1. There Shall Be Election Day Registration Available, and Election Day Shall Be Made a National Holiday
This would make a huge difference in voter turnout.
2. There Shall Be a Constitutional Amendment Confirming the Right to Vote
It's astounding to me that we don't already have this.
3. There Shall Be an Auditable, Voter-Verifiable Paper Record for All Electronic Voting Systems
Given all the fraud and errors that have occurred in the past decade or so, this is increasingly essential.
4. There Shall Be Independent, Non-partisan, and Transparent Election Administration
Determining who can vote, how they can vote, and where they can vote should not be in the hands of partisan officials who are often on the ballot themselves during the same election.
5. There Shall Be Instant Runoff Voting to Expand Voter Choice
This is a really great idea that I hadn't heard of till just now.
6. There Shall be Public Financing of Election Campaigns and Equal Media Access for All Eligible Candidates
This, more than anything, would help restore our democracy. We have to get the influence of Big Money out of our electoral system.
7. There Shall be Access to the Ballot and to Public Forums for All Eligible Candidates Regardless of Party Affiliation
When 3rd parties are able to run the people win.
8. There Shall be Uniform National Standards to Ensure Equal Application of Voting Equipment and Processes
There are currently a patchwork of separate and unequal local and state laws that cause millions of voters to be disenfranchised each election.
9. There Shall be Non-Partisan Redistricting Policies that are More Competitive and Representative of the Electorate
Incumbents shouldn't have the power to redraw the lines in ways that virtually ensure their reelection. After public financing, this more than anything would revolutionize our electoral system.
10. There Shall be Direct Election of the President; Abolish the Electoral College
The person with the most votes should become the next president. Period.
Labels: politics
At 10/29/2007 05:21:00 PM, 
Mike Clawson
Thanks for the details Clay, though I get the impression that your comment is a canned response to anyone who's blog pops up on a Google search for IRV (for instance, the fact that your comment seems like a polemic against "FairVote" even though my post didn't mention them and I've never actually heard of them).
Anyhow, you lost me with some of the math and statistics, but you're right that any proposed voting scheme bears closer scrutiny before adopting.
5. There Shall Be Instant Runoff Voting to Expand Voter Choice
Looks like the Claybot (I think?) beat me to it, but there are IRV problems. Unfortunately, there are problems in all voting systems with more than two candidates. I'll sweep the math under the carpet, but a result know as Arrow's Impossibility Theorem states that no voting system can satisfy all four of the following properties:
1. If a candidate receives a majority of the first-place votes, (s)he will be declared winner.
2. If a candidate is favored over every other candidate in pairwise races, (s)he will be declared the winner.
3. If Candidate X wins a first election in a scenario in which a runoff election (or other secondary election) is held and if Candidate X gets more votes in the second election than in the first, then Candidate X should be declared the winner.
4. If Candidate X wins a first election and a second election is held (as in #3) in which everyone votes the same, but one of the losing candidates drops out, candidate X should still win the election.
In particular, plurarity voting typically violates #2 and IRV typically violates #3 and #4. This is why IRV will never gain popular acceptance; if a candidate comes in first in the general election and then gets MORE votes in a runoff election and still loses, people are going to be understandably upset.
The biggest problem is that each electoral proposal with more than two options can be manipulated by those who understand the voting system better, handing elections over to those who best understand how to manipulate the totals rather than the "true" winner. However, it's less importance in practice since historical voting data suggests that almost all of our election results would have been the same under all of the major voting theories.
clay:
IRV is also more susceptible to fraud because it is not countable in precincts. That is, candidate A could win every individual precinct, but bizarrely lose when the ballots are all summed together
I don't see how this would make fraud easier. The idea Clay's getting at is Simpson's Paradox, which occurs when the various subgroups have vastly different sizes. If a candidate wins 48% to 47% in a huge district and by some other percentages in smaller districts, it's really not surprising that the close race in the largest district would overwhelm the results in the other districts; it just seems that way intuitively since humans are terrible at working with percentages. Furthermore, it only appears meaningful because of the arbitrary way that districts were assigned; I'm sure most people would prefer the candidate who won the most votes be elected instead of the one who won the most districts.
At 10/29/2007 08:44:00 PM, Mike Clawson
Thanks Miko. This is all very interesting stuff about IRV and it's alternatives. However, I don't want the rest of the important points of this "Bill of Rights" to get overshadowed by a debate about IRV. Even folks aren't entirely agreed about #5, I hope we can still stand together and support the other 9 Rights.
Jackson Boyd,
Well, no, IRV is not a good voting method. Objective Bayesian regret calculations show it is one of the worst. I have, on several occasions, offered you copious evidence to support that, which has been analyzed by math and political science experts who study voting methods.
Your response, which is typical of the IRV advocates, is to ignore that evidence, and keep repeating "IRV is good, IRV is good".
Among election method experts, there is a near-unanimous consensus that Range Voting and Approval Voting are better than IRV. And Bayesian regret calculations by Princeton math Ph.D. Warren D. Smith back that up empirically.
Clay
Miko,
Yes, no voting method is perfect, since no one has yet invented a hedonimeter. But the net effect of the qualities and flaws of all election method criteria can be expressed in terms of Bayesian regret. Range Voting is impressively dominant over other methods, whereas IRV is one of the worst.
Your mention of Arrow's theorem is irrelevant, since it applies to ordinal voting methods, whereas Range Voting is cardinal. And the Gibbard-Satterthwaite theorem is arguably much more important, at least among election method experts.
2. If a candidate is favored over every other candidate in pairwise races, (s)he will be declared the winner.
There's no reason this should be so. The Condorcet candidate is not necessarily the social utility maximizer, as you can see here in chart 5.
This is why IRV will never gain popular acceptance
Australia and Ireland have been using IRV since the 30's, and several prominent U.S. cities (e.g. NYC and Cincinnati) once used it. We use it here in San Francisco (or rather, a variant called "ranked choice voting"). It was recently approved in Oakland, Minneapolis, and Pierce County, Washington. While it's a terrible voting method, I wouldn't say it has no potential to gain wider acceptance.
if a candidate comes in first in the general election and then gets MORE votes in a runoff election and still loses, people are going to be understandably upset.
No, not "understandably" upset. Upset because they are uneducated about utility theory.
The biggest problem is that each electoral proposal with more than two options can be manipulated by those who understand the voting system better, handing elections over to those who best understand how to manipulate the totals rather than the "true" winner.
Which is why Bayesian regret calculations have to take strategic voting into account.
However, it's less importance in practice since historical voting data suggests that almost all of our election results would have been the same under all of the major voting theories.
Nonsense. These are just a few examples, and most elections conducted with plurality voting leave us with far too little information to know how e.g. Range Voting would have resulted.
I don't see how this would make fraud easier.
Let a Princeton math Ph.D. at the forefront of the election reform/integrity movement put it in more simple language then:
IRV cannot be "counted in precincts" – only centralized counting is possible – necessitating a big change to procedures and a security hit because a small central conspiracy then can throw elections.
With summable methods like plurality voting, or Range Voting, each precinct can send its totals up the hierarchy, meaning that fraud generally requires work at the precinct level. But with IRV, there is effectively no such thing as a precinct sub-total. Ballots have to be centrally tabulated, increasing susceptibility to centralized conspiracy. Don't take my word for it - look up the work of computer science Ph.D. and election integrity expert Rebecca Mercuri.
The idea Clay's getting at is Simpson's Paradox, which occurs when the various subgroups have vastly different sizes.
My point has nothing to do with Simpson's paradox, so far as I'm aware. IRV non-additivity is illustrated in this example.
If a candidate wins 48% to 47% in a huge district and by some other percentages in smaller districts, it's really not surprising that the close race in the largest district would overwhelm the results in the other districts;
Well, yes it is "surprising". If a candidate wins every district, it follows from trivial axiomatic logic that he should win the election. But with IRV, we can have a bizarre paradox where this logically imperative expectation breaks down. This is one of many severe pathologies with IRV.
it just seems that way intuitively since humans are terrible at working with percentages.
No, it's not intuitively wrong. It's logically/axiomatically wrong. IRV can tell us that candidate X is better than all other candidates, in every precinct - yet tell us that X is not the best for the whole electorate.
You are misunderstanding, and thinking I am talking about a scenario where the candidate wins in a majority of districts. No. Let me repeat, I'm talking about a situation in which a candidate wins in every district - but still does not win the election.
Furthermore, it only appears meaningful because of the arbitrary way that districts were assigned; I'm sure most people would prefer the candidate who won the most votes be elected instead of the one who won the most districts.
The term "most votes" is a vestige of our horrendously archaic plurality (vote for one) voting method. In many other (often better) voting methods, "most votes" isn't even a meaningful concept.
Clay Shentrup
San Francisco, CA
clay@electopia.org
415.240.1973
At 10/30/2007 12:37:00 AM, Mike Clawson
This is all very interesting, and obviously Clay and Jack have had this conversation elsewhere already. I didn't realize this was such a controversial issue.
Of course, I'd hate to see us get so bogged down in debate over percentages and funky theorems and not actually get unified enough to change the things that we can all agree need changing. Let's not get lost in the details.
jazzy,
If you think voter misidentification is a significant factor in election fraud (which is a comparatively minor problem relative to e.g. bad voting methods), then I think you would do well to look at some historical examples of fraud. Maybe it's not as rampant in Canada, but if it can happen in these examples, I think there's a lesson to be learned for any democracy.
http://rangevoting.org/PresFraud.html
http://rangevoting.org/LessFraud.html
Instant Runoff Voting is one of the worst voting methods. It can lead to the election of candidate X, even when candidate Y is preferred to X by a huge majority. Consider this hypothetical IRV election.
#voters - their vote
10 G > C > P > M
3 C > G > P > M
5 C > P > M > G
6 M > P > C > G
4 P > M > C > G
C is the clear Condorcet (condor-SAY) winner, meaning he is preferred by a landslide majority over all his individual rivals. He is preferred over G, P, and M all by an 18-10 margin.
But... M wins, even though he also has fewer first-place votes (6 voters) than C with 8.
Also:
1. P is preferred to M by 22 of the 28 voters, yet he's the first candidate eliminated.
2. G also has more first-place votes (10) than M's 6.
3. So M either loses pairwise to, or has fewer first-place votes than (or both) every rival, but still IRV elects M.
Notice that the first group of voters could have caused C to win if they had only "lied", and put him first in their list. That would mean they'd get their second favorite instead of their fourth favorite. Statistical analysis reveals that this strategy is advised for all candidates who don't appear to have at least a 20% chance of winning. That means that, contrary to FairVote propaganda, IRV does not let you "vote your hopes, not your fears". And this means that IRV effectively degrades toward plain old plurality (vote-for-one) voting. This is explained in more detail here, by math experts:
http://rangevoting.org/TarrIrv.html
Election integrity experts and activists, like computer science Ph.D. Rebecca Mercuri disapprove of IRV because it is conducive to the adoption of fraud-susceptible electronic voting machines. IRV is also more susceptible to fraud because it is not countable in precincts. That is, candidate A could win every individual precinct, but bizarrely lose when the ballots are all summed together - which enforces centralized tabulation, which is more susceptible to central fraud conspiracy. And IRV typically causes spoiled ballots to go up by a factor of about 7.
http://rangevoting.org/SPRates.html
But FairVote is right about one thing. We need a better voting method than the incredibly terrible plurality system. The best combination of quality and simplicity is called Approval Voting. It's just like the current system, except that there is no limit on the number of candidates one may vote for.
While it may seem initially less intuitive than IRV, deep scrutiny shows that Approval Voting produces a far more representative outcome. This is shown through an objective economic measure called Bayesian regret, which shows how well a particular voting method tends to satisfy the preferences of the voters. The improvement gotten by Approval Voting relative to IRV is especially large if the voters are strategic, as was described above (although FairVote promoters will often falsely claim that the best strategy with Approval Voting is to "bullet vote"). See:
http://rangevoting.org/BayRegDum.html
If we don't mind a somewhat more cluttered ballot, we can upgrade to Range Voting, which uses a ratings scale, like Olympics scoring. It is arguably more intuitive, and produces phenomenal Bayesian regret results, meaning more satisfied voters, and more competitive nominees!
For a look at how the major parties could become dramatically more competitive by merely adopting Range Voting or Approval Voting, see:
http://rangevoting.org/ForDems.html
http://rangevoting.org/ForReps.html
Election reformers must be diligent and do their research. Don't be misled by FairVote's clever marketing. Look at what Ivy League mathematicians and political science experts such as Steve Brams, who write entire books on this stuff, say. FairVote has an agenda, and it's definitely not in the pubic's best interest.
Clay Shentrup
San Francisco, CA
415.240.1973
clay@electopia.org