Actually, last time we had people voting under approval voting in an IRV election. The Prime Minister declarations period ended without any campaigns, so nominations were automatically-reopened, and Prime Minister voting ended up after Delegate voting on the forums.
Because it just feels messy 
ā but also, I think the above example is a decent one of how even when we had Prime Minister voting after Delegate voting, it didnāt seem to alleviate the issue. I donāt know if itās worth the trade-off of having a more complex schedule of elections.[1]
Sure, but when we had people voting approval in an IRV election, that wouldnāt have been a solution. (If you really want to go down the route of real-life parallels, Example 22 from the UK elections pamphlet you linked seems illustrative.) In any case, Iām glad you intend to introduce legislation because I think you are bringing your familiarity with real-world UK election rules and even in this topic there are at least two people not named Kris with conflicting views on how ballots should be counted.
Iāll make two brief thoughts, though Iām not sure how much theyād actually help.
The simpler thought: perhaps we could put simple instructions on the ballots, like the example approval ballot and preferential ballot on Wikipedia.
The broader thought: perhaps we could choose a different voting system.
That can certainly be a chase-your-own-tail kind of discussion. I think itās worth prefacing this thought with something Glen said in 2019:
Can we just pick one without getting philosophical about it? We are a community of less than 40-50 people at max population. As someone who was forced to study this stuff to get my degree, it doesnāt really matter what method we use with such a small population. [source]
There are a lot of options for our voting system. None of them are (mathematically) perfect. I get that.
But I think intent matters. Even if our voting population is small, the voting system we choose can set the tone.
Before we had approval voting for the delegacy, we had IRV, with the specific intent to have two more distinct candidates going into the second round. For instance, Nwahs, your second link to the previous discussion seems like a center squeeze ā which, arguably, could be a āfeature not a bugā if the goal is to identify two candidates with more distinct political positions.
Similarly, we adopted approval voting with the specific intent to identify candidates with broad support and less opposition, with the view that the Delegate should be someone who is a consensus candidate and who is, in Tsuās words at the time, āwell-known and trustedā to have access to all the regional power that the Delegate has.
If weāre fine with somewhere in between, a Condorcet method or expanding approvals could also be options. It also depends on what we actually value from our system; for example, later-no-harm is a property of IRV but I donāt know if I really care for it in Delegate elections because I think consensus-building is valuable for a Delegate and endorsement requirements already provide some level of incumbency advantage.
A digression: I actually think this complexity does matter more than one might think. Even quite politically involved and experienced figures arenāt always mentally keeping track of when the next election cycle is. (Like that one time literally all of us forgot about a Chair election. For like an entire month or something.) And that makes it especially hard for a newcomer who has ambitions for higher office to hit the ground running, because they might not start writing their campaign until they see nominations have opened, and their opponent might already have their campaign drafted because theyāre more experienced with a more complicated electoral calendar. ā©ļø