Would love to hear some feedback. Anybody have a better way of determining how much of a team's offense a certain player accounts for? _________________ NBA.com
Joined: 13 Oct 2005 Posts: 463 Location: Atlanta, GA
Posted: Thu Jan 17, 2008 12:24 am Post subject: Re: The Blazers' Turnaround
Welcome!
schubee wrote:
Anybody have a better way of determining how much of a team's offense a certain player accounts for?
How about %Poss (aka "Usage"), the percentage of team possessions used by a player while he's on the floor? Kenpom talks a bit about it here; I think it's by far the best way to measure the size of a player's role in his team's offense.
Also, if you're looking not for just usage (which is independent of efficiency) but also the quality of those possessions used, it might be better to divide a player's individual points produced (see Basketball on Paper for the formula) by total team points scored.
Joined: 14 Jan 2005 Posts: 1712 Location: Delphi, Indiana
Posted: Thu Jan 17, 2008 8:53 am Post subject:
Quote:
..."added his points and assists (multiplied by 2.19*) and divided that by his team's points to see how much of their offense he was directly responsible for."
This gives dual credit for the assister and the assistee. The more assisted FG a team has, the more over 100% their 'portions of responsibility' will add up to.
This we call 'double counting'. You may assign some fractional credit for an assisted FG to the passer, and the remainder to the shooter.
You might say, for example, that an assist is worth 1 point, the scorer getting credit for the other 1 (or 2) for making the shot. The apportioning of credit is open to debate, and each play might have a unique pair of fractions.
Well, since I'm just comparing Roy of the first 17 games to Roy of the next 20, I'm OK with assigning extra responsibility, so to speak.
Thanks for the input though.
In general, I have issues with making the leap from team statistical analysis to individual statistical analysis (PER, etc.), because there is so much that goes on in a basketball game that isn't accounted for by individual stats. _________________ NBA.com
Joined: 31 Dec 2004 Posts: 958 Location: Durham, NC
Posted: Thu Jan 17, 2008 1:00 pm Post subject:
Overall I thought it was a really interesting article that covered most of the major topics and wove them into the narrative quite well. A few comments...
Mike G wrote:
Quote:
..."added his points and assists (multiplied by 2.19*) and divided that by his team's points to see how much of their offense he was directly responsible for."
This gives dual credit for the assister and the assistee. The more assisted FG a team has, the more over 100% their 'portions of responsibility' will add up to.
This we call 'double counting'. You may assign some fractional credit for an assisted FG to the passer, and the remainder to the shooter.
You might say, for example, that an assist is worth 1 point, the scorer getting credit for the other 1 (or 2) for making the shot. The apportioning of credit is open to debate, and each play might have a unique pair of fractions.
I was going to say the same thing, but Mike G beat me to it. Furthering the point, I would be curious what you came up with if you tried to determine how much of the offense each player was responsible for, and then totalled that up.
I would suggest you first figure out what % of the time Roy was on the floor during each period (First 17 games, Last 20 games), and then use that as the denominator. So, if during the first 17 games the team scored (hypothetically) 500 points during the time Roy was on the floor, and he directly scored 100 points, and assisted on another 120 points, you might try something like:
(100 + (120 * .5) ) / 500 = 32% of his team's scoring while he was on the floor.
Also, you wrote:
Quote:
an improvement of nearly three points per 100 possessions on the defensive end is significant (it's the difference between being ranked 24th in the league in defense and being ranked 11th)
Significant by what means? Or do you just mean "impressive"?
After reading your article I poked around a little.
One thing I found was that Roy / Aldridge are taking close to 40% of the shots and the Blazers are getting efficient shooting from the other 3 positions (better than from Roy and Aldridge in fact, but they set the table and FTs also matter). It is a successful team blend of stars and supporting cast, primary option / shot creators and those who feed off the action of the lead guys.
(That 40% of shots from top 2 is pretty consistent across the west. There have been articles about top 2s and 3s and some that look at rest of casts. I find those quite useful for getting a better sense of team construction and performance.)
I talked about it somewhere before but I think it is interesting that Roy and Aldridge have 2 of the 3 worst raw +/-s on the team and they are among the worst player pairs for each other (and Roy was one of Aldridge's worst pairs last saeason too). The defense is 4-5 pts worse with either of them on the court and the offense doesnt improve more than 1 pt. The context of shooters with them and good defense from others especially when they are off supports them as much as they support the rest of the guys with their primary offense.
Team results with both on court together are not stronger than with just one. I will be curious if next year with Oden in the frontcourt mix Nate spreads them out so one of these primary scorers is on court at all times and maybe the % of time with both on court together goes down. Maybe it won't but whereas sometimes there is synergy between stars so far there does not appear to be between these two. But the good news perspective would be that the Blazers appear to have enough to get the total job done and don't depend heavily on having both on court or both at top form to function well.
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
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
