The critics of the STV argue that it's not enough:
The numbers or rankings that are marked on each ballot are not separate votes but are instructions to be used in the complicated counting system.What they don't say is that those "instructions" as they call them effectively answer the most important question: If your first choice doesn't get elected or if he's already been elected - what's your second choice? Specifying multiple preferences on a ballot is much more informative than marking an X or even - multiple Xs.
Here's another common myth:
...the complicated formula discriminates between the percentage of votes needed to elect MLAs.What's never mentioned is: 12.5% (14.3%, 20% or 33.3%) of what? Let's not forget that a 7-seat Capital Region riding will have much bigger population than a 2-seat North East riding. What if we count the actual number of votes required to get elected under STV and compare it to the number of votes one needs to win under the current system?
In the capital region a candidate needs 12.5 per cent to be elected. In the Northeast region it requires 33.3 per cent, in Columbia-Kootenay 20 per cent and in Vancouver West 14.3 per cent.
The proposed Capital Region constituency would have a population of 345,164. That's 49,309 residents per MLA. Obviously, that includes children and permanent residents who aren't yet eligible to vote. And of course, not every one of those voters is actually going to get out and cast the ballot. Let's assume that the number of active voters equals to 1/2 of the population. So we're talking about 172,582 ballots cast. The quota in a 7-member riding is 1/8+1 vote = 21573 votes.
Now, what about existing single member ridings? The most populous of the 7 single-member constituencies that would make up the Capital Region riding is Saanich North and the Islands. Its population is 55,201 which gives us 27,600 active voters. So the maximum number of votes one would need to get elected is 50%+1=13,801 votes; that - if we only have 2 candidates on the ballot which almost never happens nowadays. As for the minimum number - it could be... just anything, depending on the actual number of candidates and on how well they perform. Usually, it takes about 40% of the votes to get elected under FPTP. In our example that would be ~11,000 votes or almost half the number of voters one would need under STV.
So, if anything - it should be the "yes" side arguing that the existing First Past The Post system discriminates between the percentage of votes needed to elect MLAs. After all, while the STV quota is based on the number of elected MLAs (which is pre-defined for every constituency,) the percentage quota for FPTP is determined by how people vote.
And finally - this:
Your second preference could get counted as 10 per cent of a vote while your neighbour's second preference could get counted as a full vote, or as some value in between, or not at all. With STV you cannot control what fraction of your vote is given to each of your preferences, because how your vote is counted is determined by how other people vote.What wasn't mentioned is: the only time your subsequent choice can be counted as anything less than a full vote is if your preceding choice has already been elected. In other words - you've already got one of your top choice candidates elected and now, since that guy actually had some surplus votes - part of your vote goes further, benefiting your next choice. If anything - that's an advantage.