This isn’t a pedantic debate. If the math works and the results demonstrate the math works, what more is there?

> However, the sample size is only large enough to draw conclusions on maybe a county level.

Again, this is simply incorrect, as demonstrated by the math and backed up by actual results from polls conducted in this manner. You’re wrong and burying your head in the sand. Their polls of 1200 people arrive to the same conclusions within a fraction of a percent as the actual results. Is that really so hard to believe?