Past history also indicates that Gallup is correct. Any fool and his goat who knows how to read will tell you that Gallup is not.
Further inspection of Gallup's weekend (10-29 - 10/31) polling indicates that Bush has only a 4 point lead among men (50-46) while Kerry maintains a 6 point lead among women (50-44). That to me is quite odd, as Bush in almost every poll taken has almost a ten point gap men voters preference.
Furthermore, Gallup indicates that Bush holds not a single lead in any age-range demographic. At best he is tied in the 50-64 age range (48-48). This to me also seems quite odd.
Kerry leads by 17 points in the Eastern geographical region, yet his state poll indicates that Bush is up 4 in Pennsylvania. Furthermore, in Bush's strongest geographical region, he only has a net +6 gain (51-45). This to me also seems odd. To compare to Gallup's last poll, Kerry was +11 in the East, Bush was +15 in the South.
Along Gallup's Partisanship line, Bush carries R's 93-6, Kerry carries D's 91-7, and I's break 48-40 Kerry. In Gallup's last survey, they had basically the same results, except I's broke 49-44 for Kerry. I'm left wondering, why is there only an 88% sum for Independents this time around, when R's = 99% and D's 98%.
Each of those numbers, of course, has a smaller sample size, and thus a larger MOE than the internals I posted, which were coming from the entire sample, not just a subset. So that at least partially explains any weirdness.
And I agree that Gallup's state polls are wacky.
