post you quoted, therefore let me give it to you again https://www.statisticshowto.com/large-enough-sample-condition/

So you feel that with Ant's sample size:

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

Ant having 1 bad game is enough to disqualify it from 16 being enough. Which he's had.
Depending on which data you're using. Let's go by basketball-reference's "game score" as a catch all metric for Ant's playoff game. Symmetric distribution would call for all of Ant's data points to fall within 1 point of 22.8 (Ant's average). Ant's data ranges from 8.9 to 36, making symmetric distribution impossible. You could widen the range, let's say to anything 10 points within 22.8 (a huge range) and still not accomplish the bare minimum for 16 games to become "large enough" by the definition you're trying to use.

I'm not a math guy or a statistician. I'm just reading the link you're using to justify being a jerk to another user.

[Godymas]

I'm not going to teach you basic math, friend. Perhaps take your own advice? The burden here is on you to explain why 16 games is a meaningful sample, especially in the context of visibly-unsustainable shooting percentages with the weight of nearly a century of basketball behind them (even factoring in modern trends).

16 games is an inherently noisy sample. It's the same line you hear every year at the beginning of the season when someone is championing something happening over the first 10 or 20 games of the season, before that eventually evens out. Same same with a single playoff run. It's something to be viewed with at least SOME skepticism until and unless it is repeated. And especially so when there are some eye-popping numbers so discordant with RS production (and league/personal averages).

In any case, take care.

it appears that you did not see the article i shared with you in the post you quoted, therefore let me give it to you again https://www.statisticshowto.com/large-enough-sample-condition/

So you feel that with Ant's sample size:

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

Ant having 1 bad game is enough to disqualify it from 16 being enough. Which he's had.
Depending on which data you're using. Let's go by basketball-reference's "game score" as a catch all metric for Ant's playoff game. Symmetric distribution would call for all of Ant's data points to fall within 1 point of 22.8 (Ant's average). Ant's data ranges from 8.9 to 36, making symmetric distribution impossible. You could widen the range, let's say to anything 10 points within 22.8 (a huge range) and still not accomplish the bare minimum for 16 games to become "large enough" by the definition you're trying to use.

I'm not a math guy or a statistician. I'm just reading the link you're using to justify being a jerk to another user.

no, i’m looking at his PPG as the baseline because advanced stats are dictated by things such as teammates and opponent.

Ultimately, I have no issue bumping this thread a year from now when it’s even more wrong.

[cupcakesnake]

it appears that you did not see the article i shared with you in the post you quoted, therefore let me give it to you again https://www.statisticshowto.com/large-enough-sample-condition/

So you feel that with Ant's sample size:

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

Ant having 1 bad game is enough to disqualify it from 16 being enough. Which he's had.
Depending on which data you're using. Let's go by basketball-reference's "game score" as a catch all metric for Ant's playoff game. Symmetric distribution would call for all of Ant's data points to fall within 1 point of 22.8 (Ant's average). Ant's data ranges from 8.9 to 36, making symmetric distribution impossible. You could widen the range, let's say to anything 10 points within 22.8 (a huge range) and still not accomplish the bare minimum for 16 games to become "large enough" by the definition you're trying to use.

I'm not a math guy or a statistician. I'm just reading the link you're using to justify being a jerk to another user.

no, i’m looking at his PPG as the baseline because advanced stats are dictated by things such as teammates and opponent.

Ultimately, I have no issue bumping this thread a year from now when it’s even more wrong.

Ok so his PPG has a range of 15-43. You still can't use your proposed statistical rule to establish 16 games is enough.

I'm pro-Ant and a Wolves fan. There's nothing wrong in believing in him. But there's also nothing wrong with questioning the sample size right now. You know how quick fans turn on a player after 1 bad series.

[tsherkin]

So you feel that with Ant's sample size:

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

Ant having 1 bad game is enough to disqualify it from 16 being enough. Which he's had.

And his shooting percentages outside the RA all qualify as "outliers" this season, especially the 69% from 16-23 feet. But he isn't looking for reasonable discourse.

I'm not a math guy or a statistician. I'm just reading the link you're using to justify being a jerk to another user.

Indeed.

In any case, I acknowledge that it's quite possible that Ant will continue to be a very good playoff performer. And to be honest, even in the RS where his scoring efficiency and some of his metrics over the much larger sample are uninspiring, he's also clearly trending upward in a variety of ways. He's irrefutably improving, and he's been excellent consistently in the playoffs. But these are things I've said earlier in this thread, and some are simply not paying attention, you know? You can tell very quickly what someone's purpose is in a given thread much of the time.

[Godymas]

So you feel that with Ant's sample size:


Ant having 1 bad game is enough to disqualify it from 16 being enough. Which he's had.
Depending on which data you're using. Let's go by basketball-reference's "game score" as a catch all metric for Ant's playoff game. Symmetric distribution would call for all of Ant's data points to fall within 1 point of 22.8 (Ant's average). Ant's data ranges from 8.9 to 36, making symmetric distribution impossible. You could widen the range, let's say to anything 10 points within 22.8 (a huge range) and still not accomplish the bare minimum for 16 games to become "large enough" by the definition you're trying to use.

I'm not a math guy or a statistician. I'm just reading the link you're using to justify being a jerk to another user.

no, i’m looking at his PPG as the baseline because advanced stats are dictated by things such as teammates and opponent.

Ultimately, I have no issue bumping this thread a year from now when it’s even more wrong.

Ok so his PPG has a range of 15-43. You still can't use your proposed statistical rule to establish 16 games is enough.

I'm pro-Ant and a Wolves fan. There's nothing wrong in believing in him. But there's also nothing wrong with questioning the sample size right now. You know how quick fans turn on a player after 1 bad series.

why are you quoting the range? why don't you put it in a graph and see if the distribution is symmetric? the range of his averages is so small that it has to be symmetric. the 15 low end is obviously an outlier otherwise he'd be averaging closer to that number for one of his seasons, but no, his average in the playoffs for his career is 29.8. I'd bet the 25th and 75th quartiles are around 26 and 33 his median and mean are going to be borderline equal. this is a symmetric data set.

[cupcakesnake]

no, i’m looking at his PPG as the baseline because advanced stats are dictated by things such as teammates and opponent.

Ultimately, I have no issue bumping this thread a year from now when it’s even more wrong.

Ok so his PPG has a range of 15-43. You still can't use your proposed statistical rule to establish 16 games is enough.

I'm pro-Ant and a Wolves fan. There's nothing wrong in believing in him. But there's also nothing wrong with questioning the sample size right now. You know how quick fans turn on a player after 1 bad series.

why are you quoting the range? why don't you put it in a graph and see if the distribution is symmetric? the range of his averages is so small that it has to be symmetric. the 15 low end is obviously an outlier otherwise he'd be averaging closer to that number for one of his seasons, but no, his average in the playoffs for his career is 29.8. I'd bet the 25th and 75th quartiles are around 26 and 33

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

I'm just quoting the model you posted.

[Godymas]

Ok so his PPG has a range of 15-43. You still can't use your proposed statistical rule to establish 16 games is enough.

I'm pro-Ant and a Wolves fan. There's nothing wrong in believing in him. But there's also nothing wrong with questioning the sample size right now. You know how quick fans turn on a player after 1 bad series.

why are you quoting the range? why don't you put it in a graph and see if the distribution is symmetric? the range of his averages is so small that it has to be symmetric. the 15 low end is obviously an outlier otherwise he'd be averaging closer to that number for one of his seasons, but no, his average in the playoffs for his career is 29.8. I'd bet the 25th and 75th quartiles are around 26 and 33

You have a symmetric distribution or unimodal distribution without outliers: a sample size of 15 is “large enough.”

I'm just quoting the model you posted.

now you're doing the same thing that OP did, ignoring the argument with no meaningful input, please if you don't understand the statistics then don't try and make it up

even if you remove the outlier of 15, you have 15 data points.

[babyjax13]

Because he is really young and so a lot of what he does that is great is muted by the things he does that are not. As he gains more experience (i.e., now) this will improve. I think next year is going to see what a more fully realized Ant looks like. We are lucky to have someone so exciting, especially since the league is clearly looking for a 'face' with LeBron about to retire. Edwards is entertaining to watch play and has a great personality.

[tsherkin]

Because he is really young and so a lot of what he does that is great is muted by the things he does that are not. As he gains more experience (i.e., now) this will improve. I think next year is going to see what a more fully realized Ant looks like.

This is quite possible. He's made some obvious improvements year to year, so it would match the trend.

[runtmc]

Since theres a lot of talk about what the math says without anyone actually doing it, I figured doing the math might help a bit:

A 95% CI on Ant's playoff FG% is (43.7%, 54.3%), which includes his career RS FG%, so as of now, it is a bit too early to say there's something there beyond luck/streakiness. That said, its relatively close. He's currently sitting at 335 FGA for his playoffs career, for a 1-tailed test he needs to get to about ~505FGA, or a little over 8 more games at the same rate for it to be a significant result. You can do a similar analysis for his TS%/3p% etc.

OTOH, if you do a two-proportion z-test with a two tailed hypothesis that p1!=p2, you end up with p = ~.12, and need somewhere around 540 shots, or about 10 more games.

Essentially, there is some good evidence so far that Ant does better in the playoffs, but its not yet significant, and you'd need somewhere between 8-10 games at the same level for it to become a significant result (meaning there's only a 5% or less chance its due to normal random noise/luck).

Edit: also just to point out, games played is really only a very rough guide for this, as usage will heavily influence things. A guy taking 20 shots per game will converge much faster than someone taking 5 or 10 shots a game, and Ant does have relatively high usage in the playoffs, which is why it might seem odd a relatively small sample size of ~25 games would be enough for a 4.4% FG% difference to show significance. For Gobert for example, you'd need closer to 70 games for a 4.4% gap to show significance.

[KembaWalker]

The fact that you can win a championship on what is deemed a “too noisy”, small sample size of games pretty much justifies me not giving a singular **** about any metrics. Lemme know when they start citing VORP for the hall of fame

[Special_Puppy]

i give 0 **** what his regular season #s are, his playoff #s are all superstar impact, he just dropped 43 on the defending champions, there is not one team in the West that has a guy that can shut him down. If the TWolves lose a series it won't be cause of Ant.

Right, but the issue here is one of sample size and expected regression. He's balling like an absolute demon right now, as he has for the first 5 or 6 games the past two seasons. This is not debatable. The question is more, what happens when he plays enough games in the playoffs that being on a heater starts to calm down? Because there are some specific shooting numbers he is very much not likely to maintain over an extended period of time, particularly as he sees better defenses.

The answer may well be "it doesn't affect him that much," in which case it will begin to confirm his ability and simultaneously reflect somewhat poorly on his approach in the RS, for whatever that is worth. The answer may also be that his shooting regresses enough that he looks more like his RS self later on, and so the questions about maintenance of shooting were valid.

We'll just had to see.

Is a player going from fringe top 20 player in the regular season to literal BITW candidate in the playoffs a believable story?