<<Points Created (PC) was designed in an attempt to measure what Randy Buehler . . . called the "shadow effect" . . . and to correct for that effect in player ranking. . . . On the face of it, it had problems. While some players had their PC scored rated positively, most ended up with scores that had minus signs in front of them.>> There were two reasons for this: (1) PC went further by attributing bonus point creation as well as tossups (i.e., a TU counted more than 10, and an INT much more than -5). (2) It then tried to compare the player either to an empty chair or to the fourth player on a team of four identical "average" players (I don't recall which). *This* is where most of the negative values came from, and, as Pat Matthews pointed out a while back, could easily have been corrected by dropping the "normalization" factor. <<(The problems with this of course are 1) actual category distribution effects can only be measured by precise _a priori_ measurements which no one has, to date, released, 2) and difficulty is so highly subjective that no one has yet bothered to propose a way to objectively measure it - but I digress.)>> I actually tried to come up with a quasi-objective system for Penn Bowl 10. I will discuss *that* in another post. [It worked--in an odd sort of way--in a manner that will help with editing for PB11.] As far as testing the validity of PATH, I'm not well-versed enough in statistics to know exactly *how* to test it, although I'm sure a fair test could be devised.... --STI
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:44 AM EST EST