Single-winner Asset (Election Theory)


Sorry, I started the “Single-winner Asset” topic in the wrong category.

If there are two or more candidates I would approve and two or more I wouldn’t, Approval lets me express that straightforwardly without my having to gamble that one candidate I like would transfer assets to another candidate I like.

Also, Approval is a very simple system. It is FPtP less one constraint.

I hold that Approval gives the voters the same power as any finer-grained Score Voting (Range Voting) system based on addition (not averaging), if the electorate is “large”.


Why would it be a gamble whether your two preferred candidates support each other? I don’t think politicians are that bad!

Approval sounds simpler, but it forces voters to evaluate and approve multiple options to get their preferred outcome. Asset is much easier from a voting perspective: pick your favorite.

The problem with giving voters power to evaluate multiple options is that they inevitably have to focus on evaluating the centrists rather than their favorites, which isn’t motivating for them. That’s not how democracy should work. Every voter should get an equal say without knowing what other voters will pick.


I have a concern.

I might run as a “conservative”, fill my website with cut-taxes-for-everyone allow-guns free-market propaganda, and transfer my votes to a leading Democrat.

Should we perhaps instead let everyone cast as many votes as there are seats? (In a 5-district, you can probably find 5 people that are on your side that are not “centrists”.)


Perhaps, though I don’t think this resolves candidate trust issues.


@waugh Rather then making two separate threads, you could have just edited your previous single-winner asset post. You can change the category of a post when editing it.


There is one seat. Governor of the State, or POTUS, or the one US Senate seat that is coming up on staggered schedule for the given State.


Eventually I figured out how to do that.


“In Approval, you’d have to think about which candidates are centrist, disregarding the ones you truly like, which doesn’t seem quite as interesting for the voters.”

There would be no need in Approval to disregard the candidates you truly like. You should approve them unconditionally.


That’s true, but it is generally important to consider whether there is a lesser of two evils candidate that you prefer as well under Approval. That requirement doesn’t apply to Asset.


Agreed; with Asset, you would be delegating to someone else the job of picking the compromise candidate.


Yup. I just wish I could prove that there’s nothing wrong with delegation and politicians trading votes versus voters doing it themselves. We make the argument all the time that we need representatives to help us govern, so why can’t we go ahead and say we need representatives to help us vote? If we can’t trust them with our vote, it seems like we really shouldn’t trust them with anything.


But the version you describe looks as though it could plausibly degenerate to FPtP.

What about my compromise version? I take back the amendment about making electors publish their backup choices before the fact. Just vote in Approval style, then tally by rounds, and each Representative Denied Office (RDO) could give proxy votes to candidates who haven’t been eliminated yet.


Your compromise version is quite a bit more complex, and arguably, you’d be more likely to pass simple Approval than this. It’s a good idea to relieve people, balances both worlds, but I don’t think it changes performance.
The truth is, if candidates are unwilling to compromise when they trade votes, then they are totally unfit to govern. In that case, no voting system can fix the voters’ choices.


You think my compromise wouldn’t perform better than plain Approval?


Sorry, it is better, I meant that it’s not better than Asset (maybe like 1% better) :slight_smile:


101% as good as plain Approval?


No no, it’s a lot better than Approval, but perhaps 1% better than Asset. I’m not sure I want to put a number between Approval and Asset, since I’m still learning so much about Asset’s pros and cons. Right now, I’d say 60% better? That’s for the negotiation between skilled candidates, rather than uninformed voters.


A way to speed up my plan for if there are hundreds of candidates is to make it run in about log time instead of linear time. Instead of denying one candidate at a time, deny the bottom half of them. Then collect proxy decisions from all those, then on to the next round. But when only two candidates remain, elect the one with more accumulated approvals.


Basing assets on approval totals causes issues with cloning. A party running 5 candidates would get 5 times more assets than had they ran one candidate.


He’s thinking of keeping it approval-proxy instead of approval-asset, which fixes the problem, since a proxy can only approve a candidate that their voter did not.