He said no such thing to insult Spoelstra. I get the feeling that you probably didn't pay attention or watch past the first few minutes, since that's where the brief Lebron and Battier comparison popped up. Starting with a few simple, easily understood points in your introduction is basic writing. People are aware of differences within traditional basketball positions, but this seemed to be about quantifying things and adding details. He also repeatedly stated that this was a simple, initial step. Some of the applications that he hoped for are what you're asking for here. Such as figuring out winning ways to construct a team, which lineups to play, and more informed decisions on which type of player to match up against which.
It is not, he's just claiming that it is. He used data collected by other sources so he hasn't quantified anything. All he's done is "discover" that some players are good at different measurable achievements and a few are good at all of them. He hasn't done anything new or even added insight to what already exists, he's only produced pretty graphs. Yet he said nothing to suggest that he knows how to go about that. In fact, he showed precisely the opposite. It was clear that people in the audience, and people here as well, are far ahead of him intuitively. Did he really discover that some players are shooters for the first time? Ramming the same old data into a new visualization tool is only helpful when it provides additional insight. 13 positions instead of 5 isn't a new insight, it's a failure of insight. It's a case of the tail is wagging the dog. If you want to advance understanding you have to figure out what to measure. Sometimes that means measuring new things and other times it means deriving data out of existing measurements. Thinly disguising rebounds, shots, and steals as rebounding, shooting, and defense is a fraud and an arrogant one at that. This guy is full of himself, his only skill is self-promotion.
I feel that a study has to provide some benefits to be useful. I don't see anything useful with what he said. Nevermind that he shouldn't have said "positions" but rather "skillset." Even discounting that, what's the point? If there were further studies done, like which skillset performs optimally with a player of a different skillset, then he'd be on to something. For example, on the Rockets, which players would perform best when Scola's on the court as opposed to Patterson? What about Lee vs KMart? Of course these are things that coaches can figure out. But it may be helpful if a team is going for a trade and looking at an optimal player to add.
Its no different then a batter watching the spin on a baseball, although I suppose its easier with the ball coming at you. I was a great rebounder because, while I couldnt see the laces, I could spot the rotation of the ball.
So compiling, grouping, and analyzing data isn't quantifying things...because the raw data originated somewhere else. He showed that based on raw box score statistics, the vast majority of players fall into one of 13 distinct categories. You're dismissing all of his work just by reducing everything down to the vaguest, most basic point, that players are different. Yes, but HOW different? By how much? In how many different ways? How are the various teams constructed, and which ways tend to win more against which other types of lineups? This would be like someone doing a detailed analysis of the best places on the floor to focus your offense, and you dismissing everything because everyone already "knew" that some shots are harder than others. Uh no, he actually said that with their software, they could easily plug in statistics going back decades to come up a profile on all the teams. From there, it wouldn't be hard to look for correlations between certain kinds of rosters, and winning %. Also, people who analyze the performance of specific 5-man lineups, and make game flows, could combine that data and group different teams based on statistical classifications like this, to look for trends in performance. And yet he won a conference award there for "best Evolution of Sport." So clearly, people there liked what he was on to. Well no duh some shots are easier than others, so screw further analysis! Forget about bother looking at shot charts and efficiencies! Now you're at the level of talking trash. He's a "fraud" now? I also like how you so casually ascribe personality traits to him like "arrogant" and "full of himself." I could just as easily say that your posts show that you're "full of yourself."
No, not unless something quantitatively new comes from it. And those categories are how good they are at the specific metrics. The data says that directly. Nice straw man. Nothing I've said is predicated on "players are different". I dismiss the guy's work because he has provided no insight. A player who rebounds and defends well being classified as playing the "defensive rebounder" position isn't insight, it's hubris. He didn't answer those questions, he posed them...as have others before him. It would be if he claimed that he had discovered the never before realized fact that some shots are harder than others. And you believe that's going to produce the answer? Did he do that? That is greatly disturbing and says a lot about the quality of information available. Another nice strawman from you. You are bitter. Yes, the guy is an arrogant self-promoter and his claim that this represents new research is fraudulent. He is a college undergrad working with tools developed by others which are over his head. He is nothing more than a glorified lab tech entering data into software but unable to realize the lack of insight that results. If people this represents innovation then that tells you the sad state of the art.
Please try to argue the points, not the speaker. The question is, will any of this information help people play basketball better, or manage a roster more effectively? If the answer's no, which is my conclusion, then move on, nothing to see here, next story.
Excuse me, you said "some players are good at different measurable achievements and a few are good at all of them." BIG difference. What's hubris is your posts. You're continuall bashing him, insulting him by making up a personality for him, all without any real point. His insight is that players can fall into quantifiable classifications based on performance, and he suggests that we can use this to analyze team performance and as a basis for drafts, trades, etc. And "others" didn't present the data in the way that he did, to the extent that he did. Can you tell me what kind of point this is supposed to be here? He explained how his stats could be applied, I repeated it to you...and your supposed rebuttal is to play ignorant repeat your question? He said that this is a STARTING point. He's laying out an approach to things. This would be like someone way back coming up with the bright idea of tracking individual player performance and suggesting that people do that from then on, and you bashing him for not doing everything by himself, RIGHT then. "Greatly disturbing." Seriously, your word choices, combined with your near total lack of a point, is just odd. Oh man, you just absolutely RUINED my day! :grin: Your idea of a rebuttal is to call me "bitter," when you're the one with an axe to grind here. When you're the one going out of your way to put down and insult this guy, who you've probably never met and who's done way more work in this subject than you have. Your posts have no substance. Just insults, pointless complaining about how he hasn't already done everything akreadt despite clearly proposing a mere starting point, and simplistic reductions of everything down to vague and useless statements. I will no longer be responding point-by-point with you, if this is how you are going to act. Take a good look at yourself. You come across as quite the "bitter" and "arrogant" person. Of course, it seems very easy for you to throw those labels at someone presenting work at an actual conference, when you're just some guy with a handful of posts on an internet forum.
I agree with this part. csj seems to have latched on to this speaker as some kind of target for him to insult and make himself look better than. How have you reached this conclusion though? Have you put his methods to the test, and ran the numbers? That's the problem I'm seeing here. People are quick to dismiss this guy right away. I'm even detecting some strong anti-intellectualism here, with some people saying that he needs to "dumb things down," or that we should all just rely on "intuition." They fault him for shortcomings to his admittedly initial work, or denounce him for not automatically having finished all the work himself and attained the wisdom of God. Meanwhile, the alternative we have in place are five meaningless and almost arbitrary positions, which many coaches think they have to comply with. To the extent that no-skill seven foot stiffs see playing time just because the team "needs" a center.
There are only 5 positions because only five can play. There are several skill sets. A pocket passer and a scrambling quarterback are still both quarterbacks.
I've been talking about his presentation while you have been talking about me. There is a point to to understanding who the speaker is. You can't separate the message from the motivations of the person who provides it. Right, and furthermore he says that this is new when it is most certainly as old as any statistical analysis. He didn't answer the questions you raised nor did he present any "data" in that regard. His stats? What were his stats? He used common stats and derived common conclusions from them. Doing more of the same would accomplish what? Yu should spend more time actually thinking about what he said and less time going hyperbolic on criticisms of it. Laying out an approach that is different how? No, it's not like that at all. He's taking shot statistics, classifying players on how good at shooting they are, then presenting it as new. Lather, rinse, repeat. It is common online for people to levy personal attacks and then cry foul when it repeated back at them. You are attempting to turn this into a personal argument against me rather than talk issues. I've presented a case for my claims and as far as experience goes, you do realize this guy is an undergraduate, right? He is pre-intern! Look who's talking. Typical projection. More projection. And more personal attacks. I typically don't post here because of the intellect of posters like you. Chest pounding is the only way you can make your point.
It is you that is interpreting criticism as being opposed to statistical analysis. "Dumbing down" is not anti-intellectual nor are statistics and intuition mutually exclusive. These are more of your strawman arguments. You have no evidence of this. There are fewer than five positions in basketball and I suspect most, perhaps all, coaches understand that. This presentation provides no alternative to that.
Really? Because he took a lot of time to say something along the lines of: Battier and LeBron play same position, Spoelstra must have a hard time during substitutions because he can't tell the difference between LeBron and Battier! Not to knock on the guy, but after a year of training camp, practices, games, and getting to know the team. I'm pretty sure Spoelstra knows his players well enough without the need to use advanced positional topology. Honestly, the study is cool but entirely useless.
csj, I told you I would not bother quotingpoint-by-point with you if you keep acting this way. Honestly, even that statement was being too generous to you because you don't seem to have a point. You are acting in the most childish way possible here. Your idea of a rebuttal actually seems to boil down to kindergarten level "I know you are but what am I?" You talk a ton of trash about this guy, and when I point that out I'm suddenly "bitter." When I point out that unfounded "bitter" insult as well, now I'm the offensive person here talking trash and trying to turn things around on you? Yeah, you're such an innocent victim here. Your entire attack on this guy seems to be based on: 1) Made-up statements about his supposed personality 2) Saying what amounts to "no duh" and faulting him for not having everything done already even though he's merely suggesting a starting point. You claim that he derived "common" conclusions that were already obvious, when you could not have made as specific statements as he did before. Really, you knew that based on stats, most players fell within 1 of 13 categories before? You suggested that players be classfied based on stats, that these stats should inform lineups and matchups?
No, you accused him of calling Spoelstra "mentally r****ded" and unable to differentiate LeBron and Battier. Which was not what he was saying. What he was saying was that LeBron and Battier, both arbitrary "small forwards" are the types of guys who would be substituted for each other. Even though they both have extremely different skill sets. Spoelstra was just an example. We all know that that's how many coaches run their substitutions. A SF will come in for a SF, a C will go in for a C, etc. Maybe if LeBron sits, another ball handler should be substituted in so that the skill isn't completely lost from the lineup. Maybe instead of arbitrarily putting in a 7-foot scrub as the "center," coaches should just put in a player with a skill profile that fits with winning basketball, as determined by statistical analysis. Now I haven't been following the Heat all year, but in these Finals Spoelstra seems to have abandoned "traditional" basketball lineups. "Center" Joel Anthony, who started the vast majority of regular season games, has seen a total of 2 minutes in 2 games. Battier and LeBron, both "small forwards," are both starting now and are both averaging over 40 mpg.
Everyone loves to see quote arguments that go line by line back and forth with each other! But seriously, this really does provide no new information nor does it reveal anything that any avid basketball fan didn't already know.
How do you know whats "right" and "wrong" in positions? Agree that you shouldnt keep things RIGIDLY defined by a position's description on-paper. Like running QBs in the NFL, they don't HAVE to be pocket passers. So what you're saying is....positions and skillsets are "versatile". The next EVOLUTION of basketball on that one.
Exactly. The point of studying these things is that they often dispel what was once considered obvious and accepted. Maybe 90% of this analysis can be classified as known information, but he uncovers 1 or 2 unusual trends. That makes it worth the trouble. I would actually be much more curious if they did the groupings with high level stats rather than conventional categories. That seems like it would be a lot more useful.
This is semantics. You're saying the same thing. Naming things can often help describe and understand them better.
Actually the article was first published some time ago. I found it interesting. If I remember correctly the thrust was - "if we further refine the definition of position in the NBA, which ones produce the best results in combination with which other refined postitions"? It certainly seemed to offer some insight into which types players are best paired on the court. It was not a refutation of conventional basketball wisdom, but an intellectual refinement. But then it may just be the statistician in me.
