mesmerism! presidential assassinations! circuses on fire!
BlogPoll Navel Gazing 1.0
So, based on everyone's input on the BlogPoll's direction I think there were some overriding themes:
- Weighting is no good.
- Tangible repercussions for weird/biased ballots should definitely be delayed for a good while while we work out whether they're necessary or not.
- Decisions on what's weird/biased should be decided democratically.
That's where it stands at the moment. If I indulge in any weighting stuff it'll be strictly experimental. The straight-no-chaser poll will remain the primary one for the foreseeable future. Scorn will be our weapon of choice against the unreasonable.
Several pollsters expressed dismay at the overt similarities between the coaches/AP and blog polls. Joe summed it up best: "we aren't as cool as we think we are." True. (I'm A Realist has a useful side-by-side-by side comparison.) I was a little disappointed with the lack of influence some well-reasoned arguments had on the poll at large. NC State and BGSU were teams that had cases made for them but got only small bumps relative to the other polls and were not able to crack the top 25. I don't know what percentage of pollsters bothered to read the "rank 'em" roundtable, but it doesn't seem like it was many... or at least there weren't many convinced by the arguments presented.
But, as this was the first-ever BlogPoll, I'm not too surprised. It will take time to develop a culture around this thing, and it'll take some time before people become familiar enough with each other to change their minds based on someone else's opinion. Changing minds is not easy, no matter the fragility of the ideas they hold.
This year is AA ball. We ain't ready for the major leagues yet. Rob got this comment on his blogpoll post:
What was your criteria for voting? Any of these people for that matter. I'm willing to wager that the majority of people voting can't even name one player on every team they ranked. Awesome.
P.S. You do realize that ESPN has had a fan poll for about 10 years now, right?
The standard deviation with 50000+ voters is much much less than with about 100.
and I've got no answer for that. Developing one is job one. Job two is beer.
Hey, at least we've got Ray Ratto on our side.