Scoring Inconsistancy Statistics

deeney0 · Post #1 » by **deeney0** » Mon Jan 7, 2008 11:21 pm

There has been a lot of talk about individual scoring inconsistency on this board, but does anyone know of a good way to quantify it? The best way that came to my mind was standard deviation in points per 48 in each game:

Player Standard Deviation in PP48
Buckner 5.82
Doleac 6.27
Brewer 6.88
Jefferson 7.19
Telfair 7.27
Gomes 7.52
Madsen 7.62
Richard 9.41
Walker 10.03
McCants 10.81
Smith 11.07
Green 13.31

No surprise that Walker, McCants, Smith, and Green are the most inconsistent scorers using this statistic; what did surprise me was Brewer's and Telfair's relative consistency. Anyway, this might not be a very accurate statistic, if anyone has a better idea, don't keep it to yourself.

theGreatRC · Post #2 » by **theGreatRC** » Mon Jan 7, 2008 11:27 pm

It's no surprise with Brewer's consistency, he is scoring less than 10 per almost every game.

JBoog35 · Post #3 » by **JBoog35** » Tue Jan 8, 2008 12:21 am

theGreatRC wrote:It's no surprise with Brewer's consistency, he is scoring less than 10 per almost every game.

haha

I think what's going to be more telling is scoring inconsistency compared to minutes inconsistency. I bet they're parallel.

Post #4 » by **Calinks** » Tue Jan 8, 2008 12:41 am

Yea I'm not surprised with Brewer or Telfair either. Clearly their scoring has been consistently bad.

deeney0 · Post #5 » by **deeney0** » Tue Jan 8, 2008 1:43 am

Here's the standard deviation in minutes per game:

Brewer 11.75
Buckner 12.36
Doleac 6.98
Gomes 8.77
Green 8.89
Jaric 13.37
Jefferson 3.37
Madsen 5.82
McCants 11.41
Richard 6.94
Smith 7.95
Telfair 9.85
Walker 10.93

And a scatter plot comparing it with the first statistic:

I wouldn't say there's a correlation.

the_bruce · Post #6 » by **the_bruce** » Tue Jan 8, 2008 5:35 am

If you include with this deviation with the high/low and/or a median or average value I think it makes more sense. As one can just glance at the other numbers to infer oh brewer always cant score etc.

Post #7 » by **C.lupus** » Tue Jan 8, 2008 1:52 pm

Sweet! We're getting into statistics now.

I think the standard deviation is the easiest measure and is sufficient. It needs to be reported with the mean, though, for it to make any sense.

Post #8 » by **revprodeji** » Tue Jan 8, 2008 6:41 pm

So, we have stats that show we have bad offensive players?

sweeeeeeeet

mandurugo · Post #9 » by **mandurugo** » Tue Jan 8, 2008 11:53 pm

you could try normalizing your results (dividing each players deviation by his max per 48), that way players couldn't hide their deviations by deliberately scoring fewer points per game. Brewer didn't think we knew what he was doing... but numbers never lie. I think Mark Twain said that.

deeney0 · Post #10 » by **deeney0** » Wed Jan 9, 2008 12:40 am

I don't think the mean has anything to do with the deviation - We're talking about inconsistancy here, if Corey can only be counted on for 6 points a game, that's consistant, whether or not its good.

shrink · Post #11 » by **shrink** » Wed Jan 9, 2008 3:42 am

Can you insert Jaric's number in your first post?

deeney0 · Post #12 » by **deeney0** » Wed Jan 9, 2008 5:11 am

Interesting, I don't know how he's missing. I'd love to, but my hard drive crashed today... I'd just backed up a few days ago, so pretty much the only thing I'm going to lose is the hour or so I put into that file. Oh well. Based on the graph, he's just a little under 9.