Those are crushing numbers. 90% of 43% of voters who made a choice to vote and leave. Such overwhelming numbers. Do you think the other side would have gotten everyone who didn't vote?
And it's not an hypothesis: polls and parliament votes have shown for years that the support for independence (or for independentist parties) has been fairly stable at around 45-50%. Support for a referendum vote has been wider (up to 70-80% depending on the poll and how you count some party's positions) but similarly fairly stable over the years.
Surely if people thought the referendum "didn't matter", then they'd be less, not more likely to cast a ballot? I'd say that 40% is a pretty strong vote, in those circumstances (and given the risk of being beaten up by riot police).
It was practically guaranteed that the votes would be replaced later in the cardboard urns guarded exclusively by the separatists. The obvious plan was to drag as many millions of people as possible to vote and then use this number as new upper limit to made up a separatist support number (see? 80% of this new cipher voted separatist, we counted the votes ourselves, [invented a number], and this is two million more of supporters than before, so I'm right. Now lets negotiate how money you own me to not secede).
It must be noticed that they printed 10 millions of ballots, for a pool of voters much lower. Why they needed to have like two ballots for each possible vote?... well, fill the dots.
The only good move in that situation is avoid voting
In most places that wouldn't even be considered a valid referendum.