I wrote an eMail to 82games.com asking them to provide better analysis of the wonderful statistics they publish. Roland@82games Response: I requested to publish his response on this board: Roland@82games response: It seems we have drawn attention all over!!!
I'm an idiot with numbers, but it's almost like those posters with the hidden image that you have to find: Just keep staring at the screen, and you'll figure it out. I've been pouring over that site for three years, and I've just gotten the hang of it. Just keep going over and over it, and you'll start to figure it out. And next season, every defensive possession will be charted. Awe-some.
I love 82games stats.. i did my own detailed stat track one day for mike james when he first arrived here. It took me 4 hours it's a lot of work but it pays off.
I don't mean to confuse the folks on the board that just prefer to have opinions. Present company excepted.
...they saved a lot of money on car insurance? Sorry, couldn't help myself Interesting. I have mixed feelings about their posting conclusions, but I must admit that I don't get nearly as much out of the stats as I probably would if they explained some of them to me.
Ya, I've signed up to be one of their charters for next year. I worry that it will change the way I watch the games and I won't be able to just enjoy the game. any more.
As someone who has "had" to watch games for years while "working," believe me, you'll be just fine. Crap, I've had to scout players on Central Connecticut State and the like for scouting services, writing down page after page about players who'd be lucky to walk on at NC State, and it was still fun. You'll dig it, and believe you me, your work is appreciated.
I'd prefer they limit their practice to presenting the statistics. Often time (actually almost always), there are many interpretations to statistics. Insisting on getting a conclusion is ridiculous. Not only I don't trust these guys in getting all possible interpretations, but also I don't believe their personal biases won't affect their pick among possible interpretations. If the audience is familiar with statistics, and can read the analysis with critical eyes, it would've been fine. But that is not the case. Most people who visit that site are going to quote what they read there as if it were bible.
I am sorry that you believe they will use that data in a malevelant manner. IMO most of the people already misquote what they read there as if it were bible.
I didn't mean to say "malevelant". "subjective" may be better word to describe it. Yeah, people are already misquoting data, but that reflects their own personal interpretation of the data. As the provider or the data, you want to leave room for audience to interperate data in different ways. Just to give you an example. Federal Reserve provide public access to its comprehensive data sets on economy, but you won't find any subjetive opinions post on their database websites. I am alreday tired of people abusing the data. Let's not to make it worse.
I think I understand what you are implying. But, subjective interpretation of their own data [that was my question, not their answer: "we'll be overhauling the site for 05-06"] would have to be with malevalent intent, wouldn't it? There are perhaps only a handful of persons out of the 100k people on this board who have the tools to 'properly' interpret their statistical data. Rox management probably can, since I heard a quote attributable to JVG that smacked of 82games type stats. I majored in math which included alot of stat courses and I couldn't possibly come up with analysis that I would live or die by. People quote the site all the time on this board and don't know what they are talking about, so I would be quite happy to have some hint of what the data means. Like someone above said "look at the data long enough and thngs begin to mean something". I'm just Lazy and Dumb, I guess. I don't mean to 'rant' on you. This was suppose to be a happy thread. Sorry!!
Not trying to create a conflict, but I'm curious as to your examples of any writers or fans that you've seen abuse the data 82games.com provides for us.
"abusing" might have been too strong a word for this. But we have seen fans quoting various indexis to judge who is a better defensive of offensive players, or which team is hottest based on records and schedule, etc. Honestly all those statistics are very preliminary and problematic. They lack proper statistical controls (if there is a way to do it to start with).
If they're quoting Hollinger's defensive PER system, then they're probably off. The system is in its infancy, Hollinger admits this, and tends to overrate the big steal/block guys. Manu Ginobili looks like a star, while Kirk Hinrich gets the shaft.
I didn't mean to make this an unhappy thread. Sorry if that is the impression I left you. My work is statistical analysis, mostly on economic and financial data. To tell you the truth, often time there is no proper control. You just have to leave the data alone and let people interpretate them in different ways, because there is no way to test which interpretaion is correct. In principle, I'd like the provider of the data stay out of the interpretation game. For example, the bush administration can point at some statistics and say "the economy wasn't so bad in the first 4 years" and democrats can come up with the opposite conclusion using the same set of data. There is no point and it will be bad for the Fed to get involved into this spining.
