APBRmetrics

I'm new, but have been reading through the forum for about a week. Having encountered some interesting threads/thoughts/references on basketball, I'm interested in learning a bit more about the metrics are originally constructed, modified and applied to get some background knowledge on the topic so I can make more sense of the content.
I've bounced this question around other forums and I was directed towards Hollingers Prospectus (03) and Basketball on Paper from Oliver, which I've ordered and am waiting on. The caveat to this was that the motivation behind the construction of the various formulas was not discussed in detail, which is a point of interest for me.
An example of one of my problems:
When I first started trying to break down the PER formula on my own, I kept running into the problem of seemingly arbitrary designations of value to statistical categories. The best explanation I could come to was that the components in the formula are constructed from conclusions drawn from data (which I couldn't find referenced). I'd like to know why the formula looks like how it does, not just how good of a fit it is and how to use it.

Summary:
I'd like some advice on where a newbie to basketball statistics should begin; books, articles online, journals, anything that would help. I don't need a background in the mathematical aspects of statistics (math/physics student already), so a reference to a stats textbook is not what I'm looking for, but more specifically applying statistics to basketball.

Thanks in advance, and I hope that I can contribute positively to this board in the future.

Hey, there. I just wrote this primer for someone such as yourself:

http://thecity2.com/2011/12/01/the-city ... al-primer/

Thank you.

Evan, that's nicely written.

Mike G wrote:Evan, that's nicely written.

Thanks.

And still, it would be nice if people explaining "advanced" stats would either :
a) be consistent with multiplying ratios by 100, so they can call it "percent"; or
b) abandon this practice, leaving those of us with 6th grade and better educations to understand that .647 is 64.7%, but without the annoying and often duplicate % signs.

Not singling you out, Evan. Everyone does this: after committing to per-100 (possessions, free throws, etc), and further committing to the % sign, giving a formula that does not multiply by 100.

If any reader is unsteady with basic math, this just makes it harder.

I had thought that (for example) FG% were shown in .468 form for decades, and now we see it as 46.8% -- sometimes.
A more complicated formula and more screen clutter is what I see. And confusion when you forget to multiply by 100 to get to that clutter.

You can still write, "He made 47% of those shots", without any injection into the formula.

I totally agree with you.

Instead of FG%, it would be better if it was simply field goal ratio (FGR). I'd rather work in ratios all the time, but I have no idea what level of mathematics the audience has that is reading that article, so I err on the side of assuming they are really stupid.

EvanZ wrote:I totally agree with you.

Instead of FG%, it would be better if it was simply field goal ratio (FGR). I'd rather work in ratios all the time, but I have no idea what level of mathematics the audience has that is reading that article, so I err on the side of assuming they are really stupid.

Great work as always.