I started off thinking about this after looking at San Jose, CA, which requires that all city officials win with a majority of votes. At first glance, it seems simple enough to say that the STAR runoff winner would have a majority of votes, therefore they qualify under that requirement. But none of the non-Top 2 candidates received binary votes; they only have scores. If we ignore those scores, we would end up excluding candidates who received votes from the results. If we try to focus on the scores, then the candidate with the most votes (highest score) might not win. If we somehow try to use only the pairwise comparisons for the results, we have to figure out which runoffs should be used for the final results and why in a legally consistent and electorally reasonable manner, and this might contradict the STAR list of results. And if we take a hybrid approach, then a candidate outside the Top 2 might receive far more support (say, 70% of total possible points) than either of the Top 2 (whose votes we are counting via binary votes, thus likely to stay to <60%.)
I’m not sure how election codes, city charters, and state/federal Constitutions might interact with the idea that a voting method could be enacted where someone with the most votes might not win.
Additionally, for those interested in the NPVIC, one of the main proponents mentioned that alternate voting methods should all adhere to the principle that the “most votes wins”; however, they also said:
There are numerous alternative voting systems (IRV, approval) and it’s not clear how to combine them. Presumably, the first state enacting an alternative voting system would specify, in its own legislation, how to interpret votes from that state.
So would there be legal issues with STAR? Or can one escape by perhaps publishing a “public” set of results holding to STAR’s methodology, while taking an alternative approach with the “official” results that bypasses legal issues, such as declaring by fiat that the STAR winner should win, against any “most votes wins” legal principles that might exist?
If necessary, maybe Score’s points could be used similarly to “votes”, so that if someone gets a majority of possible points, that is counted as a majority of votes. It would maybe even not be too difficult to maybe use some kind of severability to allow the runoff part of STAR to be struck down in court while keeping the scoring part.
Also, as an aside, I am curious whether anyone would seriously advocate bloc STAR as opposed to bloc Score, as it might get very confusing to explain how bloc STAR’s results were tabulated, where maybe a 6th or 7th place scored candidate might make it in the top 3 winners.