Rules of the Democratic Primary, and how could they be improved?

So the rules of the primaries are:

  • Some (25%) of delegates are awarded at the state level (“at large delegates”). Some are awarded at the Congressional district level (75%). There is also a second category of elected delegates awarded based on the statewide vote called “Pledged Party Leaders and Elected Officials”, awarded in parallel to the first category. The number of Pledged PLEOs a state gets is 15% of the combined number of district level and at-large elected delegates.
  • States are required to distribute delegates among the Congressional districts based on a mix of population and votes for Democrats in Presidential and/or Gubernatorial elections.
  • In addition, there is an unelected category of delegates, known as “Automatic Party Leaders and Elected Officials”. (This is what the party calls superdelegates.) It includes DNC members, Democratic holders of the following offices: President, VP, members of Congress, Governors (and mayor of DC), as well as past holders of the following offices: President, VP, Senate Leader (i.e. Majority or Minority), Speaker of the House, Minority Leader, DNC Chair. They don’t get to vote on the first ballot.
  • The presidential campaigns submit lists of people they approve of as district level delegates from a list of those who have filed to the state party an application to become a district level convention delegate. The rules for how the delegates are chosen from those who were approved varies by state. In some cases, they may be elected by voters in the primary, in which case this step will need to occur before ballots are printed.
  • After the primary, and after the district level delegates have been selected, the campaigns approve people who apply to be delegates in one of the two statewide categories (at large and pledged PLEO). They must approve at least one candidate per seat won in that category. In some states they must approve at least two candidates per seat won in that category. The delegates are chosen from this list by either the district-level delegates, or a committee of party officials elected not before 2016.
  • In the primaries themselves, voters pick one candidate.
  • For each level at which delegates are selected, candidates are awarded delegates using the largest remainders rule of proportional representation with Hare Quotas, however, candidates with less than 15% of the vote do not receive seats, and their votes calculating quota are ignored. In addition, for at-large and PLEO delegates, if a candidate otherwise entitled to delegates drops out, then they will also be ignored.
  • For the first ballot of the convention, pledged delegates must vote for the candidate that they pledged to. However, if a candidate with pledged delegates drops out and releases their delegates, they do not have to. At the convention, the winner must get a majority, or there is another ballot.
  • Caucuses still exist, unfortunately.

I think this is close to being a good system (i.e. a few changes away; without them, it’s still pretty bad), despite only allowing voters to choose one candidate, because delegates are assigned by a proportional rule, nomination requires a majority vote at the convention, and delegates are not bound to candidates who drop out. This allows later preferences to be expressed to a certain extent via delegation, since ideally the later preferences of a candidate’s delegates should reflect those of their voters. However, in practice, this may not be the case, since the rules for selecting delegates from those approved by the campaign allow nonsupporters of a campaign to have influence over the delegation of that campaign. State level delegates are chosen by the state party committees or the state’s district level elected delegates. They may choose delegates whose later preferences reflect their interests rather than that of the voters for that candidate.
This is even true when district level delegates are selected by voters. My state allows voters to directly select district-level delegates. It uses a modified version of PAL, in which the delegate candidates are separated by gender, but not candidate preference (which is listed on the ballot). Delegates are elected in a manner that preserves a preset gender composition, reflects the composition of pledged first preferences implied by the primary vote, while prioritizing those delegates with the most votes. This has all sorts of weird strategic implications that most voters are probably unaware of. (Just selecting the delegates that your first presidential choice’s campaign approves deprives you of any say as to who they will be, since it is standard practice for a campaign to approve only as many candidates are there are seats to win.)
Campaign approval of candidates should act as a bulwark against “Let’s choose the only Sanders>Biden>others voter in the country for the Sanders delegate”, planning for defeat is probably not a campaign’s greatest priority, so they may give enough attention to later preferences.

In general, the DNC should embrace the principle of “delegation as a way of expressing later preferences”, rather than “assume voters have no opinions on what should happen if their favorite doesn’t win.” Embracing the former principle would involve making it easier to find out what the later preferences of a candidate’s approved delegates are, so that voters may take it into account. A blatant case of the latter principle can be shown in the rules for excluding candidates from receiving delegates in certain cases. Voters who select candidates with less than 15% of the vote don’t have any influence on the delegates. Also, when a candidate drops out, they are excluded from receiving statewide elected delegates in those states which haven’t selected their specific statewide delegates. The DNC likely knows the latter principle is false. The real motivation for these rules is likely to generate an illusion of consensus, so the party appears united at the convention. However, allowing supporters of minor candidates to have their voices heard would mean a nominee supported by a greater actual consensus. The 15% threshold should be scrapped. As long as there is a threshold, especially one that high, claiming that candidates are awarded delegates “proportionally” is misleading. It can only be said that delegates are awarded to candidates above the threshold proportionally.

There are too many (3) completely separate categories of elected delegates. The consequences of having this many categories is that the ruleset for awarding delegates is overly byzantine, and fineness of proportionality is sacrificed. Two of the categories are redundant: both “at-large” and pledged PLEO delegates are awarded based on the statewide vote, however, separate applications of the largest remainder rule are applied. For the purposes of determining the number of state level delegates that each candidate is entitled to, these categories should be merged. At first, I was also opposed to the existence of district level delegates. In terms of providing local representation at the convention, dividing them among congressional districts is silly, since due to gerrymandering, they often don’t correspond to actual communities with common interests. They sacrifice fineness of proportionality, since a state’s total number of delegates will likely be two or three digits, but a single district’s delegate share will probably only be one digit. They also harm candidates slightly over 15%, as they will lose delegates in their weaker districts. On the other hand, they help candidates slightly under 15%. Another purpose they seem to serve is to make some voters more equal than others. Since apportionment among Congressional districts is determined in advance, if a lot more voters show up in single Congressional district than expected, their influence will be diluted. This is not necessarily bad, since it helps for things like the weather depressing turnout in a region. So my opinion about this is mixed. I will note that adding the fractional remainders for each category would allow for fineness of proportionality to be maintained while still having separate categories of delegates. It’s not like the average voter knows how the delegate math works anyway.

Potentially worthwhile actions:
If we can find the later preferences of the delegates approved by campaigns and likely to be selected, then make them publicly available, it could help voters cast a more tactical informed vote in the primary. If we can do it with perfect success, we could find the pairwise delegate counts.

Sources for rules:

Actually, I should probably explain the strategic implications of this, because the tactical voting only takes a relatively tiny number of voters (like 2% of the electorate) to exploit. Basically, while saying “I’ll vote for all the [candidate] delegates, because I want [candidate] to win” is a wasted vote for the reasons I explained in the quote, it’s also the most common type of vote. I would estimate that about 80% of the votes in the 2016 Maryland primary were like that (examine the results here and here). Thus only a relatively tiny number of votes differentiate the candidates for delegate who are running to represent the same presidential preference. Also, you do not have to vote for only candidates who represent your presidential preference (if you did it would be an Approval-at-large election with an unknown at time of vote number of winners.) So it takes only a few tactical voters to swing the delegations for each presidential candidate. If you know the later preferences of each delegate, and there is sufficient diversity of secondary preferences within a single presidential candidate’s delegate slate, then you can use this to influence the pairwise preferences of the delegation. If there is a brokered convention, or if you influence the slate of a candidate who drops out before the convention, such tactics would matter.

Oh come on people. You don’t know that the Democrat Party is 100% just as fascist as the Repug Party? What utter nonsense. They are both 100% owned and operated by the corporatist pluto/security complex. Find some way to remove both of these cancers. Such as a strategic hedge simple score election system.

PS –
Please pay attention to what you are doing. Merely labeling this thread as ‘Off-Topic’ will not protect CES from the legal consequences of gratuitous partisanship. If this behavior continues, CES will likely lose its 501( c )( 3 ) tax exempt status. Remember there are people out there who will be happy to cause that to happen.

So are you suggesting that the winner of the primary is entirely unimportant because the candidates are completely interchangeable? Or that the primary is rigged “Iran 2009” style in which the votes have no actual effect on the outcome, in which case obviously tactical voting would be doomed to fail? If neither are true, then there is at least some use in understanding the rules- which most people clearly don’t. If the latter is true, well, what makes you think they can’t rig a score vote?

I think rkjoyce is arguing that the entire Democratic Party is owned by corporate elite along with the Republicans, and we need to get entirely around both the Democratic Party, the Republican Party, and all other parties by using simple score voting in the general election.

(I am not sure what would be the implications of having an actually fair Score election in party primaries and a C1V general election.)

I flagged rkjoyce’s post as off-topic, because it seems like it just totally ignores the whole point of this discussion. If you disagree with that decision let me know.


How is it partisan to improve a democratic election’s outcomes from the perspective of its voters?

=/ How is it partisan to improve a democratic election’s outcomes from the perspective of its voters? /= – above

It isn’t partisan at all. Provided of course that the same treatment is applied to all the other relevant political parties.

But otherwise, it is partisan and in this context, illegal*.

*Please note: I am not a lawyer, so please consult your own attorney.

The Republican Party is probably less amenable to primary reform, so does that excuse focusing on the Democratic primary?

First of all, the Internal Revenue Service will not likely be interested in what party is or is not amenable to primary reform. But I am not a lawyer.

Second of all, it is unscientific for alleged election methods analysts to indulge in any partisanship, or even the appearance of such, in any forum pertaining to that subject. (Except perhaps to a very small extent.)

Therefor, my initial comment, which was so gratuitously dismissed, was an implicit assertion that this thread ought not exist in this forum.

How come FairVote haven’t been affected by this? A large organization like that probably has some partisan connections or other.

Let me explain something. I live in Massachusetts, and for the most part, there is only one party (Mitt Romney notwithstanding). If my reading of the party’s rules and bylaws is accurate, the Massachusetts Democratic State Committee has the power to completely override the results of all of the local caucus nominations, even if those results are arrived at by a unanimous consensus.

Political parties are simply hierarchical private associations that can do whatever they wish. For all practical purposes, ‘major’ parties currently are motivated by financial concerns, and not so much by ideological ones. They do have Bosses. It is very odd to imagine that high party officials of the ‘majors’ will be at all interested in election methods for primary elections without some (probably underhanded) very strong motivations. I have never seen them to have any genuine concern for fairness in the primary rituals.

The proper way to discuss election methods for party primary elections is to propose hypothetical new parties, and then to analyze election methods for those. It would be most beneficial to assume that a social entropy effect applies to such organizations – that is, they tend to become totally corrupt in, very roughly, five years.

(FairVote has no need for any partisan connections. Like the major parties (and, increasingly, the minor ones), it is very tightly connected to the rich class.)

Online platforms, like this one, are not considered legally responsible for the content posted by their users. If I write something on the CES forum, they can’t be held accountable (though they may be asked to take it down if it’s illegal, like copyright infringement).

As for analyzing the GOP to be fair, what’s the point? As long as Trump has a pulse, he’s getting the nomination with >80% of the vote. 5 state parties canceled their presidential primary. The GOP primaries themselves typically award some delegates to the statewide FPTP winner and a few to each district FPTP winner. The campaigns only tend to approve as many candidates for delegate as there are available positions, which, when combined with winner-take-all means that any opportunity to vote on delegates tends to be “Soviet style”. There’s nothing happening that is currently of interest.

=/ Online platforms, like this one, are not considered legally responsible for the content posted by their users. /= – above

This platform happens to be controlled by a 501( c )( 3 ) tax exempt organization. Please think about that.

Anyway, my opinion is that attempting to reform the procedures of a vast mountain of corruption such as any major political party is akin to planting a tree at the top of an active landfill. This may be attempted, but there’s virtually no chance that it will accomplish anything.

I am fairly sure that if one were to insist upon devising election methods for hypothetical political parties, those methods would not be very unlike those that are appropriate for organizations in general. And if this is so, there is not much to talk about. As should be noticed, In regard to election methods I tend to insist upon keeping everything as simple as possible.

Also, the more governments become entangled in the procedures of political parties, the more opportunities for corruption will arise.


  1. Reforming the party to use Score Votingcould set an example for reforming the actual government. We would be able to point to the success of how it worked in practice in an election voters care about. (Despite your own stance, many people genuinely care about which Democrat wins. The whole reason the elites are so successful is that they convince many voters that they are actually good.)
  2. The use of hedge strategy within a party primary could allow disruption of elite capture effect within that party, and then that candidate merely has to win a general election with the advantage of being honest and actually for the people… (That is, primary voters choose candidates not controlled by the elite because the primary does not use C1V. I am thinking of Tulsi Gabbard here, because she gives the impression of an honest person who the establishment hates and who is suffering in the polls, although you might consider her a bad example.)
  3. It is not obviously more or less difficult to change a party versus changing the government. We might as well try both. How come you seem so interested in trying to get general elections changed to SSV but when we talk about parties you just say it is a waste of time? Both are controlled by the elite.

The issue with reforming a party primary is that most parties have final say over whether they’ll approve their election results. It may not be possible to get lasting change from a party, which is legally sovereign, compared to a government, which is bound by ballot measures.

CES is an electoral reform non-profit. If the GOP was having an actual election (instead of canceling half their primaries) it might be worth talking about their primary system. Trying to claim partisan ship for an electoral reform group talking about the most prominent election currently underway would get laughed out of court. Also, the FEC (which currently doesn’t even have enough members to have a quorum and wouldn’t do anything even if they did) would be the body responsible for adjudicating that IIRC.

1 Like

Statement above, with added references:
:~~~~~~~~~~~~~~~~~~~~ //
CES is an electoral reform non-profit. If the GOP was having an actual election (instead of canceling half their primaries) it might be worth talking about their primary system. Trying to claim partisan ship for an electoral reform group talking about the most prominent election currently underway would get laughed out of court [ref 1]. Also, the FEC (which currently doesn’t even have enough members to have a quorum and wouldn’t do anything even if they did) would be the body responsible for adjudicating that IIRC. [ref 2]
// ~~~~~~~~~~~~~~~~~~~~:

Regarding [ref 1]; there would not be any court proceeding. The IRS would simply revoke CES’ 501( c )( 3 ) ‘tax exempt’ status. Regarding [ref 2]; the Federal Election Commission (FEC) has nothing to do with this, it would simply be an IRS ruling (IANAL).

RepresentUs is an organization that strongly promotes and lobbies for ‘RCV’/‘IRV’.

Please note that =/ Below is the full list of donors to RepresentUs Education Fund 501( c )( 3 ) and/or RepresentUs 501( c )( 4 ), December 2018 – June 2019. /= –
Take a look at all those BIG donors!

501( c )( 3 ) organizations have (theoretically) extremely limited privileges to influence elections. 501( c )( 4 ) organizations have extensive privileges to influence elections.

Y’all are free to talk about whatever you want. You are not staff, your actions are not condoned by the board of the Center for Election Science. Nothing you say or do is likely to affect our 501©(3) status.
That said, please don’t push it, or we’ll have to moderate comments.
@rkjoyce your comments aren’t constructive. Whether you believe that the parties are controlled by lizards or not is irrelevant to looking to how they could be more democratic. I would appreciate discussion from the Republican party’s perspective also.

I am deeply saddened by your responses to my suggestions and comments here. I have been seriously studying the issues relevant to election methods since 2004, and as I have proven elsewhere here, have published quite a lot of material since 2006. I have also been president and chairman of a non-profit that oversaw the operations of both a health food store and a health food restaurant. I worked very hard on those projects. I think I am well aware of the nature of 501( c )( 3 ) organizational operations. (It was very difficult to avoid being pushed into the co-op category.)

See what I said:

This tells me that RepresentUS probably must have two independent, but different, non-profit corporations. I have even mentioned elsewhere that CES ought to consider doing likewise. Now is that any of way making comments that are ‘not constructive’?

I also worked very hard on the mathematics, the logistics, and the strategies for implementation of election systems that can truly represent the people. Very hard. I have three main ‘pieces’: strategic hedge simple score, rectangular tiling re-districting, and add-on (virtual) proportional representation. (I am becoming more and more convinced that societies that exhibit any need for proportional representation are simply too defective to survive. There must be some minimal level of co-operation between the minorities if a society is to endure.)

Years of effort have gone into the proper development of these methods. They appear to be rather different than other methods that have been proposed here. But you did say you wanted to hear from people ‘even if they disagree with us’ (somewhat paraphrasing).

Why should I not recommend caution in dealing with institutions I have come to distrust? Am I doing any favor by not recommending due caution in dealing with them?

I have been working with election method analysis very hard for a long time. So CES is not he only coal in my furnace. I have long neglected two of my tiny but relevant blogs for a very long time. They are:

Simple Score Voting


Community Survival

I can also be contacted at ‘reaver’ a+ polarismail do+ net (omit the single quotes).

I have put most of my effort into the CES forum, but have received very little positive feedback. And I feel some individuals have been rather rude. So maybe I should put more effort into my own projects. I do not accept that it was said that “your comments aren’t constructive.” Considering the amount of effort I have expended, that is not truly justified.

I am still with you, but I feel I must ‘branch out’ and do a few things my own way.

I do appreciate that you have provided a service for me, and some of the commenters have been most helpful. I wish you all well.

Sorry about that @rkjoyce - I meant in this specific thread. Not constructive in the manner of not commenting directly on the rules of the democratic party. Not all your work, not all your comments. I’m sorry that I was ambiguous with my speech, and I hope you’ll forgive my lack of precise language, and of tact.