Why is electoral-vote.com lopsided towards Bush?

[coloradodem2005]

I don't understand because Kerry and Bush are dead even again. None of these polls are Strategic Vision so what is the deal?

[21winner]

1. Don't they have email?

No one can answer this here.

[clonebot]

2. i dunno - i wouldnt worry

i've seen it extremely favorable to kerry some days - it just depends on who was polled that week. i used to volunteer with pollsters - most people quit halfway through the poll or don't even do it. dont worry too much. i'm working on my own spss project and stats gathering system (wrote it myself in c++) to give myself a clearer view of the poll findings and 98% of the time kerry comes out on top. and i obviously have no reason to be bias with my own statistics projects since it is for my personal fun and enjoyment. i know my findings are accurate since I've aced graduate level stats and biometric courses.

i'm a nerd.

[karnac]

4. it's so flawed even their *averaged* projection map is off

http://www.electoral-vote.com/fin/sep17p.html

NY tied and CA weak? I DON'T THINK SO!

Karnac

[last_texas_dem]

3. I think a lot is just coincidental

The polls released in recent days have been slightly more favorable towards * than they had been, but many have shown neck-and-neck races and a few (mostly Gallups) have seemed a bit unbelievable. (Then again, there were some slightly unbelievable ones for Kerry a while back, like where he was leading in TN and whatnot, so perhaps things are just that volatile.) FL was tied for quite a while and now has a slight * lead, PA has an insignificant 1-pt. * lead, and some fluke poll of NJ has a * lead. Those are all electoral vote-rich states and having them over in *'s column make him look like he's stomping Kerry when it's really still pretty damn close.

[clonebot]

5. i agree

weekly polls are hardly scientific - they're just cranking them out with an old land line based autodialer and an archaic formula that barely works anymore. whats interesting is that the overall success of a polling group is how accurately they predict the final outcome or result. if they are dead on election day then whatever crazy poll results they had months before don't really matter. which unfortunately means that a pollster can be totally biased and falsify results during the election but as long as they accurately predict the final outcome down to the percent their poll will be praised as "one of the good ones" and they'll get more contracts next year. ;)