The process is poorly described. The Brennan Center [1] does a better job. The key point in the efficiency gap is that you don't calculate district by district.

The process is poorly described. The Brennan Center [1] does a better job. The key point in the efficiency gap is that you don't calculate district by district. You calculate the efficiency gap for all of the districts in a state. Yes, if a single party won all districts by a 2% margin, it would be gerrymandering. It's not statistically believable for a party to win that many districts "by chance", especially if the vote margins are that narrow in each district. If Party A wins District 1 by 2%, I'd expect Party B to win District 2, etc. If Party A wins Districts 1-100 by 2%, then yes, I'd strongly suspect gerrymandering. [1]: https://www.brennancenter.org/sites/default/files/legal-work/How_the_Efficiency_Gap_Standard_Works.pdf https://www.brennancenter.org/sites/default/files/legal-work...