News reports about polling will often say that a candidates lead is outside the margin of error to indicate that a candidates lead is greater than what we would expect from sampling error, or that a race is a statistical tie if its too close to call. It is not enough for one candidate to be ahead by more than the margin of error that is reported for individual candidates (i.e., ahead by more than 3 points, in our example). To determine whether or not the race is too close to call, we need to calculate a new margin of error for the difference between the two candidates levels of support. The size of this margin is generally about twice that of the margin for an individual candidate. The larger margin of error is due to the fact that if the Republican share is too high by chance, it follows that the Democratic share is likely too low, and vice versa.
Polls report levels of support for
each candidate, with the MOE applied to each Individual candidates support (graph at link). You must
double the MOE to determine if the lead is beyond a statistical tie.
Thats why Im not too excited about Bidens 5 point lead in a poll with a 4.7% MOE, because he would need a lead greater than
9.4% to be outside the margin of error.
Its also why I
am excited about national polls showing Biden with a 7-10% lead with 2-3% MOE (538 gives 99% odds of winning EC with pop vote margin greater than 7%), and why I am excited about Rust belt polls giving Biden 8-10% leads with 2-3% MOEs.