Bonn1997 wrote:

BigDaddyG wrote:

Bonn1997 wrote:I think Conley had pretty good college stats, especially for a freshman. If you switch the rbs and assists, his production looks pretty similar to Winslow's.

Yeah, but they were cancelled out by his shooting stats. A lot of people thought he was a reach when Memphis picked.

Shooting stats? I'd disagree. His 2 point shooting (58%) was phenomenal. Yeah, his 3 point shooting was low but you're talking about less than 70 total shots. His overall efficiency was excellent (because of the 2s). He had a true shooting % of .584 and effective field goal % of .552. Winslow is at .570 and .559 for those two stats.

FistOfOakley wrote:

WaltLongmire wrote:
Do your 2 point shooting stats consider whether they were jump shots or drives to the basket?

I would think you have look at where the shot is coming from. 3s are 3s, no matter how you slice it...not the same with a 2pt shot.

Winslow probably got to the basket a lot, and hit on his 3s at a good clip, but I have doubts about him continuing to shoot the 3 with as much success from the longer distance, and with more defensive pressure, and he is not just going to overpower NBA players, or drive by the smaller players he will face at SF. His inability to hit a shot off the dribble is documented, and I don't know much about his midrange game, but I can't think he took a ton of midrange jumpers this year, but I have no proof of this.

winslow didn't shoot a ton of midrange.. i think 80% of his shots were either 3s or in the restricted area... that's good... it means he's taking the most efficient shots on the floor...

russell took a lot of midrange shots but that's not necessarily good... if you're not confident in your finishing ability you will pull up a lot more often..

Russell also took a good number of 3s.

The issue with Winslow is whether he can do some of the things in the NBA at the 3, or maybe 2, against quicker NBA defenders. The criticism of some has been that he was playing against bigger/slower guys.

I like him- consider him an intangibles guy who's a winner, but there are still questions about his game, more so on offense, than D.

EwingsGlass wrote:

Isolating the value of the workout video - I cannot argue with you -- I would expect it to be near worthless. The same value as a real estate listing in the Sunday classifieds. They show what they want to show you - nothing more.

Without really statistically analyzing it, if I am rating prospects, I would put college stats as maybe a 10% indicator. They are biased by the player's role on the team. By the team's competition in the conference. Certain skill players are very well suited to the height and athleticism in the NCAA. Its that point that makes me scared of Russell. For me, I look for the players with elite athleticism. That's one of the reasons I am switching to a Hezonja fan. I am basically between Winslow and Hezonja... the likelihood of Winslow crapping out is pretty low. I feel like we would be getting a smaller Jimmy Butler. That said, Hezonja's ceiling is superstar in my opinion. He may be SuperMario. Or he could be a duck. But I am not as interested in guys that are not elite athletes.

For the PG position, I don't think you need elite atheleticism. That's one reason Chris Paul fell - he supposedly wasn't the athlete Deron Williams was... well, who cares, he's the best PG of his generation. Same with Nash and Steph Curry.

I think D'Angelo Russell is in that realm for PGs. He's just damn good. He has the tools and more importantly the floor vision and understanding of tempo to be a dominant PG. He really is just a classic classic PG.

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:Off the top of my head, Russell Westbrook. His stat line was not amazing. That said, he was considered a top defensive tweener based on athleticism. I can remember what the videos showed of him, but to half answer the question, mediocre stats are not determinative to the extent that elite athleticism exists IF player development continues in the NBA. If I am agreeing with you, for me, the videos of practice mean nothing to me. But combine stats mean something to me.

Russell Westbrook Stat Summary - UCLA Bruins:

										
Season	GP	MPG	PPG	FG%	3FG%	FT%	APG	RPG	BPG	SPG
2007-08	39	33.8	12.7	46.5	33.8	71.3	4.3	3.9	0.2	1.6
2006-07	36	9	3.4	45.7	40.9	54.8	0.7	0.8	0	0.4
										
										
Career	75	21.9	8.3	46.3	35.3	68.5	2.5	2.4	0.1	1

This would support the Stein campaign, but for me, I am still leaning toward Hezonja.

Thanks. I still suspect it's rare but that looks like a pretty good example actually.
I definitely agree with your statement that mediocre college stats are not automatically determinative of a poor NBA career. Some guys mature and work hard and grow beyond what could have been projected from their college stats. I just was (and still am) questioning that these workout videos are a good way to identify which guys will have that growth.

Isolating the value of the workout video - I cannot argue with you -- I would expect it to be near worthless. The same value as a real estate listing in the Sunday classifieds. They show what they want to show you - nothing more.

Without really statistically analyzing it, if I am rating prospects, I would put college stats as maybe a 10% indicator. They are biased by the player's role on the team. By the team's competition in the conference. Certain skill players are very well suited to the height and athleticism in the NCAA. Its that point that makes me scared of Russell. For me, I look for the players with elite athleticism. That's one of the reasons I am switching to a Hezonja fan. I am basically between Winslow and Hezonja... the likelihood of Winslow crapping out is pretty low. I feel like we would be getting a smaller Jimmy Butler. That said, Hezonja's ceiling is superstar in my opinion. He may be SuperMario. Or he could be a duck. But I am not as interested in guys that are not elite athletes.

There are stats that can adjust for the level of competition, though, and the advanced stats like win shares should be minimally influenced by the role on the team. I'm not saying the college stats are 100% but I'd put them way higher than 10% - probably at least 50%. -- said the guy who drafted Adam Morrison...

Stats are not necessarily determined by characteristics that lend themselves to success in the NBA. Sure, Lebron who has the gift of size, length, speed, athleticism/agility, skill and IQ is a generation changing talent that no one would miss out on. As you have to make choices on priority for these -- picking the stats is a bad way to do it. I tend to believe that college stats are mainly a function of skill at that level. But, at the NBA level, skill (while important, everyone has to be skilled at their role) takes a lesser role to the other characteristics. I think the NCAA stats are biased in favor of skill players instead of elite athletes. Not to say the athletes can't make their stats, but that they will develop NBA level skills in the NBA, while skill players don't tend to develop elite athleticism.

I've got Winslow with length (not size), athleticism, speed, and IQ. Could use more skill and well, his size isn't changing. The question is whether he is an "elite" athlete. Hezonja, is roughly the same, but with the elite athleticism, but you must question is IQ as he tends to try and add flourish to the game with useless behind the back passes. If Hezonja smartens up, super star. If not, pine rider. Winslow? He'll be like Butler. Maybe as good, maybe not. But he has a career in the NBA in front of him.

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:Off the top of my head, Russell Westbrook. His stat line was not amazing. That said, he was considered a top defensive tweener based on athleticism. I can remember what the videos showed of him, but to half answer the question, mediocre stats are not determinative to the extent that elite athleticism exists IF player development continues in the NBA. If I am agreeing with you, for me, the videos of practice mean nothing to me. But combine stats mean something to me.

Russell Westbrook Stat Summary - UCLA Bruins:

										
Season	GP	MPG	PPG	FG%	3FG%	FT%	APG	RPG	BPG	SPG
2007-08	39	33.8	12.7	46.5	33.8	71.3	4.3	3.9	0.2	1.6
2006-07	36	9	3.4	45.7	40.9	54.8	0.7	0.8	0	0.4
										
										
Career	75	21.9	8.3	46.3	35.3	68.5	2.5	2.4	0.1	1

This would support the Stein campaign, but for me, I am still leaning toward Hezonja.

Thanks. I still suspect it's rare but that looks like a pretty good example actually.
I definitely agree with your statement that mediocre college stats are not automatically determinative of a poor NBA career. Some guys mature and work hard and grow beyond what could have been projected from their college stats. I just was (and still am) questioning that these workout videos are a good way to identify which guys will have that growth.

Isolating the value of the workout video - I cannot argue with you -- I would expect it to be near worthless. The same value as a real estate listing in the Sunday classifieds. They show what they want to show you - nothing more.

Without really statistically analyzing it, if I am rating prospects, I would put college stats as maybe a 10% indicator. They are biased by the player's role on the team. By the team's competition in the conference. Certain skill players are very well suited to the height and athleticism in the NCAA. Its that point that makes me scared of Russell. For me, I look for the players with elite athleticism. That's one of the reasons I am switching to a Hezonja fan. I am basically between Winslow and Hezonja... the likelihood of Winslow crapping out is pretty low. I feel like we would be getting a smaller Jimmy Butler. That said, Hezonja's ceiling is superstar in my opinion. He may be SuperMario. Or he could be a duck. But I am not as interested in guys that are not elite athletes.

There are stats that can adjust for the level of competition, though, and the advanced stats like win shares should be minimally influenced by the role on the team. I'm not saying the college stats are 100% but I'd put them way higher than 10% - probably at least 50%. -- said the guy who drafted Adam Morrison...

Stats are not necessarily determined by characteristics that lend themselves to success in the NBA. Sure, Lebron who has the gift of size, length, speed, athleticism/agility, skill and IQ is a generation changing talent that no one would miss out on. As you have to make choices on priority for these -- picking the stats is a bad way to do it. I tend to believe that college stats are mainly a function of skill at that level. But, at the NBA level, skill (while important, everyone has to be skilled at their role) takes a lesser role to the other characteristics. I think the NCAA stats are biased in favor of skill players instead of elite athletes. Not to say the athletes can't make their stats, but that they will develop NBA level skills in the NBA, while skill players don't tend to develop elite athleticism.

I've got Winslow with length (not size), athleticism, speed, and IQ. Could use more skill and well, his size isn't changing. The question is whether he is an "elite" athlete. Hezonja, is roughly the same, but with the elite athleticism, but you must question is IQ as he tends to try and add flourish to the game with useless behind the back passes. If Hezonja smartens up, super star. If not, pine rider. Winslow? He'll be like Butler. Maybe as good, maybe not. But he has a career in the NBA in front of him.

Below are the R^2’s for the different correlations:

Points per minute: 0.3405
Field goal attempts per minute: 0.3522
Field goal percentage: 0.3436
Three-point attempts per minute: 0.6391
Three-point percentage: 0.7941
Free throw attempts per minute: 0.286
Free throw percentage: 0.7615
Rebounds per minute: 0.8312
Assists per minute: 0.8823
Steals per minute: 0.5981
Blocks per minute: 0.9327
Turnovers per minute: 0.4535
Personal fouls per minute: 0.4447

http://basketball-statistics.com/howdoncaastatisticstranslatetothenba.html
I think my 50% was a reasonable middle ground but the exact number obvious depends on the specific stat in question.
It's interesting that the rbs, assists, and blocks in college are almost perfectly correlated with the NBA #s.

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:Off the top of my head, Russell Westbrook. His stat line was not amazing. That said, he was considered a top defensive tweener based on athleticism. I can remember what the videos showed of him, but to half answer the question, mediocre stats are not determinative to the extent that elite athleticism exists IF player development continues in the NBA. If I am agreeing with you, for me, the videos of practice mean nothing to me. But combine stats mean something to me.

Russell Westbrook Stat Summary - UCLA Bruins:

										
Season	GP	MPG	PPG	FG%	3FG%	FT%	APG	RPG	BPG	SPG
2007-08	39	33.8	12.7	46.5	33.8	71.3	4.3	3.9	0.2	1.6
2006-07	36	9	3.4	45.7	40.9	54.8	0.7	0.8	0	0.4
										
										
Career	75	21.9	8.3	46.3	35.3	68.5	2.5	2.4	0.1	1

This would support the Stein campaign, but for me, I am still leaning toward Hezonja.

Thanks. I still suspect it's rare but that looks like a pretty good example actually.
I definitely agree with your statement that mediocre college stats are not automatically determinative of a poor NBA career. Some guys mature and work hard and grow beyond what could have been projected from their college stats. I just was (and still am) questioning that these workout videos are a good way to identify which guys will have that growth.

Isolating the value of the workout video - I cannot argue with you -- I would expect it to be near worthless. The same value as a real estate listing in the Sunday classifieds. They show what they want to show you - nothing more.

Without really statistically analyzing it, if I am rating prospects, I would put college stats as maybe a 10% indicator. They are biased by the player's role on the team. By the team's competition in the conference. Certain skill players are very well suited to the height and athleticism in the NCAA. Its that point that makes me scared of Russell. For me, I look for the players with elite athleticism. That's one of the reasons I am switching to a Hezonja fan. I am basically between Winslow and Hezonja... the likelihood of Winslow crapping out is pretty low. I feel like we would be getting a smaller Jimmy Butler. That said, Hezonja's ceiling is superstar in my opinion. He may be SuperMario. Or he could be a duck. But I am not as interested in guys that are not elite athletes.

There are stats that can adjust for the level of competition, though, and the advanced stats like win shares should be minimally influenced by the role on the team. I'm not saying the college stats are 100% but I'd put them way higher than 10% - probably at least 50%. -- said the guy who drafted Adam Morrison...

Stats are not necessarily determined by characteristics that lend themselves to success in the NBA. Sure, Lebron who has the gift of size, length, speed, athleticism/agility, skill and IQ is a generation changing talent that no one would miss out on. As you have to make choices on priority for these -- picking the stats is a bad way to do it. I tend to believe that college stats are mainly a function of skill at that level. But, at the NBA level, skill (while important, everyone has to be skilled at their role) takes a lesser role to the other characteristics. I think the NCAA stats are biased in favor of skill players instead of elite athletes. Not to say the athletes can't make their stats, but that they will develop NBA level skills in the NBA, while skill players don't tend to develop elite athleticism.

I've got Winslow with length (not size), athleticism, speed, and IQ. Could use more skill and well, his size isn't changing. The question is whether he is an "elite" athlete. Hezonja, is roughly the same, but with the elite athleticism, but you must question is IQ as he tends to try and add flourish to the game with useless behind the back passes. If Hezonja smartens up, super star. If not, pine rider. Winslow? He'll be like Butler. Maybe as good, maybe not. But he has a career in the NBA in front of him.

Well I found this, which is exactly a measure of what we're discussing. It's the R squared for the correlation between college and NBA stats. You have to multiply the R squared by 100 to get the percentage of the variability in the NBA that is predicted by the college stats since R squared is out of 1.00.

Below are the R^2’s for the different correlations:

Points per minute: 0.3405
Field goal attempts per minute: 0.3522
Field goal percentage: 0.3436
Three-point attempts per minute: 0.6391
Three-point percentage: 0.7941
Free throw attempts per minute: 0.286
Free throw percentage: 0.7615
Rebounds per minute: 0.8312
Assists per minute: 0.8823
Steals per minute: 0.5981
Blocks per minute: 0.9327
Turnovers per minute: 0.4535
Personal fouls per minute: 0.4447

http://basketball-statistics.com/howdoncaastatisticstranslatetothenba.html
I think my 50% was a reasonable middle ground but the exact number obvious depends on the specific stat in question.
It's interesting that the rbs, assists, and blocks in college are almost perfectly correlated with the NBA #s.

This stat is helpful. It seems to me that stats like Rebounds, Assists and Blocks, and to a lesser extent Three point percentage and FT percentage appear to be highly correlated, but points and everything else appear to be mostly uncorrelated -- at least so uncorrelated as to not use as a real decision making point.

Look at it like this --- whatever players have that allow them to rebound in college are exactly the same characteristics they need in the NBA. However, the characteristics of players in college that allow them to score in college only makes up about 35% of the characteristics that allow them to score in the NBA.

To me, that doesn't mean to me that I should use that stat to make 35% of my 'buying decision' -- but instead that the points stats may vary by 65% in either direction. So a 10 pt scorer in college may be a 3 point scorer in the NBA or a 17 point scorer. That deviation makes it uncorrelated.

But for Blocks, Rebounds and Assists, it appears to be very indicative.

I'm not sure how much you've studied the behavioral sciences but r squared values around .3 or .4 are definitely not dismissed. They're considered pretty large still. I saw the exact figure a long time ago but I believe the correlation between smoking fequency and lung cancer risk is only about that high.

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:

Bonn1997 wrote:

EwingsGlass wrote:Off the top of my head, Russell Westbrook. His stat line was not amazing. That said, he was considered a top defensive tweener based on athleticism. I can remember what the videos showed of him, but to half answer the question, mediocre stats are not determinative to the extent that elite athleticism exists IF player development continues in the NBA. If I am agreeing with you, for me, the videos of practice mean nothing to me. But combine stats mean something to me.

Russell Westbrook Stat Summary - UCLA Bruins:

										
Season	GP	MPG	PPG	FG%	3FG%	FT%	APG	RPG	BPG	SPG
2007-08	39	33.8	12.7	46.5	33.8	71.3	4.3	3.9	0.2	1.6
2006-07	36	9	3.4	45.7	40.9	54.8	0.7	0.8	0	0.4
										
										
Career	75	21.9	8.3	46.3	35.3	68.5	2.5	2.4	0.1	1

This would support the Stein campaign, but for me, I am still leaning toward Hezonja.

Thanks. I still suspect it's rare but that looks like a pretty good example actually.
I definitely agree with your statement that mediocre college stats are not automatically determinative of a poor NBA career. Some guys mature and work hard and grow beyond what could have been projected from their college stats. I just was (and still am) questioning that these workout videos are a good way to identify which guys will have that growth.

Isolating the value of the workout video - I cannot argue with you -- I would expect it to be near worthless. The same value as a real estate listing in the Sunday classifieds. They show what they want to show you - nothing more.

Without really statistically analyzing it, if I am rating prospects, I would put college stats as maybe a 10% indicator. They are biased by the player's role on the team. By the team's competition in the conference. Certain skill players are very well suited to the height and athleticism in the NCAA. Its that point that makes me scared of Russell. For me, I look for the players with elite athleticism. That's one of the reasons I am switching to a Hezonja fan. I am basically between Winslow and Hezonja... the likelihood of Winslow crapping out is pretty low. I feel like we would be getting a smaller Jimmy Butler. That said, Hezonja's ceiling is superstar in my opinion. He may be SuperMario. Or he could be a duck. But I am not as interested in guys that are not elite athletes.

There are stats that can adjust for the level of competition, though, and the advanced stats like win shares should be minimally influenced by the role on the team. I'm not saying the college stats are 100% but I'd put them way higher than 10% - probably at least 50%. -- said the guy who drafted Adam Morrison...

Stats are not necessarily determined by characteristics that lend themselves to success in the NBA. Sure, Lebron who has the gift of size, length, speed, athleticism/agility, skill and IQ is a generation changing talent that no one would miss out on. As you have to make choices on priority for these -- picking the stats is a bad way to do it. I tend to believe that college stats are mainly a function of skill at that level. But, at the NBA level, skill (while important, everyone has to be skilled at their role) takes a lesser role to the other characteristics. I think the NCAA stats are biased in favor of skill players instead of elite athletes. Not to say the athletes can't make their stats, but that they will develop NBA level skills in the NBA, while skill players don't tend to develop elite athleticism.

I've got Winslow with length (not size), athleticism, speed, and IQ. Could use more skill and well, his size isn't changing. The question is whether he is an "elite" athlete. Hezonja, is roughly the same, but with the elite athleticism, but you must question is IQ as he tends to try and add flourish to the game with useless behind the back passes. If Hezonja smartens up, super star. If not, pine rider. Winslow? He'll be like Butler. Maybe as good, maybe not. But he has a career in the NBA in front of him.

Well I found this, which is exactly a measure of what we're discussing. It's the R squared for the correlation between college and NBA stats. You have to multiply the R squared by 100 to get the percentage of the variability in the NBA that is predicted by the college stats since R squared is out of 1.00.

Below are the R^2’s for the different correlations:

Points per minute: 0.3405
Field goal attempts per minute: 0.3522
Field goal percentage: 0.3436
Three-point attempts per minute: 0.6391
Three-point percentage: 0.7941
Free throw attempts per minute: 0.286
Free throw percentage: 0.7615
Rebounds per minute: 0.8312
Assists per minute: 0.8823
Steals per minute: 0.5981
Blocks per minute: 0.9327
Turnovers per minute: 0.4535
Personal fouls per minute: 0.4447

http://basketball-statistics.com/howdoncaastatisticstranslatetothenba.html
I think my 50% was a reasonable middle ground but the exact number obvious depends on the specific stat in question.
It's interesting that the rbs, assists, and blocks in college are almost perfectly correlated with the NBA #s.

This stat is helpful. It seems to me that stats like Rebounds, Assists and Blocks, and to a lesser extent Three point percentage and FT percentage appear to be highly correlated, but points and everything else appear to be mostly uncorrelated -- at least so uncorrelated as to not use as a real decision making point.

Look at it like this --- whatever players have that allow them to rebound in college are exactly the same characteristics they need in the NBA. However, the characteristics of players in college that allow them to score in college only makes up about 35% of the characteristics that allow them to score in the NBA.

To me, that doesn't mean to me that I should use that stat to make 35% of my 'buying decision' -- but instead that the points stats may vary by 65% in either direction. So a 10 pt scorer in college may be a 3 point scorer in the NBA or a 17 point scorer. That deviation makes it uncorrelated.

But for Blocks, Rebounds and Assists, it appears to be very indicative.

Note also that a lot of the other 65% of the variance may just be considered random and unpredictable from the information available at age being drafted. I have not seen any other measures that predict with a higher accuracy the NBA stats than the college stats do. You're better off picking the factor with 35% predictive validity than factors with zero established connection to NBA production.
Also, the 35% figure refers to the closeness of one's ranking in college to one's NBA ranking. It's not meant to be multiplied to determine the range of the player's NBA stats.

Not dismissed, no. But in our situation, I am clearly data dredging to support my case that basing draft decisions based on college scoring is not only risky, but careless and harmful. Using my handful of examples mostly centered around Adam Morrison, I am trying to convince you that the obvious basketball characteristics that anyone with a NBA video game could determine are more obviously tied to NBA success than college stats because college stats obviously have other factors that bias the result. I'm left with the final conclusion that college stats are helpful but not controlling but if I'm choosing between two interesting players, I'm going to take the more athletic one even if the other guy scored more points in college. And after all those words, anyone who finished reading this diatribe should think to themselves... "Well, goink"