Your unpopular takes? (PC Board Edition)

Moderators: trex_8063, penbeast0, PaulieWal, Clyde Frazier, Doctor MJ

User avatar
Jaivl
Head Coach
Posts: 7,108
And1: 6,761
Joined: Jan 28, 2014
Location: A Coruña, Spain
Contact:
   

Re: Your unpopular takes? (PC Board Edition) 

Post#301 » by Jaivl » Sun Apr 9, 2023 1:02 pm

OhayoKD wrote:
MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that

I never said you did?
If I just guessed the net-rating of every team was 0, I would get an error-rate around 3. Actually IIRC, Unibro did that calc in January and got an error rate of 2.8 which you might notice is "lower" than 2.85. But hey, let's just be nice and go with 3. Maybe the blowouts got more frequent since?

Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.
This place is a cesspool of mindless ineptitude, mental decrepitude, and intellectual lassitude. I refuse to be sucked any deeper into this whirlpool of groupthink sewage. My opinions have been expressed. I'm going to go take a shower.
OhayoKD
Head Coach
Posts: 6,042
And1: 3,933
Joined: Jun 22, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#302 » by OhayoKD » Sun Apr 9, 2023 2:11 pm

Jaivl wrote:
OhayoKD wrote:
MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that

I never said you did?
If I just guessed the net-rating of every team was 0, I would get an error-rate around 3. Actually IIRC, Unibro did that calc in January and got an error rate of 2.8 which you might notice is "lower" than 2.85. But hey, let's just be nice and go with 3. Maybe the blowouts got more frequent since?

Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.

Was probabaly MAE then. Point was the "10%" difference doesn't really mean anything in a vacuum, but fair enough. It's not a like for like calculation so you can't make positive claims regarding the gap
f4p
Sixth Man
Posts: 1,879
And1: 1,870
Joined: Sep 19, 2021
 

Re: Your unpopular takes? (PC Board Edition) 

Post#303 » by f4p » Sun Apr 9, 2023 6:07 pm

MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that


assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?

Jaivl wrote:Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.


for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.
MyUniBroDavis
General Manager
Posts: 7,827
And1: 5,034
Joined: Jan 14, 2013

Re: Your unpopular takes? (PC Board Edition) 

Post#304 » by MyUniBroDavis » Sun Apr 9, 2023 6:47 pm

f4p wrote:
MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that


assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?

Jaivl wrote:Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.


for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.




I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post
User avatar
Narigo
Veteran
Posts: 2,799
And1: 883
Joined: Sep 20, 2010
     

Re: Your unpopular takes? (PC Board Edition) 

Post#305 » by Narigo » Sun Apr 9, 2023 7:07 pm

Kevin Durant is the greatest shooter ever not Steph

Shawn Marion is extremely underrated and doesnt get enough credit on the Suns
Narigo's Fantasy Team

PG: Damian Lillard
SG: Sidney Moncrief
SF:
PF: James Worthy
C: Tim Duncan

BE: Robert Horry
BE:
BE:
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#306 » by ShaqAttac » Sun Apr 9, 2023 7:27 pm

MyUniBroDavis wrote:
f4p wrote:
MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that


assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?

Jaivl wrote:Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.


for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.




I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post

wuts mae n rmse.

n why pf keep askin whats wrong like it hasnt alr been said a bunch
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#307 » by ShaqAttac » Sun Apr 9, 2023 7:35 pm

Narigo wrote:Kevin Durant is the greatest shooter ever not Steph

Shawn Marion is extremely underrated and doesnt get enough credit on the Suns

da fuq
f4p
Sixth Man
Posts: 1,879
And1: 1,870
Joined: Sep 19, 2021
 

Re: Your unpopular takes? (PC Board Edition) 

Post#308 » by f4p » Sun Apr 9, 2023 7:35 pm

MyUniBroDavis wrote:
f4p wrote:
MyUniBroDavis wrote:^ this is a horrible discussion going on up here btw

Also I didn’t calculate the RMSE lol how would I even do that


assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?

Jaivl wrote:Show your process.

If the root mean squared error of picking 41 wins for everybody is 2.85 I'll happily eat my pants and film it on camera.


for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.




I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post


lol, 2 whole posts about statistics on the realgm pc board, oh my. maybe one of the 10 lebron vs MJ threads would be more your speed.
User avatar
Narigo
Veteran
Posts: 2,799
And1: 883
Joined: Sep 20, 2010
     

Re: Your unpopular takes? (PC Board Edition) 

Post#309 » by Narigo » Sun Apr 9, 2023 7:43 pm

ShaqAttac wrote:
Narigo wrote:Kevin Durant is the greatest shooter ever not Steph

Shawn Marion is extremely underrated and doesnt get enough credit on the Suns

da fuq


whats so offensive about this statement? I think when considering the mid range, I think KD is a better overall shooter than Steph Curry
Narigo's Fantasy Team

PG: Damian Lillard
SG: Sidney Moncrief
SF:
PF: James Worthy
C: Tim Duncan

BE: Robert Horry
BE:
BE:
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#310 » by ShaqAttac » Sun Apr 9, 2023 7:57 pm

Narigo wrote:
ShaqAttac wrote:
Narigo wrote:Kevin Durant is the greatest shooter ever not Steph

Shawn Marion is extremely underrated and doesnt get enough credit on the Suns

da fuq


whats so offensive about this statement? I think when considering the mid range, I think KD is a better overall shooter than Steph Curry

curry got an all-time middy, is a goat-lvl ft shootah, and shoots way better from all the other spots.

bros really be takin warrior numbers at face value. AD shot as good when we won without the 73-win tm behind him
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#311 » by ShaqAttac » Sun Apr 9, 2023 8:01 pm

f4p wrote:
MyUniBroDavis wrote:
f4p wrote:
assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?



for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.




I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post


lol, 2 whole posts about statistics on the realgm pc board, oh my. maybe one of the 10 lebron vs MJ threads would be more your speed.

i mean ur just a broken clock rn. it was alr explained with da weather ****
tsherkin
Forum Mod - Raptors
Forum Mod - Raptors
Posts: 92,310
And1: 31,884
Joined: Oct 14, 2003
 

Re: Your unpopular takes? (PC Board Edition) 

Post#312 » by tsherkin » Sun Apr 9, 2023 8:01 pm

ShaqAttac wrote:
Narigo wrote:
ShaqAttac wrote:da fuq


whats so offensive about this statement? I think when considering the mid range, I think KD is a better overall shooter than Steph Curry

curry got an all-time middy, is a goat-lvl ft shootah, and shoots way better from all the other spots.

bros really be takin warrior numbers at face value. AD shot as good when we won without the 73-win tm behind him


It isn't controversial to think of Durant as a better mid-range shooter than Steph. It's statistically true, particularly from 16-23 feet. And not surprising, given Durant's release point and general shooting proficiency. He can't smash threes like Steph but he can sure as Hell crush the middle game like no one else.
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#313 » by ShaqAttac » Sun Apr 9, 2023 8:02 pm

tsherkin wrote:
ShaqAttac wrote:
Narigo wrote:
whats so offensive about this statement? I think when considering the mid range, I think KD is a better overall shooter than Steph Curry

curry got an all-time middy, is a goat-lvl ft shootah, and shoots way better from all the other spots.

bros really be takin warrior numbers at face value. AD shot as good when we won without the 73-win tm behind him


It isn't controversial to think of Durant as a better mid-range shooter than Steph. It's statistically true, particularly from 16-23 feet. And not surprising, given Durant's release point and general shooting proficiency. He can't smash threes like Steph but he can sure as Hell crush the middle game like no one else.

but rigo sayin he shoot better OVERALL
tsherkin
Forum Mod - Raptors
Forum Mod - Raptors
Posts: 92,310
And1: 31,884
Joined: Oct 14, 2003
 

Re: Your unpopular takes? (PC Board Edition) 

Post#314 » by tsherkin » Sun Apr 9, 2023 8:04 pm

ShaqAttac wrote:but rigo sayin he shoot better OVERALL


It depends on how you evaluate shooters. Steph is clearly a better 3pt shooter. Durant seems generally more proficient as a shooter everywhere else. Durant is able to be a 65% TS player over the past half decade on the strength of average FTr and his absurd mid-range shooting coupled to good but not-Steph level shooting. I think that speaks to a level of shooting prowess coupled to volume that merits the discussion, if nothing else.
MyUniBroDavis
General Manager
Posts: 7,827
And1: 5,034
Joined: Jan 14, 2013

Re: Your unpopular takes? (PC Board Edition) 

Post#315 » by MyUniBroDavis » Sun Apr 9, 2023 8:08 pm

f4p wrote:
MyUniBroDavis wrote:
f4p wrote:
assuming you were including me and not just hayo, what exactly did i get wrong from the article?

their methodology was:

1. take values of a metric for all players from season 1
2. weight those values by the minutes played in season 2
3. use that to predict a net rating for a team in season 2
4. compare that to the actual net rating for that team in season 2 to get an error
5. use that error for all 30 teams to calculate the RMSE in units of net rating
6. rinse and repeat for other metrics

does that not accurately describe the paper?

from there, with the errors for net rating, we can easily just change it to error for wins by multiplying by 2.7. if a metric has an error of 2.6 and another has an error of 2.8, then we can estimate that they only differ in accuracy for win predictions by 0.54, correct?



for 2023, i get an RMSE for the league of 3.97 using a guess of 0 for every team. the 2.85 is for net rating, not wins. so 3.97 in net rating RMSE would be 10 or 11 wins. so you can keep your pants. if this was 1977 though, the RMSE would be 2.81 (for net rating), so you might have to at least eat one leg.


now i'm not enough of a statistician to explain why just guessing the mean and getting an RMSE equal to the RMSE of all those metrics (at least for 1977) is not the correct way to look at things, but i suspect it is. if we had a league with the same talent distribution amongst players as it is currently, but the players just all happened to be distributed on teams in a way that made every team a 41 win team, the mean value guess error would be 0. it would be perfect. and all metrics would essentially look worthless. if however, that same league had talent distributed such that half the league went 82-0 and half went 0-82, the mean value guess would have a horrific RMSE of about 15 net rating points. meanwhile the metrics would have the same (presumably) error in both cases. so the effectiveness of the mean value theorem (trademarked) would depend only on the distribution of talent on the teams.




I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post


lol, 2 whole posts about statistics on the realgm pc board, oh my. maybe one of the 10 lebron vs MJ threads would be more your speed.


Im seeing an argument about statistics form one guy talking about hurricane and shohei and another guy that keeps saying “oh yeah I’m not a stats guy but” when this thread was pretty fun earlier and y’all are now arguing about interpretations of a blog post because neither of y’all want to just ask the author about the post itself lol
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#316 » by ShaqAttac » Sun Apr 9, 2023 8:09 pm

tsherkin wrote:
ShaqAttac wrote:but rigo sayin he shoot better OVERALL


It depends on how you evaluate shooters. Steph is clearly a better 3pt shooter. Durant seems generally more proficient as a shooter everywhere else. Durant is able to be a 65% TS player over the past half decade on the strength of average FTr and his absurd mid-range shooting coupled to good but not-Steph level shooting. I think that speaks to a level of shooting prowess coupled to volume that merits the discussion, if nothing else.

steph way better at middies than kd is at trips and > at ft's too

how does that add up to kd having a better shot
ShaqAttac
Rookie
Posts: 1,189
And1: 370
Joined: Oct 18, 2022

Re: Your unpopular takes? (PC Board Edition) 

Post#317 » by ShaqAttac » Sun Apr 9, 2023 8:20 pm

MyUniBroDavis wrote:
f4p wrote:
MyUniBroDavis wrote:


I’m just saying this is a horrible discussion in general lmao

The guy that made the blog post isn’t hard to contact cant y’all just dm him man no ones trying to hear either of y’all argue about stats from a blog post


lol, 2 whole posts about statistics on the realgm pc board, oh my. maybe one of the 10 lebron vs MJ threads would be more your speed.


Im seeing an argument about statistics form one guy talking about hurricane and shohei and another guy that keeps saying “oh yeah I’m not a stats guy but” when this thread was pretty fun earlier and y’all are now arguing about interpretations of a blog post because neither of y’all want to just ask the author about what the test means lmao

unicorns, mages and per def hit different

i think i got the idea but dam
tsherkin
Forum Mod - Raptors
Forum Mod - Raptors
Posts: 92,310
And1: 31,884
Joined: Oct 14, 2003
 

Re: Your unpopular takes? (PC Board Edition) 

Post#318 » by tsherkin » Sun Apr 9, 2023 8:24 pm

ShaqAttac wrote:steph way better at middies than kd is at trips and > at ft's too


I think you underestimate the difference in their mid-range shooting. From 17-23, Durant has taken fully 41% of his shots (or almost 7.4 per game) from 10-23 feet, shooting 52.0% from 10-16 and 53.1% from 16-23. That's... absurd.

Steph over the same time frame has taken about 15% of his shots (or just under 3 per game, shooting 49.8% and 47.1% from 10-16 and 16-23 respectively. Add to that Durant shooting a shade over 39% from 3 and over 89% from the FT line during this time frame while generating more FTA/g than Steph and you don't really have an argument about Steph's FT shooting being relevant.

Yes, Steph is the ultimate 3pt shooter in terms of range, volume and raw percentage. He's the best 3pt shooter we've ever seen, to an outlier degree and he's better at that than Durant. But KD leverages the mid-range very, very effectively and he's noticeably better at it on more than double Steph's volume... and that's not considering 3-10 feet, where he takes another 12.5% of his shots at 52.7% FG compared to Steph taking 11.5% (quite similar) and shooting 50.6%. So again, another evident advantage for Durant over a large block of time.

I'm not saying that that you can't make a good pro-Steph argument, I'm saying your incredulity over the idea that the comparison should be made is misplaced. There's a very good pro-KD argument in terms of total shooting ability. Whether that's enough that lots of people will accept it or not is another story, because Steph's 3pt shooting is wild stuff, to be sure. But there's no sense wailing at the basic premise that someone drew the comparison and has the dissenting opinion, because Durant is better at shooting from everywhere under the arc and that does matter to some extent.
User avatar
Joao Saraiva
RealGM
Posts: 13,450
And1: 6,218
Joined: Feb 09, 2011
   

Re: Your unpopular takes? (PC Board Edition) 

Post#319 » by Joao Saraiva » Mon Apr 10, 2023 1:59 pm

breezypeezy wrote:
Joao Saraiva wrote:1. Billups is a top 50 player ever. He should be ranked above Allen Iverson and Isiah Thomas to start. I think he's very underrated, even if people recognize his value.

2. Hakeem Olajuwon has the best playoff run ever in 94, and depending on how much you value playoffs he has a great case for the best peak ever.

3. Kobe Bryant was a better player than Larry Bird.

4. Per position, Andrei Kirilenko should be ranked among the best defenders ever.

5. LBJ's contributions on defense in his earlier years (07-13) got underappreciated with time and he should be regarded as en elite defender in that period.

(1) and (3) are odd views, at least someone agrees, your thesis did get a "and 1" I sée.

My question is did you see Isiah or Larry play or is your contention from a statistical measurement. These 2 were magical elements (along with Magic himself) to any court they stepped on.
Not certain theres a statistical capture over how much they elevated everyone around them. They were Mel Gibsons "Braveheart" level leaders, the guys to win lead and win battles. Those games were football level bruising contests back then and to a teammate everyone they played with knew who was in charge the minute they hit the court.
The intangible value of such players doesnt bleed through a stat sheet, they were weighed in fearlessness, drive, and mercy-less will to win and intimate, they engendered fear and surrender from opponents who knew they could not overcome this opponent.
In their primes they were every bit the legend they have.

Kobe, Chauncey great leaders and players, yet nothing on Isiah or Larry's level. Jordan and Magic also had this factor as well, im just not certain how measureable it was but their peers and opponents knew what they were in store for whenever they faced off.


If I watched them play... not live. I followed NBA since the mid/late 90s. I've seen altough as much as I can from previous eras, specially playoffs. So I'd say I've seen a good bunch of games from Larry, Magic and Isiah.

I understand the aura of Larry Bird, I understand his passing was superb and that he contributed in a lot of ways but Larry failed a lot of times in the playoffs. He had a ton of subpar series, so I give the edge to Kobe on consistency and maintaining his level of play.

Also Kobe's longevity might be a bit underrated, but he's still better than Bird in that regard. So I guess consistency, better longevity overtakes Bird despite Bird's higher peak and prime.

I also think Kobe has become a bit underrated arround the PC board or that he's always been. I understand he didn't always make the correct play, I understand that the "clutch" ability is also elevated by casuals. But he did get off a shot against any type of defense, and that ability is more important that it seems. His ts% for someone who took so many bad shots is actually pretty high, specially when you take into consideration his low TOV%.




About Billups... I believe he was a fantastic floor general and pace controller. That's why those Pistons teams had a ton of success with him running the PG spot, and also why when he went to Denver they made it to the WCF and Carmelo seemed to be the best of himself. Billups could score, he would score, but he would certainly put his teammates in position to succeed.

Also he took advantage of mismatches, read em correctly and was patient. He was a good 3 point shooter and a good defender as a PG. He's the definition of a guy who I think did everything correct on a basketball court.

I think IT is a bit overrated. He had the explosion as a scorer, but I don't see him as such a general as Billups was, or even consistently being a better scorer than Billups.
“These guys have been criticized the last few years for not getting to where we’re going, but I’ve always said that the most important thing in sports is to keep trying. Let this be an example of what it means to say it’s never over.” - Jerry Sloan
tsherkin
Forum Mod - Raptors
Forum Mod - Raptors
Posts: 92,310
And1: 31,884
Joined: Oct 14, 2003
 

Re: Your unpopular takes? (PC Board Edition) 

Post#320 » by tsherkin » Mon Apr 10, 2023 3:15 pm

Joao Saraiva wrote:I also think Kobe has become a bit underrated arround the PC board or that he's always been. I understand he didn't always make the correct play, I understand that the "clutch" ability is also elevated by casuals. But he did get off a shot against any type of defense, and that ability is more important that it seems. His ts% for someone who took so many bad shots is actually pretty high, specially when you take into consideration his low TOV%.


I think the problem with Kobe discussions is more that you get his stans who wildly overreach, and then a lot of conversational backlash as people discuss why he wasn't God's gift to basketball, even if the folks making counterpoints have a healthy respect for Kobe's ability. And then there's the whole discussion of Kobe's ability versus how he elected to play, which is another little wrinkle in all of that, because he could have been better if he didn't focus on several of the wrong things. Even still, a magnificent player and ATG, which speaks volumes of his talent and development. Also, Kobe's TS% isn't bad; 01-09, he was under 104 TS+ once (and it was 103). 2010-2012, not ideal, but then back to 107 in 2013. He was actually more efficient than a lot of his peers, and at much higher volume than most of them, including the more efficient ones. You're talking about Wade, Peja, Pierce, Lebron, Vince, Melo and Ray Allen as his most consistent competitors through 2013 as far as guards and wings at volume. Certainly pre-05. Couple of seasons here and there from like Jason Richardson, Corey Maggette, etc, etc but by and large, Kobe was in a very elite tier in terms of scoring efficiency and volume during his hey-day. He lagged behind Lebron and Wade, and surely behind Ray (though Allen didn't really touch Kobe's volume outside of his earliest days in Seattle).

Kobe has this weird reputation where people think he was inefficient, which isn't true. He was a guy who made frustrating decisions at inopportune times. His overall profile as a scorer was quite good, and could have been better had he not grown a fondness for the dumbest shots and sets in basketball. But even still, he was incredible.


About Billups... I believe he was a fantastic floor general and pace controller. That's why those Pistons teams had a ton of success with him running the PG spot, and also why when he went to Denver they made it to the WCF and Carmelo seemed to be the best of himself. Billups could score, he would score, but he would certainly put his teammates in position to succeed.


Billups was outstanding. He was an early adopter of the FTA/3PA type model. He wasn't a very good scorer in between the rim and the 3pt line but damn did he put pressure on a defense in his day, and as you say, he was a smart floor general. He also didn't call his own number too frequently.

I think IT is a bit overrated. He had the explosion as a scorer, but I don't see him as such a general as Billups was, or even consistently being a better scorer than Billups.


Isiah wasn't actually a particularly good scorer relative to his volume-scoring peers. For a few years when he was a volume playmaker, he had other major offensive value, but I also consider him one of the more overrated players. Only so far, as he managed to integrate into good teams, buy into the defensive philosophy, and in as much as we can't apply numbers to leadership, he didn't have his team's ear pretty clearly.

Return to Player Comparisons