We’re getting plenty of hints that CJ is playing as close attention to this blog as any other website, so I should caution everyone who contributes here to choose their words wisely. I bring this up because it appears that I’ve given CJ an idea, and in blog warfare one should keep their cards close to their chest, and always be aware that what you reveal can be turned against you and/or used to the opponent’s advantage.  So here I have to put up a thread to get the ball back.

My mistake this time was a comment I made yesterday, where I elaborated on the numbers presented in Stats Shocka: LGF Pages vs. DoD:

That’s pretty simple logic, and it seemed harmless enough.  But compare that to what CJ posted today (the “stats” thread must have left a mark):

Yes, he just realized it.

Anyway,   I’ll hand it to CJ here for agreeing with me and attempting to come up with a good way to more accurately quantify the real views of individual articles, but I’m afraid to say this is a FAIL.

Before I explain, CJ should clarify re: the Pages, because seeing a hyperlinked thread title in a sidebar shouldn’t constitute a “view” in anyone’s book.   In fact, it fails the criteria that CJ laid out in the preceding sentence (containing full text), and by that rationale would mean that any view of any page with an RSS feed on it should count as a “view” for those linked threads.   That’d be more blatant cheating, but that’s how it reads in CJ’s comment.

On the other hand, if CJ meant to say that they’ll count when someone views the LGF Pages’ main page,  then that’d be consistent.  But he didn’t specify that, and we stopped giving CJ the benefit of the doubt a long time ago.

Either way, its a sham unless its used as a separate and internal statistic.   The technique (and the resulting statistic) could no longer be used as comparison to other sites unless it is known that the same technique is used.   But even then, the thread count on the front page must be equal for a truly accurate comparison (if I’m correct, each additional article should have an exponential effect on the increase; I’ll call this the “scroll-assume” variable.)  Complicating things further, a front page post will register once initially, then again when clicked on, counting twice for essentially one “read”.   I’d expect the numbers to go up quite a bit.

Bottom line:  I just gave CJ an idea on how to inflate his visible stats, and how to make it sound good to the horde.  

Sorry ’bout that.