Poll Analysis

RealClear Politics does something I hate when it takes an aggregate of polls, it mixes unalike polls, and it gives all polls equal weighting. That's not realistic.  In order to do a more reasoned polling average in the swing states, it should be a weighted average of like polls.  I've taken a partial step towards filtering that way this cycle for the swing states. I haven't yet weighted the polls but what I have done is filtered the polls to include only the following in the various swing states;  Only likely voter polls with a sample size >=500 a margin of error <=4.25%,  and only polls taken since 06-Jul-24.

Not all swing states have had polls that qualify with that as a minimum criteria. But, given those criteria, here's what I am seeing:

Given these likely voter views, and assuming no other inherent bias in these polls, I have Arizona, Georgia, Michigan, Nevada, Pennsylvania and Wisconsin all looking like Trump wins. It's still not a complete picture mind you because some of these are within the average (unweighted) margin of error.

But if this holds up and all else is held equal (and yes, more work needs to be done on my analysis, and it's still only July), the electoral college would look like this: