Whether you believe the top numbers or not, the trend is unmistakable. Every state is moving away from Clinton.
And I see a lot of young people on this forum who have grown overconfident in 2-week/1-month plus prior to election day polling. Because of the last month stability of the polling in the last four elections, we've convinced ourselves that we can rely on mid-October polling to tell us what we need to.
But there may be few here who remember a race like 1980. Which had a hugely dramatic and extremely late break toward Reagan. Even discarding Gallup's famous Carter +8 penultimate poll, the other (then regarded as amateur) pollsters were all showing an extremely tight, within the margin race. The actual outcome (Reagan winning by 10 points) was well out of the margin of error for nearly every public poll.
Perhaps significant to this outcome was that 1980 had persistently high undecided / third party numbers for nearly the entire GE campaign. The late break of undecideds was sharp and shocking. It possible that if we had the prolific public polling then as we do now, it would have appeared much like we are seeing today; a series of wild, volatile perceived outliers pointing to utterly irreconcilable disagreement over the state of the race in the last week.
If this forum existed in 1980, the Gallup one week move from Carter+8 to Regan+3 in their last 2 polls would have probably prompted quite a few dismissive "junk poll" retorts.