This is exactly what was done in the VE5ZX/CT1BOH analysis. The trouble
with it is that even with the really striking outliers, it's not direct
evidence of any particular behavior that is contrary to the rules.
That's why you need the possibility of inspection.
73, Pete N4ZR
New Articles Daily - the Contesting Compendium at http://wiki.contesting.com
The World Contest Station Database, updated daily at www.conteststations.com
The Reverse Beacon Network at http://reversebeacon.net
On 8/18/2009 6:27 PM, Charlie Gallo wrote:
> On 8/18/2009 David Kopacz wrote:
>
>
>> I have asked the CC to divulge how they are identifying cheaters, not
>> specifically to me, but publicly. They refuse to divulge that
>> information in the interest of protecting their methods from being
>> circumvented by cheaters. This I can understand to some degree.
>>
>
> They are probably running some sort of fairly basic statistical analysis, and
> knowing exactly what they do will tend to change things
>
> My wife is a statistician, and I've seen her pull some real "interesting"
> trends out of data. The huge problem is when everyone knows the model and
> uses the same one (which is pretty much what lead to the current financial
> mess - everyone used the same model, developed by Chase)
>
> Except that she is busy as heck, I've been tempted to download the logs from
> CQWW, and set her at it saying "what are the outliers - which stations are
> 'odd'". It won't tell you that they are cheating per-se, but it will tell
> you that "something is different at that station, be it gear, location, or
> operating practices"
>
> I suspect what the CC has done is already IDed a bunch of "outliers",
> stations that make them say "Hey, that's Odd.." (which is the REAL sound of
> Science being done, not Eureka), and they will show up to see WHAT is odd.
>
> Now, what I would HOPE is that if they found something "odd" that was within
> the rules, THEY would not tell the world exactly what the other station is
> doing, that is a competitive secret for the station to give out or not
>
>
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
|