Remember Ayyadurai said in the "normal, expected" case, you'd see a flat line.

I'm just showing that, to get a flat line, you'd have to have the % of straight-ticket Republican voters be perfectly correlated with the % of split-ticket Trump voters. This is very unlikely, and is the ONLY way to get it flat.

Which means, plainly, that he's lying/being intentionally deceptive.

In every other case except that, you'd see a line with a negative slope, given any data. So when Ayyadurai says that a negative line is evidence of fraud, he's again being deceptive. It's not evidence of fraud, it's what you expect under most cases, given the values he's decided to plot. Does that make sense?

If not I can dig in deeper.