Analyzing the voting data for the dueling city auditor ballot measures shows that the majority of voters did want an auditor, but vote-splitting resulted in no auditor being selected.
There is a block of about 11,000 voters who voted no on both measures. This group was opposed to hiring any auditor. This number is derived by assuming those who voted yes for elected also voted no for appointed, and visa versa.
So, subtract the yes-appointed votes from the no-elected votes, and visa versa. This shows the roughly 11,000 block of additional no votes.
There were about 38,000 votes cast for each measure. Subtracting the 11,000 no-auditor votes leaves 27,000 voters who did want an auditor. The data confirms this, showing a combined total of 27,014 yes votes between the two measures. But, the minority no-auditor block won because the yes votes were
split between the two measures (with nearly twice as many yes-elected than yes-appointed votes).
It is well known that offering a competing ballot measure often results in the most popular measure being defeated because of vote-splitting. I don’t think using vote-splitting (a side effect of our antiquated voting system) is a good exercise of democracy.
Only if all yes-appointed voters wanted their measure or no auditor at all, has democracy prevailed. But if at least one-third of them would prefer an elected auditor over no auditor, then democracy has been
And it has been cheated knowingly by those on the appointed campaign who are aware of the vote-splitting effect.