I live in a major swing state which decisively went for Trump. Here's the same kind of precinct plot for a moderately Democrat county (metro area).

Take-aways:

1. The county shows a U-shaped distribution, as opposed to initial plateau then linear slope in MI. This rings true to me, and I don't think there's any weighting going on here.
2. Overall, Trump is running behind the straight-R ticket by 4%
3. Strong polarization among precincts, with clusters < R20% and > R50%
4. There are really 2 distinct trend lines: < 20% and > 50%
5. Above 50%, the stronger a precinct goes R, the more surplus (crossover) votes there are
6. But even more interesting, below 20%, the stronger a precinct goes D, the better Trump does. In R10% counties, Trump only lagged behind the straight-R vote by < 2%.

I would think it's a pretty big indication of artificiality if I can perfectly recreate that trend line with excel: https://imgur.com/ByFaAkC

I tried to replicate a chart in excel in order to reverse engineer the formula, I think I got pretty close.

The formula is:

if (STRAIGHT_PCT < STRAIGHT_THRESHOLD) then SUBTRACT_PCT = 0

else SUBTRACT_PCT = (STRAIGHT_PCT - STRAIGHT_THRESHOLD) * WEIGHT

SUBTRACT_VOTES = TRUMP_VOTES * SUBTRACT_PCT

TRUMP_VOTES = TRUMP_VOTES - SUBTRACT_VOTES

BIDEN_VOTES = BIDEN_VOTES + SUBTRACT_VOTES

excel image vs shiva's chart

Macomb County's numbers:

STRAIGHT_THRESHOLD = 22% // don't adjust until above this number

WEIGHT = 0.5 // for every 2% above the threshold, subtract 1% from TRUMP_VOTES

You're correct that as straight Republican ballot percentages approach 100% then the surplus Trump vote percentage must approach 0%, for the sum of both cannot be greater than 100%.

But that doesn't explain the near uniformity of the slope of the problematic counties, nor the fact that the algorithm looks to be "if republican_straight_ticket > 20%, then force slope to -1/2"

I don't think this is any "custom code" or "hack" or anything. They mention in the video that every voting system has a "weighted race" functionality built in.

What I think happened then is that a modified configuration was used in targeted precincts to enable the weighted race feature and configured it to kick in for Republicans at 20% straight party detection and continue downward at a rate of (eyeballing it) -1/2.

This is it. This is the game, set, match. I joined TDDW just to say this.

DOJ needs to impound the Dominion machines and arrest their developers ASAP!