A Stanford student presented at the Sloan Sport conference, claiming that after examining every player in the NBA with statistical analysis, there are 13 positions, not 5. Many present and former Rockets are mentioned...the video was pretty interesting. http://www.sloansportsconference.com/?p=5431
He lost me at .... "Shane Battier and Lebron James play sf but have vastly different playing styles." No ****.
It's hardly a secret that there are tons of different playing styles at each position. I'm not sure it would be a good idea trying to categorize all the different styles, but if you did you might as well do some research, so you don't end up with:
I think this guy thinks he is Bill James. Every single gm would laugh at this kid. This is stanford's best product?
He's stating the obvious...but getting "role" mixed up with "position" PG Point Forward Combo Guard SG Swingman SF Tweener Power Forward Forward/Center Center
Really? It seems to me that his analysis is overly simplistic. He's discovering things that are completely intuitive to any basketball person. We know that there are different styles of play at each position and that certain style groupings work better together than others. It's also problematic that this system doesn't take into consideration anything about matchups. I guess it's more of a macro analysis, but that would provide much more valuable information. Something that showed which types of groupings would work well against other types of groupings would be interesting. For instance, it seems overly apparent to me that Kendrick Perkins has been a huge negative for the Thunder thus far in the finals, because he can't matchup with anyone in the Heat's starting lineup. Also, he isn't a threat to score at all, so they can't take advantage of his size. He's just a non-factor out there because of his play-style type in that matchup. I believe that's been a huge factor in their slow starts. They almost always make a run when they go small. A metric that could quantify that would be helpful.
The guy said that a player with no abilities would be defined as a "One of a Kind", assuming there were no other such players. If there were enough of those players, though, they would define a 14th "position". He said exactly that when he described the result as 58 positions with 45 one-of-a-kinds. In other words, the game is defined by who plays it, not how best it should be played. Sad. It was interesting that the audience members seemed to softball obvious criticisms of his approach and he never understood that he was speaking to people who grokked the problem much better than he did. Instead, he "hypothesized" a result that was most surely incorrect and suggested that the answers would lie in adding literally thousands of independent variables, as though insufficient complexity was an issue! He claims his results are the "first ever" and are revolutionary, but the fact is that they are a testament to his failure to understand the data despite his tools. This guy can't tell the difference between player strengths and positions and draws false dichotomies because he confuses positional requirements with what the limited data is capable of saying. He's not nearly as smart as he thinks he is and has an overly inflated ego because of where he's from. His tools may be great but this is a case of garbage in, garbage out. He may be capable, but he needs to be taught what he's pretending to teach.
There are only 3 positions in the game, and a player can play several positions on the floor, depending on his size, speed, and skills: Post Wing Point
Lol. He makes the assumption that Spoelstra is mentally r****ded and can't tell the difference from LeBron or Shane. LOL. What a joke.
He might as well compute how many bounces the ball has or will have in a game, and which way the ball bounces from the rim at every shot. These are garbage stats as one poster already mentioned. Intuition in basketball has proven to be much more reliable than cold statistics. As an example, ask Rodman if he used stats to know which way the ball would bounce every time it hit the rim. He seemed to know where the ball would go every time. Stats are great to supplement basketball knowledge and intuition but they are totally bereft of any meaning if interpreted on their own or assigned with too many variables. The most unpredictable variable in any human endeavor is the human being itself, and this fact even applies to other forms of animals and their behavior. Respected statisticians in any field of study will concur with that fact. This guy should just use his great analytical gift to predict lotto numbers and also stock behavior at Wall Street. Basketball has just too much dynamic human element in it to be captured by stats.
i remember barkley saying once i think that he watched which way the ball was spinning as it came towards the basket. have no idea what rodman did to be so great at rebounding though, or him at least saying anything about it.
Haven't seen the video yet, but he's totally right if he's saying that traditional basketball positions are wrong. They're not even "positions," because players aren't restricted to actual spots on the floor. Players have certain tendencies, such as big guys generally playing closer to the basket. But even that's not set in stone, because player roles and skills vary dramtically. Baseball and football have positions. Basketball has vague labels based on a player's height. What's a small forward even supposed to be? It's funny how that gets lumped into "forward" along with power forward, even though small and power forwards are often nothing alike. The PF is basically a slightly shorter center, and at least there are classic stereotypes for what a PG, SG, and C is supposed to be like. I don't know if there's even a classical definition of a SF out there. We have guys who are basically shooting guards (Paul Pierce). Guys an inch or two taller than that, without the outside shooting skills. 6'8 players who can shoot. 6'10 players who could be playing a big man position, if they weren't so skinny and/or inclined to play on the outside. Role-based classifications should be the way to go. Someone shouldn't be a "small forward," he should be thought of as a "3pt specialist" or something like that.
Back in the hack-a-shaq days, there would be an extra position called "end-of-bench-scrub-whose-only-job-is-to-get-shaq-to-the-line"
This is basically a refusal to put effort into studying the game. What is this "intuition" you speak of? Can you define it? How do you account for guys who have vastly different intuition? Also, intuition can be flat out wrong. Many players take stupid shots, because of their "intuition." Synergy Stats tracks how well players perform at various actions, such as driving left or right. There are players who drive left all the time, even when they do far better going right. Old school players and coaches who were brought up in the pre-3pt era (or the early 3pt era) put down 3pt shooting as recently as a few years ago. They advocated more mid-range jump shooting, which was supposedly purer, more skillful, and more "fundamental" basketball. Historical stats have shown that mid-range jump shooting isn't much more accurate than 3pt shooting, and are actually far less efficient when you account for the extra point that 3pt shots offer. Now, quality teams build their offenses around lots of inside and outside shots, minimizing the inefficient mid-range shots. 3 > 2. But don't listen to that nonsense, that's just the evil "statistics" talking. Rodman was quite a student of the game: 1992 LA Times article "There are solid technical reasons to explain Rodman's success. He is a student, constantly watching tapes of the players who shoot the most. He looks for tendencies. When they miss, which way does the ball usually bounce?" Which guys "shoot the most" and "which way does the ball usually bounce" are observable events that can be easily quantified with stats. So what are you getting at, exactly? You put down statistical analysis, then say stats can be useful, then go back to saying basketball is too hard to analyze so we should all fall back on undefined "intuition." No one has ever claimed to know everything based on some numbers which only quantify certain things. It's about knowing more than you would if you were too lazy to even bother with counting.