Draw the district lines cleverly and a party with a MINORITY of the votes can win a MAJORITY of the seats. The trick is two moves: PACK the other side’s voters into a few districts they win huge, and CRACK the rest thin across districts they narrowly lose. Their votes are ‘wasted’ either way. Slide the manipulation and watch 40% of votes buy 60% of seats.
Redistricting can divorce seat share from vote share through PACKING (concentrate opponents so they win a few districts by huge, wasteful margins) and CRACKING (spread the rest so they narrowly lose many). Both bury opponent votes as ‘wasted.’ The EFFICIENCY GAP measures it: (wasted votes of party A − wasted votes of party B) / total votes, where a vote is wasted if it’s for a loser or is surplus beyond 50%+1. A neutral map is near 0; gaps above ~7% flag a gerrymander (used in Gill v. Whitford). Here a fair 40% statewide is engineered into 3 of 5 seats. A fail-loud self-check throws unless a sub-50% vote share is turned into a majority of seats. ◆ real electoral math, node-verified.
A five-district toy with fixed vote splits (the exact packing/cracking and efficiency-gap logic); real gerrymanders optimise over geography and the efficiency gap has known edge-cases — the fewer-votes-more-seats result is exact.