Yes. Here's what's empirical. If you look at referenda around the world, particularly national referenda, and a year out or on the initial date that you announce you're going to put this on the ballot you look at what percentage is voting “yes”, in almost all cases what happens is that support for yes goes down.
When you first launch a referendum, it's always like mom and apple pie, or let's reform elections and make everything great. People say, okay, yes, let's do that. But then the conversation comes and maybe the “no” ads come out and say, oh, but this will ruin your life, and it's very scary. In almost all referendum campaigns around the world, support goes down. That's the empirical fact.
The anecdotal fact is that there are these relatively small professional communities that run campaigns. In California there's a very active one around referenda, and this is a rule of thumb. There are a bunch of people whose reputations and compensation depends on win-loss, so when you're in a room with those guys, when you're at conferences with those guys, if you're 55% or 60% “yes” a year in advance, that's a danger sign. They're pretty sure that one will not win.