So someone looked at the stats, actually watched the game, and had some logical insights? Even yalls boy Soul Match Malloy is too busy to be writing articles like that. You don't see many articles like that. Must be plagiarism.
82games.com has a rough data... eFG within different shot clock usage... but they didn't have anything about offensive rebounding rate, turnover, foul drawing rate, etc depending on how much is left on the clock. The Rockets' eFG with 3 seconds or less on the clock is .449, comparing with an overall eFG of .499. They run the clock down to 3 seconds or less on 16% of the shots. I have not yet read all of the other team's stats. However, I was really making a point about how Rocket's performance last nigth compared to how they shot on contested shots for the rest of the season as time was running down. It seems they did better than they usually do.
I would venture to guess you're right. What's the most contested shot on the floor? Shots around the basket. And the Rockets are near the bottom of the league in efficiency on inside shots. That indicates a lack of athleticism and finishing skill under duress. To me, those are skills you need to be good at making contested shots at the end of the shot clock.
I think thats a logical conclusion. The only counter to that would be Reggie Miller. He was average at best finishing in traffic. But dude was money outside, with a hand in his face or not, especially against the clock. Ron Artest seems to hit a lot more than his fair share of contested shots late in the clock. Its the early ones that seem to be a big problem.
I found an online sofware that did a simple scatter plot and linear regression of several sets of data from this season: http://www.wessa.net/slr.wasp http://www.basketball-reference.com/leagues/NBA_2009.html 1. Points Per Possession on Offense (Y-Axis) vs. Possessions per game (X-Axis): 2. Points Per Possession on Defense (Y-Axis) vs. Possessions per game (X-Axis): 3. Point Differential Per Possession (Y-Axis) vs. Possessions per Game (X-Axis): A few thoughts based on the very simple graphs: First, the data points are certainly rather scattered, even if there is any kind of a trend, there are significant variations from it. There are slow paced that are both good and bad on offense and defense, and fast paced teams that are both good and bad on offense and defense. Second, there pretty mucn no evidence shoting that faster paced teams scored more efficiently this season. To the extent there is a "trend line" the slope is essentially flat (0.031). Third, there seems to be somewhat of a trend that, defensively, the faster paced teams have sucked more with a trend line slope of .432. Fourth, combining the two points above, in terms of points differential per possession, the faster faced teams, to the extent there is a trend, have done worse (slope of -.401). Though there are certainly teams that have done well playing any kind of pace. I didn't get into the coefficient of correlation, but that can certainly be calculated fairly easily. EDIT: Follow up: Correlation coefficient between Possessions Per Game and... Points per possession on Offense = 0.032685149 Points per possession on Defense = 0.374759439 Point Per possession Differential = -0.227825909 Not suprising.
Oh, i understand the difference. I think it's useful to see how a player MAY play given more minutes/faster pace but I think a common assertion is that Player X is better than Player Y because their numbers are better on a per minute/pace adjusted basis. I think it's only relevant for people who have a lack of appreciation for what a player does in a certain time frame. It becomes misleading at a certain point. EXAMPLE Chris Anderson plays 20 minutes a night and has a PER of 18.75. That's better than quite a few notable players (Caron Butler, Joe Johnson, Mehmet Okur, Paul Pierce, Ray Allen, Carlos Boozer, Landry, Scola, Lamar Odom, etc.) The Birdman has performed well while he was on the court. On a per minute/pace adjusted basis Chris Anderson was a better player than the players listed above. While technically true, if you agree with PER...isn't that misleading? I'm definitely not saying that box score stats very useful on their own. Quite the contrary, i'm saying no stats are useful alone. They only become useful with good analysis, anything can be manipulated to reflect a desired opinion. ...and, i find little wrong with pace adjusted stats for an entire team, it seems much more relevant. The variables don't disappear but they seem to cancel out. There may be an efficient player who produces in limited time but may not be able to sustain that level for a long period (like Carl Landry) and there may be players who are seemingly less efficient but play most of the game (like Shane Battier).
Sure, that's misleading. I think PER has its use, but shouldn't be blindly used to argue one player is better than another. If I'm a GM and I need to choose between bringing in two players at simitlar salary, its not just about their numbers, per minute or per game. You have to have an idea of what role they'd fill, and how well they'd fill it. Now, if I'm looking at the boxscore to get a quick idea of what they're good at, I'd probably separate it into MPG and than a bunch of per-minute stats, instead of just looking at per game numbers. So, that tells me both their productivity rates, as well as how long they're playing each game. I think that's a more useful way of framing the information in the boxscore.
