when i make a forecast, efficiency at cheating is a part of the process - i explicitly, purposefully, soberly weigh how i think the cheating will happen, and what will work itself out in the end.
and, a big part of the reason i gave the south to trump and the north to biden was a perception on how the cheating will work itself out.
so, were the polls wrong? no - the polls weren't wrong. but, not every vote gets counted. so, when the polling suggests that biden is ahead by a certain amount, you have to balance that out with the amount of cheating that's going to happen on the ground - is that enough to act as a buffer for all the democrats that will try to vote by mail and just have their ballots thrown out? well, that's the question i'm asking - that's what my projection is about.
i decided that biden was far ahead enough in the north that the cheating couldn't undo it, but he wasn't in the south, that the republican advantage on the ground (meaning: republican corruption) would be decisive, in the end.
and, right now, it looks like i nailed it - but the margins are very small.