| Ryuu96 said: I'm more and more convinced that pollsters don't know what's happening either, that's why they're all practically merging on the same thing, if Harris wins then they're right because of MoE, if Trump wins then they're right because of MoE. It's only a landslide victory which will make pollsters wrong this time around, which makes me a little suspicious. |
Early voting is a new thing for the USA, especially on the scale that we are seeing this election. Pollsters rely on historic precedence to adjust their models for higher accuracy, but when there's no precedence, they are basically lost, regardless of if they are biased or not.
It's kinda like Shadow1980, NPD numbers and Nintendo Switch. He has no bias one way or another when it comes to sales data, but none of the historic data is properly applicable to a hybrid console, so he has gotten it wrong with his predictions repeatedly.
So pollsters hedging their bets when it comes to polls this election cycle wouldn't be surprising at all. Operating within the margin of error is the safest thing they can do in a time of great uncertainty. Everyone doing the same thing also covers each other's backs. At the end of the day they all can say that nobody really knew what was going to happen.
Legend11 correctly predicted that GTA IV will outsell Super Smash Bros. Brawl. I was wrong.
