@dentangle@chaos.social @millihertz@oldbytes.space @onepict@chaos.social Thought I'd toss a theoretical perspective in for good measure. A lot of (good faith) voting systems are designed with the notion that each voter's vote, whether it's a ranking or a choice, will reflect their preferences for the candidates, independent of most other context. The problem of the system is then to aggregate these votes into a final choice that has fairness properties and other good features. Arrow's Impossibility Theorem and similar results show that this is tricky to do even under idealized assumptions. "Choice theory" is the keyword if you wanted to look up the fancy theory (decision theory, public choice, social choice theory are also good keywords).

As soon as people start trying to put their finger to the wind and guess how others will vote, and then warp their preferences according to that guess ("tactical voting"), the voting system's theoretical guarantees about fairness etc would be expected to break down. In some cases you can prove this mathematically. The structure of the voting system will not save you from this conclusion; ranked-choice voting, first past the post voting, and a bunch of other aggregation methods suffer the same general problem. You end up with a dynamic not unlike asset bubbles. Your standard economic analysis suggests asset bubbles shouldn't happen, roughly because people buy and sell according to their preferences, not according to what they think other people's preferences are. Of course experience shows asset bubbles happen constantly, as investors and speculators try to figure out whether they should throw money at the same things they see other people throwing money at (cough AI cough). Tactical voting produces the equivalent of asset bubbles in politics. It's really better for all involved if everybody votes and everybody votes their conscience.

I suspect many people don't because the level of trust you'd need has eroded too far: if I think the other guy's trying to game the system then maybe I ought to try gaming the system too so I don't end up screwed. Which closes the loop back to "X can't win here", a transparent attempt to game the system that works on several levels (the immediate level of warping preferences and therefore vote patterns; the secondary level of eroding trust and making the same play work again in the future).