Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
I wonder how reliable stats compiled from 2 years at San Diego State are? Isn't the sample size a bit small to be basing your franchises future on? What about the 1 and done players? I think drafting is still as much a crap shoot as it's always been.
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
I wonder how reliable stats compiled from 2 years at San Diego State are? Isn't the sample size a bit small to be basing your franchises future on? What about the 1 and done players? I think drafting is still as much a crap shoot as it's always been.
Most of us present to observe Kawhi's almost 70 games at SDSU believed that he had the capacity to thrive in the association. Would the institution or data size have been more compelling if he was at Kentucky or Memphis? If that's your position, you might care to explain how he's proven to be that much better than guys like MKG or CDR ... or cancel that thought: maybe you can just explain the career arc of someone like Scottie Pippen from Central Arkansas.
SDSU has been a credible program for 10+ years; your comment seems to suggest otherwise ...
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
I wonder how reliable stats compiled from 2 years at San Diego State are? Isn't the sample size a bit small to be basing your franchises future on? What about the 1 and done players? I think drafting is still as much a crap shoot as it's always been.
Most of us present to observe Kawhi's almost 70 games at SDSU believed that he had the capacity to thrive in the association. Would the institution or data size have been more compelling if he was at Kentucky or Memphis? If that's your position, you might care to explain how he's proven to be that much better than guys like MKG or CDR ... or cancel that thought: maybe you can just explain the career arc of someone like Scottie Pippen from Central Arkansas.
SDSU has been a credible program for 10+ years; your comment seems to suggest otherwise ...
It wasn't a question of where he went to school, rather the sample size. 70 games isn't a lot is it, when compiling advanced stats to project future success?
Simply watching Leonard play isn't what the analytics is based on, so your observation here isn't part of the discussion.
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
No, it's just good drafting. Analytics may have played a role in the drafting, but none of us know how much of a role.
It's funny how this whole thread quickly dissolve into people "proving" analytics failed or succeeded by just naming good or bad draft choices. All that shows me is few people in this thread have any idea what analytics are.
Too many people assume that if you use analytics, then analytics is all you use to make decisions. They assume that analytics are in opposition to observation rather than work in concert with them. They assume that analytics are either perfect/flawless or have zero value. They assume that if you use analytics and don't win a ring, then the analytics were 100% unsuccessful.
"Analytics creates an environment in which talent is no longer necessary to win championships. It also renders all other forms of talent evaluation meaningless. Acquiring talented players and analytics are two mutually exclusive concepts, and one must choose one or the other."
-Dr. Nonsense von Strawman, University of Atlantis.
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
3 of whom were drafted before the era of analytics, so I'm not sure what your point is.
That said, it's funny...teams improve via analytics, then get more out of their players, and then those who are anti-analytics claim that it's a function of talent. It's a circular argument.
Look at Golden State. Kerr took over for an anti-analytics coach and took the exact same core of talent, and are now on pace to not only improve by 15 games, but to go from 51 wins to 66 with a point differential of +10.2, which is a lot harder than going from 25 to 40. But now Golden State is "the most talented team in the league." It's the same group of guys!
Same thing is true with Atlanta. I actually had someone tell me the other day that they've gotten better because they have 4 All-Stars. I give up.
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
No, it's just good drafting. Analytics may have played a role in the drafting, but none of us know how much of a role.
It's funny how this whole thread quickly dissolve into people "proving" analytics failed or succeeded by just naming good or bad draft choices. All that shows me is few people in this thread have any idea what analytics are.
Dean Oliver:
Quote:
Do you think the Spurs have stayed on top simply because of talent, aging talent, that has been predicted to fall off the last eight years? The Kawhi Leonard pick (the Spurs acquired him in 2012) was a huge analytics pick. It was very evident in the numbers.
If the argument was made by a pioneer in the field do you still think they have no idea what they're talking about?
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
If you're going to dismiss analytics because you can make comments like that, then why not dismiss the importance of good coaching as well? Let's see Popovich try to win without those players.
People would look silly if they made the same comments about coaching that they make about analytics. Who needs good coaching anyway? It's about talent. Coach so and so coached a team that lacked talent and lost so good coaching doesn't work. Guys like Kobe and Lebron won rings so coaching doesn't matter. You can take a good coach if you want. I'd rather have talented superstars.
OKC used a lot of analytics and that "smart thinking" and it led them to the bone headed decision of keeping Ibaka and Perkins over Harden. As I said at the time as an outsider that just watched the game with my eyes, it made little sense in trading Harden and paying Ibaka instead. However they did their math and determined it was better in the long run to mathematically formulate that ibaka, Perkins, durant and Westbrook would be better. I'm sure they would do this again. After all it's analytics.
I don't understand the temptation to criticize analytics because it doesn't lead to perfection. If people hold the importance of talent to the same standard, then maybe talent doesn't matter because if talent always wins out, then OKC would be champs every year.
Quote:
However they did their math and determined it was better in the long run to mathematically formulate that ibaka, Perkins, durant and Westbrook would be better
They added Kevin Martin when they traded Harden, and Martin did very well for that team. I don't think keeping Ibaka was as bad as you think because Ibaka is really good, is a big man (unlike the other stars on the team) and it was pretty obvious to anyone who watched the playoffs last year that Ibaka was really important on that team. I'm not sure they beat Memphis or the Clippers without him. They didn't do well against the Spurs without him.
Joined: 02 May 2005 Posts: 90307 Location: Formerly Known As 24
Posted: Sun Mar 08, 2015 2:01 pm Post subject:
Trading harden was a financial decision, and since they had two ball dominant guards, they chose one over the other. I would have traded Westbrook (i feel harden works better with durant), but i respect them keeping the guy they feel is the better talent, and a two way talent,malthough inwould have waited one more season to make the move. But the only analytics that were heavily in play were luxury tax analytics. _________________ “We must always take sides. Neutrality helps the oppressor, never the victim. Silence encourages the tormentor, never the tormented.” ― Elie Wiesel
Trading harden was a financial decision, and since they had two ball dominant guards, they chose one over the other. I would have traded Westbrook (i feel harden works better with durant), but i respect them keeping the guy they feel is the better talent, and a two way talent,malthough inwould have waited one more season to make the move. But the only analytics that were heavily in play were luxury tax analytics.
They always could have amnestied Perkins but Bennett didn't want to pay somebody to not play for him. It's that simple.
Joined: 23 Jun 2005 Posts: 8488 Location: The (real) short corner
Posted: Sun Mar 08, 2015 5:55 pm Post subject:
Quote:
OKC used a lot of analytics and that "smart thinking" and it led them to the bone headed decision of keeping Ibaka and Perkins over Harden. As I said at the time as an outsider that just watched the game with my eyes, it made little sense in trading Harden and paying Ibaka instead. However they did their math and determined it was better in the long run to mathematically formulate that ibaka, Perkins, durant and Westbrook would be better. I'm sure they would do this again. After all it's analytics.
But you can't use that trade as an indictment of analytics in general, not knowing how much its use was involved, and especially considering the team who they traded him to.
OKC used a lot of analytics and that "smart thinking" and it led them to the bone headed decision of keeping Ibaka and Perkins over Harden. As I said at the time as an outsider that just watched the game with my eyes, it made little sense in trading Harden and paying Ibaka instead. However they did their math and determined it was better in the long run to mathematically formulate that ibaka, Perkins, durant and Westbrook would be better. I'm sure they would do this again. After all it's analytics.
But you can't use that trade as an indictment of analytics in general, not knowing how much its use was involved, and especially considering the team who they traded him to.
They took in great detail as to what Ibakas impact was from an analytics based approach in contrast to what they could have done with Harden and no Ibaka.
If they keep Harden they have to unload Ibaka. They thought they could be ahead of the curve by keeping ibaka and replacing Hardens impact with another SG. There was a lot of calculation done as well, not just the math on Hardens contract vs Ibaka. Analytics were a part of the equation, no doubt.
As I said in the first post, I'm not against analytics like some, but the number one thing you should judge a players ability on is your own eyes. In fact it's much easier to do so with talented players (hof talent). When it comes to role players and trying it pick out of the B grade talent, that's where analytics help a great deal. You think Jerry West needed analytics to see that Kobe was a rare breed of player? The stats probably didn't even favour keeping Kobe over Eddie Jones at the early stages of KBs career. That's why you also need a visionary and someone who understands the game is about - and can do so without needing numbers to prove his case. Analytics should be used but not as the be all and end all. Sorry, don't see basketball as is if it were baseball.
Quote:
I don't understand the temptation to criticize analytics because it doesn't lead to perfection. If people hold the importance of talent to the same standard, then maybe talent doesn't matter because if talent always wins out, then OKC would be champs every year.
I'm not criticizing it. I'm saying its not the be all and end all. I rather have Jerry West as my GM with his vision than Morey. Morey is a fine GM, but I think he misses the importance of things that Jerry Wests vision can gather. The basic point is you certainly don't refuse the data. You take it in, but not IMO as your primary source of evaluation of a player. Especially not the greater players in the league.
Joined: 23 Jun 2005 Posts: 8488 Location: The (real) short corner
Posted: Sun Mar 08, 2015 7:42 pm Post subject:
wolfpaclaker wrote:
fiendishoc wrote:
Quote:
OKC used a lot of analytics and that "smart thinking" and it led them to the bone headed decision of keeping Ibaka and Perkins over Harden. As I said at the time as an outsider that just watched the game with my eyes, it made little sense in trading Harden and paying Ibaka instead. However they did their math and determined it was better in the long run to mathematically formulate that ibaka, Perkins, durant and Westbrook would be better. I'm sure they would do this again. After all it's analytics.
But you can't use that trade as an indictment of analytics in general, not knowing how much its use was involved, and especially considering the team who they traded him to.
They took in great detail as to what Ibakas impact was from an analytics based approach in contrast to what they could have done with Harden and no Ibaka.
If they keep Harden they have to unload Ibaka. They thought they could be ahead of the curve by keeping ibaka and replacing Hardens impact with another SG. There was a lot of calculation done as well, not just the math on Hardens contract vs Ibaka. Analytics were a part of the equation, no doubt.
As I said in the first post, I'm not against analytics like some, but the number one thing you should judge a players ability on is your own eyes. In fact it's much easier to do so with talented players (hof talent). When it comes to role players and trying it pick out of the B grade talent, that's where analytics help a great deal. You think Jerry West needed analytics to see that Kobe was a rare breed of player? The stats probably didn't even favour keeping Kobe over Eddie Jones at the early stages of KBs career. That's why you also need a visionary and someone who understands the game is about - and can do so without needing numbers to prove his case. Analytics should be used but not as the be all and end all. Sorry, don't see basketball as is if it were baseball.
No one is saying that basketball is going to be like baseball. It's a case of having a tool and using it versus not having it. You can have a good tool and make a bad decision. In this case Houston made a good decision, and OKC made a bad decision, and they both used analytics to varying extents.
Jerry West doesn't need it to do what he does because he's brilliant. But the Warriors do make a lot of their decisions informed by analytics, and the results speak for themselves.
I agree - use it. Never have I agreed that a GM should throw the numbers out of the window. But vision is just as important. I'll take a visionary over a stats geek who thinks he knows the game but really hasn't even played it at a high level, any day of the week.
Ideally you have both inputs at a high dose. You have the numbers guy as your assistant GM/GM under president or something. And you have the guy on top having the vision and cultural influence.
Miami I believe uses it a great deal. They have balance. Riley's vision and cultural influence from the top. With great analytics guys working for him. That's what I read at least. That they have some great numbers guys working for Riles.
Joined: 23 Jun 2005 Posts: 8488 Location: The (real) short corner
Posted: Sun Mar 08, 2015 8:06 pm Post subject:
wolfpaclaker wrote:
I agree - use it. Never have I agreed that a GM should throw the numbers out of the window. But vision is just as important. I'll take a visionary over a stats geek who thinks he knows the game but really hasn't even played it at a high level, any day of the week.
Ideally you have both inputs at a high dose. You have the numbers guy as your assistant GM/GM under president or something. And you have the guy on top having the vision and cultural influence.
Miami I believe uses it a great deal. They have balance. Riley's vision and cultural influence from the top. With great analytics guys working for him. That's what I read at least. That they have some great numbers guys working for Riles.
Yeah, Riley was in some ways a pioneer in using analytics. Here's an interview with Spoelstra talking about, it and has this great line:
Quote:
NBA.com: So Pat was the one who originally drove you into statistics?
E.S.: He always wanted to crunch numbers and look at different ways of doing it. So he met with me one time and said, "I want more information. I don't know exactly what. Find somebody who can write a program, and take it where you think it can go."
Riley has been so impressed that even amid his team's own season, amid his scheduling of visits to conference tournaments in his dual role as team president, he commissioned a study from his staff about what makes these Rockets tick.
What he found was a team playing stifling defense, with the shot-blocking of Yao-replacement Dikembe Mutombo just a small part of the overall defensive equation.
"We really studied the numbers the other day during the streak, and you find out they have what we call a lot of perfect-fit guys," he said. "There's a lot of good defenders. There's a lot of role players. There's guys that move the ball. There's one absolute superstar who's only averaging 24 points a game during this run."
And that's what has so impressed Riley, in many ways establishing a blueprint for his team's future. It hasn't been all Tracy McGrady, just as Riley knows it can't be all Dwyane Wade in South Florida in coming seasons.
What Riley sees is the type of supporting cast his team has lacked this season, the bit players who make a difference.
Drafting Tim Duncan, Tony Parker, Manu Ginobilli, and Kawhi Leonard is some snazzy analytics
I wonder how reliable stats compiled from 2 years at San Diego State are? Isn't the sample size a bit small to be basing your franchises future on? What about the 1 and done players? I think drafting is still as much a crap shoot as it's always been.
Most of us present to observe Kawhi's almost 70 games at SDSU believed that he had the capacity to thrive in the association. Would the institution or data size have been more compelling if he was at Kentucky or Memphis? If that's your position, you might care to explain how he's proven to be that much better than guys like MKG or CDR ... or cancel that thought: maybe you can just explain the career arc of someone like Scottie Pippen from Central Arkansas.
SDSU has been a credible program for 10+ years; your comment seems to suggest otherwise ...
It wasn't a question of where he went to school, rather the sample size. 70 games isn't a lot is it, when compiling advanced stats to project future success?
Simply watching Leonard play isn't what the analytics is based on, so your observation here isn't part of the discussion.
I agree that ~ 70 games is a less-than-ideal sample size. But observation entails more than mere "watching". Don't forget, most of the essential advanced statistics at the current heart of this debate are based on data compiled via observation by an official scorekeeper. I'd also point out that the Spurs FO probably wouldn't agree with you that trading for the rights to the 15th player selected in the draft rose to the level of "basing the organization's future on".
I don't know to what degree analytics factored into San Antonio's ultimate decision to part with George Hill in the trade to acquire Kawhi's draft rights. I suspect it played a meaningful role, though, because they are known to have valued him over a guy like Derrick Williams in the same draft class. You might recall that Williams passed the eyeball test with flying colors, was flashier and had better overall box score stats at U of A, and went second in the 2011 draft. And he's also been a colossal failure thus far.
By the way, I didn't think we were discussing the emerging analytics that will be derived from SportsVU, big data projects, and the like. So if the point was statistical insignificance, it might have been more insightful to point to Tony Parker, who participated in one event on U.S. soil (Nike Hoop Summit in 2000), played in less than 55 games in France, and then was drafted in the latter part of the 1st round in the 2001 draft ... analytics may not have been the hottest topic in 2001, but Hollinger and his ilk have been at this since at least the mid-90s ...
It wasn't a question of where he went to school, rather the sample size. 70 games isn't a lot is it, when compiling advanced stats to project future success?
Simply watching Leonard play isn't what the analytics is based on, so your observation here isn't part of the discussion.
I agree that ~ 70 games is a less-than-ideal sample size. But observation entails more than mere "watching". Don't forget, most of the essential advanced statistics at the current heart of this debate are based on data compiled via observation by an official scorekeeper. I'd also point out that the Spurs FO probably wouldn't agree with you that trading for the rights to the 15th player selected in the draft rose to the level of "basing the organization's future on".
I don't know to what degree analytics factored into San Antonio's ultimate decision to part with George Hill in the trade to acquire Kawhi's draft rights. I suspect it played a meaningful role, though, because they are known to have valued him over a guy like Derrick Williams in the same draft class. You might recall that Williams passed the eyeball test with flying colors, was flashier and had better overall box score stats at U of A, and went second in the 2011 draft. And he's also been a colossal failure thus far.
By the way, I didn't think we were discussing the emerging analytics that will be derived from SportsVU, big data projects, and the like. So if the point was statistical insignificance, it might have been more insightful to point to Tony Parker, who participated in one event on U.S. soil (Nike Hoop Summit in 2000), played in less than 55 games in France, and then was drafted in the latter part of the 1st round in the 2001 draft ... analytics may not have been the hottest topic in 2001, but Hollinger and his ilk have been at this since at least the mid-90s ...
I isolated Leonard here because the article someone linked had a quote from expert, specifically saying that the Spurs' Leonard pick was a "huge analytics pick, that was evident in the numbers". Also in the article were other analytic experts associating the Spurs heavy use of analytics to build around their aging stars as an example of successful implementation of analytics on a franchise level. This lead me to believe that the Spurs do indeed place a lot of their team's future success on analytics, thus the "future of their franchise".
I agree on Derrick Williams. I thought he looked "NBA ready" coming out of college, especially in the tournament. But guys that stand out in college tend to go early, so the Spurs probably would never have the chance to pass on him anyway, and I don't think they would have.
I think franchises with top 3 picks in the draft tend to go for the observably talented players. It would be tough to pick a Kwahi Leonard with the 2nd overall pick and then say "our analytics indicate future greatness here". Has any GM done that yet?
Joined: 23 Jun 2005 Posts: 8488 Location: The (real) short corner
Posted: Mon Mar 09, 2015 3:57 am Post subject:
Quote:
I think franchises with top 3 picks in the draft tend to go for the observably talented players. It would be tough to pick a Kwahi Leonard with the 2nd overall pick and then say "our analytics indicate future greatness here". Has any GM done that yet?
That was essentially what Cleveland tried with Anthony Bennett, though it was as likely to do with bad scouting and Mike Brown being a dumbass.
Everybody uses analytics, by some degree and by some terminology. But today the term means advanced analytics and yes some are successful and some are not.
Wolf mentioned bringing in Pau by using analytics. I doubt few, if any, couldn't see that Pau's numbers, the basic stuff, like points, FG%, rebounds and blocks were all good and that he was quite the missing peace.
Interestingly though, it wasn't until joining the Lakers that his, or at least some of his, more advanced analytics (that also tries to capture defense) jumped such as Win Shares WS/48, ORtg and Drtg. PER (not to say this does much to capture defense) was very constant and showed little change.
His RAPM was 4.2. 5.1 and 3.5 the three years prior to the Laks, was 4.7 (first year with Laks but split with Memphis) and then 4.2, 3.9 and 5.2: fairly consistent
His 82games (Dallas's stat guru) Production +/- and on/off +/- for those seven years were (3.9/1.0) (9.2/4.7) (8.0/-.5) ((5.8/7.3 Mem) (9.3/8.4 Laks)) (8.0/6.3) (9.3/3.7) (9.5/10.1). So here again, advanced stats show quite an improvement with the Lakers.
So, what do you use? None, some, all? What weight to each? There is not one single analytic out there. These are just some. If one guy uses one and another guy uses another, they might come to different conclusions.
Do you evaluate Grandal, now on the Dodgers, by pitch framing or by his collective ERA vs another catcher's on the team?
Regardless, back to Pau, I will have to agree with Barkley on this one. We didn't need analytics to know he was pretty good. _________________ "A metronome keeps time by using a Ringo"
It wasn't a question of where he went to school, rather the sample size. 70 games isn't a lot is it, when compiling advanced stats to project future success?
Simply watching Leonard play isn't what the analytics is based on, so your observation here isn't part of the discussion.
I agree that ~ 70 games is a less-than-ideal sample size. But observation entails more than mere "watching". Don't forget, most of the essential advanced statistics at the current heart of this debate are based on data compiled via observation by an official scorekeeper. I'd also point out that the Spurs FO probably wouldn't agree with you that trading for the rights to the 15th player selected in the draft rose to the level of "basing the organization's future on".
I don't know to what degree analytics factored into San Antonio's ultimate decision to part with George Hill in the trade to acquire Kawhi's draft rights. I suspect it played a meaningful role, though, because they are known to have valued him over a guy like Derrick Williams in the same draft class. You might recall that Williams passed the eyeball test with flying colors, was flashier and had better overall box score stats at U of A, and went second in the 2011 draft. And he's also been a colossal failure thus far.
By the way, I didn't think we were discussing the emerging analytics that will be derived from SportsVU, big data projects, and the like. So if the point was statistical insignificance, it might have been more insightful to point to Tony Parker, who participated in one event on U.S. soil (Nike Hoop Summit in 2000), played in less than 55 games in France, and then was drafted in the latter part of the 1st round in the 2001 draft ... analytics may not have been the hottest topic in 2001, but Hollinger and his ilk have been at this since at least the mid-90s ...
I isolated Leonard here because the article someone linked had a quote from expert, specifically saying that the Spurs' Leonard pick was a "huge analytics pick, that was evident in the numbers". Also in the article were other analytic experts associating the Spurs heavy use of analytics to build around their aging stars as an example of successful implementation of analytics on a franchise level. This lead me to believe that the Spurs do indeed place a lot of their team's future success on analytics, thus the "future of their franchise".
I agree on Derrick Williams. I thought he looked "NBA ready" coming out of college, especially in the tournament. But guys that stand out in college tend to go early, so the Spurs probably would never have the chance to pass on him anyway, and I don't think they would have.
I think franchises with top 3 picks in the draft tend to go for the observably talented players. It would be tough to pick a Kwahi Leonard with the 2nd overall pick and then say "our analytics indicate future greatness here". Has any GM done that yet?
I don't know whether Popovich's substitution patterns and lineup permutations are really based on the heavy use of analytics as has been suggested, but I do believe that the drafting of RC Buford is based heavily on analytics.
Kawhi was ranked as high as #6 on some draft boards, as I recall. In the end, some teams chose other options ahead of him for various ill-advised reasons (e.g., Biyombo because the Bobcats [via Sacramento] overvalued raw athletic potential; Jimmer because the Kings [via the Bucks] overvalued his offensive production at BYU and underestimated his defensive deficiencies). However, I don't agree with you that Williams sat higher than Kawhi on the Spurs draft board. I look at this situation through the lens of an Aztecs backer, so I'm not as informed as many. But it's true that the Spurs have not really been in a position to draft #2 since TD was drafted so many years ago.
Buford and his team could have traded up for Williams had they valued his rights. They had the assets to entice the Wolves into a trade. Instead, they valued Kawhi's potential based on what they had seen of his brief stop at SDSU, and despite his lackluster pre-draft "measurables" (with the exception of the hands) ...
I do agree with your last point, though. In this case, Buford freely admits that nobody on his team envisioned that Kawhi would have such a resonant effect on the organization (currently #7 in RPM, leading the league in steals per 36 min., best rebounding SF in the game, Finals MVP in his 3rd season, probably the best overall defensive SF in the game, etc.). He's looking like a steal at #15 ... Kyrie and perhaps Klay are the only two selected in that draft class that I would consider with a pick over Kawhi if a time machine was available. But he's significantly ahead of either of them on the defensive end, so it's a tough call.
All times are GMT - 8 Hours Goto page Previous1, 2, 3Next
Page 2 of 3
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
