There is some discussion about Bernie Sanders’ strategy for winning the nomination despite being significantly behind in the pledged delegate count. Most of this discussion is nearly worthless because those engaged in it (talking and listening) are, or seem, poorly informed about how the system actually works. So I thought this would be a good time to look at some of the numbers.

First, some context.

The primary process is not Constitutionally democratic. There is no legal requirement that a party nominate someone based on any sort of voting process. But, over the last several decades this has become normal. Having said that, it is also a vetting process. The reason primaries are held over a long course of time is not because no one can figure out how to get all the party’s participants to vote on the same day. Rather, it is a long and drawn out contest in which the different candidates make their respective arguments using various venues, and as this process happens, party participants vote, in groups, for one or the other. It is like a boxing match. There are several rounds and we see how the candidates do over time.

This varies across states (and between the parties, and across time), but the ways in which party participants express their opinions are myriad. There are two ends of a wide spectrum. One is to simply ask people who claim, or demonstrate (depending on state) that they are with, a member of, or associated with a particular political party which candidate they prefer, in a polling process that looks just like an election. That is a primary. At the other end of the spectrum, people gather in public spaces and go through a formalized ritual involving holding up signs, standing on furniture, yelling and wildly gesticulating, and moving around the room, and often electing or choosing a subset of individuals to represent the views of the larger number in that room. Then, usually, they do it again at a later date in a different room. Sometimes a few times. That is sometimes correctly referred to as a caucus.

But a primary and a caucus are not two distinct things. Many contests are somewhere in between. In Iowa, it goes much as I just described. In Minnesota, where we have a caucus system, it doesn’t work that way at all and is much more like a primary where people simply vote.

Party members and operatives may prefer the caucus system because it gets people involved, and also, exposes and allows discussion on a wider range of issues. A party can, and often does, define itself on the basis of issues, and without a caucus system, the issues are defined by a very small number of individuals, often not even directly involved with the party. For example, for the Republican Party, the issues are mostly defined by radio shock jocks like Rush Limbaugh and the Fox News network commentators.

But a caucus system may have the effect of limiting participation, excluding those who can’t make the caucus schedule, for whom an absentee ballot is the only way to vote, etc. That is an argument for dropping the caucus and using primaries instead. But, of course, the counter argument to that is that the caucus invites real, meaningful participation. Which sounds great in theory, until you go to a caucus and notice that only a few people are actually meaningfully participating. But, at least, someone is somewhat meaningfully participating.

There are many other things going on that matter. For example, are you under the impression that people who might participate in a primary or caucus are registered in a particular political party? And, if you are registered in a certain party, you have to change your registration to participate in a different party’s process? Well, that may be true for you, but it depends on which state you live in. For example, in Minnesota, you can’t register in a party. We simply do not do that here. Yet, in Minnesota, you must be with the party to participate. You sign off on a general agreement with the party’s principles, and you agree to participate in only one party’s activities in a given caucus cycle. Yet you don’t register. If this seems odd or even impossible, perhaps you live in New York State or some other place where one formally registers.

The different versions of “party membership” leads to a great difference in what can happen during the primary process. The following pertain.

When you go to vote or caucus,

1) Do all the people go to the same building, regardless of political party? (In some states, yes, in other states, no.)

2) Once you get there, do you have to vote or caucus in the party that you are already registered in, or can you chose at the last second?

3) If you are not registered, or if (as in some cases can happen) you are “registered as an independent”, are you excluded from participating in a particular party’s contest, or not?

4) Is your primary or caucus also the day on which actual constitutional candidates or issues are on a ballot — the official government run ballot, not the party ballot — or is this just a party thing? (Some states often do this, some states never do this, and it affects who shows up.)

4) How does “open” vs. “closed” (able or not able to switch at the moment of the primary) relate to “primary” vs. “caucus” in your state?

We often see it remarked that Sanders does better in caucus states, or in open states, etc. etc. This might be partly true, but given that the following kinds of contests theoretically exist, it is a great oversimplification.

I quickly add that not all these combinations exist, but many do, and even this accounting does not fully reflect all of the actual variation. For example, the exact ways that state-wide pledged delegates are allocated also varies across states, and that variation adds even more complexity. And, two different states may have different ways of allocating delegates, but whether or not the campaigns engage in these methods varies from year to year. For example, in Nevada this year, the Sanders campaign engaged in a ground game to capture more support after the caucus system passed through its first phase, but in Minnesota, as far as I know, neither campaign bothered. That was probably not a strategic difference, but rather, a reaction to the fact that allocation works ever so slightly differently in each of these two caucus states.

There is a reason that political experts like Rachel Maddow have highly specialized lawyers to give overviews of this sort of thing when they want to try to get it right!

Then there is the issue of pledged vs. unpledged delegates.

The following commonly held beliefs are incorrect about this issue.

1) “Superdelegates” exist only in the Democratic party. Not true.

2) “Superdelegates” declare which candidate they will vote for prior to the convention, and thus, can be counted as part of that candidate’s vote total. Not true.

3) “Superdelegates” are members of the party elite who are not responsible to the citizens or voters. Not true.

4) “Superdelegates” destroy democracy and ruin everything for everybody. Not true. Maybe a little true.

5) There is no reason to have “Superdelegates” other than for the party elite to control the outcome of a primary. Mostly not true.

Superdelegates (an unofficial term for unpledged delegates) exist because of the 1968 and 1972 Democratic Party Conventions. In 1968 all hell broke loose, and one of the problems was the perception that the nomination was mainly the product of back room politics. So a fully democratic process was deployed for 1972. But in that years, the Republicans played dirty tricks (remember Watergate?) on the Democrats and messed up the process. So, it was decided that a certain number of delegates should be responsible and involved individuals who were held accountable by the fact that they were elected to their jobs (either as constitutional officers or party officers) or otherwise potentially held in esteem or contempt based on their behavior, and that these delegates would not be pledged as part of the primary and caucus process. They would be the control rods available to fix things when things broke because of murder, dirty tricks, scandal, or any unforeseen event.

Superdelegates do not pledge for a candidate, but they are people, and more to the point, they are politicians. And Americans with First Amendment rights. So, it is easy to understand how many would want to endorse a particular candidate for the party’s nomination, and one could argue that there is no reason to stop them from doing it, and there is no legal way to stop them. Having said that, many Superdelegates choose to not endorse, because they feel, as many do, that their role as the adult in the room when things go sideways is compromised by any such endorsement.

The point is, when the process eventually moves to the convention, we should expect Superdelegates to do the right thing, but we can’t expect that everybody’s version of the right thing is the same. Should a Superdelegate vote in favor of the candidate that won their state? (Not all Superdelegates represent states, some are at large, but this would work for many.) Should a Superdelegate vote in favor on the candidate that is ahead nationally, regardless of how that candidate did in that delegate’s state? If Candidate A is ahead but was recently photographed sitting in a hot tub full of fruit jello with the Primier of North Korea and recorded making a deal for nuclear weapons, perhaps a Superdelegate may decide to switch allegiance to Candidate B. If a candidate does so much better during the last half of the primary process than the other candidate, does this indicate that this candidate is the people’s choice, even if the people who voted during the first half of the election aren’t really being asked? What if national polling indicates that everybody in the party has shifted preference from the winning candidate to the candidate that is behind, for some reason or another? Would these conditions mean that the right thing to do is to support the second (or third) place candidate?

And, when it is all said and done, and a given Superdelegate has made a decision that is at odds with the constituency of the home state or district for that delegate, will people remember that it is their job in our great democracy to exact vengeance and hold that delegate’s fee to the fire next chance they get?

OK, now back to Sanders’ strategy and the numbers. One of the things we hear today is how Hillary Clinton lost and Barack Obama won in 2008, even though it was assumed that Clinton would win. Also, at some point, Clinton threw in the towel and urged support for Obama. This year, it looks like Clinton is going to win the nomination, and Sanders is not, so at what point does Sanders support Clinton?

What Sanders should do, of course, is to keep his campaign running fully until the convention, though he could bow out sooner if he chooses. If, for example, he comes in second in New York and, especially, if Clinton picks up a lot of delegates there, one could not fault Sanders for shifting his support from himself to his opponent. At the same time, one can not fault him for choosing to not do that.

It has been suggested, vaguely by the Sanders campaign and interpreted by the press, that Sanders’ strategy at this point is shifting from winning the pledged delegate count to getting the Superdelegates to support him, thus giving him a win on the first ballot at the Convention. Short of this, he could get a sufficiently large subset of the Superdelegates to support him, forcing Clinton to not win on the first ballot.

After the first ballot, the pledged delegates are more or less free to vote for whom they want (though it may not be that simple and can vary by state, and there are rules that don’t exist yet pertaining to this). So if, for example, it is true (though there is no credible evidence of this) that Democrats across the nation have shifted their preference towards Sanders, even if they voted or caucused previously for Clinton, then Sanders might do well on the second ballot.

So now we come to some interesting numbers.

Each primary season is different, in terms of ordering of states, total delegates, total needed to secure a nomination, etc. etc. I looked at the Obama-Clinton race and found the moment where the total number of pledged delegates fought over was roughly at the point we are now in the Clinton-Sanders race. At that point in time, Obama had 1159 pledged delegates, and Clinton had 1059 pledged delegates, for a spread of 100 delegates.

At this point in the process, measured as stated, this year, Clinton has 1255 delegates and Sanders has 1012 delegates, for a spread of 243.

(These numbers are no perfectly matched because there is no way to do that for obvious reasons.)

Putting this another way, at this point in the process, Hillary Clinton is significantly more ahead of Bernie Sanders than Barack Obama was ahead of Hillary Clinton eight years ago.

I hasten to remind you that these numbers do not include the unpledged Superdelegates.

There are two reasons these numbers are important. First, this spread is large enough to argue against Superdelegates going into the convention with a shifted alliance towards Sanders in order to act as a control rod on a very close and dynamic race. That argument could weaken if Sanders does exceedingly well in New York, of course. The second reason is that the number of Superdelegates that would have to be convinced to make this move would be very large, probably beyond what is possible.