Basketball Statistics Blog

Sunday's Reflexions on Basketball Statistics:
Giving Explanations about Sort of Sample Gathered

"... When we discuss in the eBA Clinics Course about statistician's role, what the meaning of "giving explanations about what sort of sample was gathered, and how might that sample affect the result" ?

This topic is broadly discussed at these eBA Group of Sites: there is a general interpreted sensation that statistical knowledge is many times incorrectly used with intention, by finding ways to translate information from the observation that are favorably disposed to the presenter.

Photograph: Posterous

A famous quote, included in our front page, from former British Prime Minister Benjamin Disraeli is "There are three kinds of lies: lies, damned lies, and statistics."

In truth , the well-known book "How to Lie With Statistics" by Darrell Huff identify many cases of deceptive uses of statistics, concentrating on graphs designed to deceive either deliberately .

By choosing ( or rejecting, or modifying ) a certain sample, results can be manipulated; throwing out bad data values or local anomalies is one means of doing so ( eliminating one or more low performance games of a certain player from his season balance, is a known example ).

This may be the result of without restrictions fraud or of complex, balanced and unintentional bias on the part of the statistician. Do you know the case of that point guard that at home plays, with the locals guys at the computers, had 11.1 assists per game... meanwhile on the road he served only 3.2 ? ..."

tags:  basketball  ~ basketball statistics ~  basketball analysis  ~  sort of sample gathered
visit:  eBA Portal  ~  eBA Forums  ~  eBA Encyclopedia  ~  eBA Clinics