One thing you should know before discussing the primary process, there are new rules for how delegates are to be awarded.

The total number of delegates in play on the first vote will be 3,768. To gain the nomination, a candidate will have to get a majority of this this number, or 1884 plus one or more, on the first vote. There are the usual “pledged” delegate vs. “unpledged” (the latter sometimes called “superdelegates”) but with fewer of the latter than in previous years, and they will not be voting on the first ballot. If no candidate meets the 1884+ threshold on the first ballot, all the delegates are released from prior pledges, and superdelegates are thrown into the mix. Then, 4,532 delegates are in play, and a majority, or over 2266, will be required to win.

That is something of an oversimplification. If a single candidate goes into the convention with something like 2,267 pledged delegates, then superdelegates will be allowed to vote. Notice how close the supermajority of pledged and the 50% threshold of all, are. It feels like astrology, but I digress.

Among the pledged delegates, there is a 15% threshold rule per state in allocating delegates. If a candidate gets 15% or more of the vote/caucus delegates, they are in the running for allotted delegates. Then, among those who pass 15%, the delegates are allotted proportionately. If no one gets 15%, then the threshold shifts to one half of whatever the front runner got. So, if the leading candidates gets 12%, then the new threshold is 6%.

Here are two of many possible examples of what could happen in a given state.

The Iowa Caucuses are on February 3rd. In polls, Biden and Sanders are about even, with Buttigieg and Warren competing for third place and all seem to be at or above the threshold. However, the difference between public opinion polls and outcomes is potentially large in a caucus state, because the variation affected by “ground game” is directly reflected in polls when there is a primary, but not in a caucus. In Iowa, keep an eye on Klobuchar, who claims to have a wining or at least result-surprising ground game in the Corn State. That is not a false claim. In other words, anything can happen in Iowa. Iowa will be deciding the commitments for some 41 pledged delegates.

I currently predict, and this is a pure thumb suck estimate, that the four current front runners (Biden, Sanders, Buttigieg and Warren) will roughly split Iowa’s 41 pledged delegates, with Sanders taking the largest share, and Warren the smallest share.

Then comes the famous New Hampshire Primary, on February 11th. New Hampshire has 24 pledged delegates, a very small number, but the Granite State is famous for being a tail wagging the giant sausage making political dog of democracy. There is a good chance that New Hampshire will break in a very similar way as Iowa, with Biden, Sanders, Warren, and Buttigieg all reaching the threshold and sharing delegates with Biden and Sanders getting equal numbers at the top, Warren third, and Buttigieg fourth. But, either Warren or Buttigieg, or both, could fail to meet the 15% threshold. The latest Boston Herald poll has Buttigieg way below that number and Warren near it. Other recent polls have both below it. New Hampshire may well be the make or break moment for Buttigieg.

Then comes the Nevada Caucus on February 22nd. As usual, Nevada has less polling than other states, but there is enough to identify Biden, Sanders and Warren as, once again, the top tier, but with Warren repeatedly polling at just below the threshold. Buttigieg hovers just below them, and not looking like he’d get the 15% threshold. There is a good chance Biden and Sanders will split Nevada’s 36 pledged delegates roughly evenly. This could be a make or break caucus for Warren. But, maybe not.

Then, on Feb 29th, we have the South Carolina primary. The first two events are mainly white semi-rural or rural people deciding who should be president. Nevada Democrats have significant diversity but mainly Hispanic, and a strong labor component. But elections are won or lost on the basis of African American support in this country, and South Carolina is the first event with significant African American participation. Here, Biden is way ahead of everyone else, with Sanders and Warren sharing a distant second place, and hugging that 15% threshold a little too closely for comfort. It is possible that Biden will walk away with all of South Carolina’s 54 pledged delegates. Recent polling has shown Steyer as a factor in this state, and if that is correct, it could be Biden and Steyer splitting those delegates at about a 2:1 ratio. That all depends on if we believe Steyer is for real. I, personally, am not sure.

At this point, what we know now will still be true: Biden and Sanders are front runners. Warren is a factor, likely Butigieg is a factor. If nothing unusual happens, we will be entering Super Tuesday with a Biden-Sanders fight. However, Warren could outperform and pop, or Butigieg could take the threshold in three of these four states, or Steyer could buy his way in, er, I mean, well, whatever, you know what I mean. The point is this: We are watching a horse race with two odds-on horses, both old white guys but one progressive and one centrist, and one of them likely to win. But, there are these two or three other horses in the race that could woosh by either or both of them in these first four furlongs.

But then, Super Tuesday comes along. Sixteen entities, mostly states, vote on Super Tuesday, for a total of 1357 pledged delegates. Using information from polling, or if no polling exists, the thumb-suck-estimate method, assuming that no candidate has an unexpected break-through event in early states (or otherwise), and assuming that Biden, Warren, and Sanders are the only candidates likely to be viable for most of the primaries (Klobuchar will take a good number of Minnesota votes), the following shows a reasonable estimate of the outcome of Super Tuesday. Remember, this is based only on polls (this is not a predictive model) and polls are sparse in many of these states.

This is, in my view, the “null model” of what is going to happen between now and the day after Super Tuesday. It is a model to be defied by individual candidates, broken by the voters, altered by circumstances, manipulated by the Russians, etc. There is more uncertainty in this season’s Democratic primary than seen in the recent past, especially with a couple of billionaires showing up at the last second to buy the presidency, and according to the polling, making a dent.

Each of the four graphs represents polling from April to the present.

Each graph has four lowess smoothed lines representing the data, with all four candidate’s lines on each of the four graphs.

Each candidate gets its own graph, with those four lines, but with that candidate’s polling data shown.

The polling data is normalized so that each point is the proportion of polling points across just these four candidates, so there are no effects of other candidates entering or leaving the race.

All four graphics are on the same scale, but since I don’t want scale or actual polling numbers to be the focus, I did not include them.

What we see here is four distinctly different patterns. Biden is a top candidate with a declining campaign. Warren is an up and coming candidate with an expanding campaign. I used to use the term “flat line” to describe Sanders’ numbers, but I won’t do that any more. Let’s just say he has a nice, healthy, straight line that does not go up or down.

Harris has a campaign that experienced a temporary peak that went away.

Pretty cool graph, if I may say so myself.

So, in that pattern we get a middle of the country white state, an eastern white state, an eastern southern state with a large African American population, and a random labor state (Nevada Democrats are union workers in the casinos, and such), first, and thus, an early look at what some semi-representative parts of the nation think of the candidates.

But then, this year, right away, boom, Super Tuesday.

So, CBS has put together a poll on the standing of the candidates in all the Super Tuesday states plus the first states. And, it is rather amazing.

Here is the breakdown:

Tier 1
Biden 25%
Warren 20%
Harris 16%
Sanders 15%

Tier 2
Buttigieg 6%
O’Rourke 4%
Castro 2%

Tier 3
Everybody else, all under 1%

Now, we need to adjust slightly. Note that Texas is in this poll. That is why O’Rourke is in the second Tier, I suspect. And, good for him. If his candidacy could guarantee Texas it would be good. Klobuchar’s Minnesota, also a Super Tuesday state did not help here in this poll. Among the many Minnesota Democrats I hear from, the most widespread comment I get is “Amy’s great. Senator Amy is great,” and that’s about it. We may be keeping her home. I suppose California being in this group may be helping Harris. Note also how well Warren is doing. Mass is a Super Tuesday state but I would think New York Dems would be very big on her.

For this poll, I suspect that Warren is underestimated and O’Rourke is overestimated, but overall, these numbers are roughly as expected. Notice that as noted elsewhere, Sanders is a 16%er and that’s it. Not much more and not much less.

I made a very special graphic, have a look:

Following in part on the procedure discussed here, this analysis combines data from several time-overlapping polls to produce a neater and cleaner depiction of each of the top four candidates march towards the presidency … or not.

It turns out that polls come in clusters. There will be several days in a row with a bunch of polls coming out, and then there will be a few days with no polls at all. There are reasons for this I won’t go into now. And, these polls, in the clusters, tend to overlap in time. For this reason, it is easy to take a bunch of polls in such a cluster and average out the results to give a better than average snapshot of a candidate’s status for a given period of time, usually about a week. Then, these withing cluster estimates are somewhat independent from the other clusters because there is no overlap in time, for the most part. The power of each estimate is very high, the trends depicted across the estimates are very likely.

That’s what the graphic above shows for Biden, Sanders, Warren, and Harris. Trends I noted in the previous several blog posts are apparent, but more cleanly depicted.

Here is what this graphic, based on 38 national polls, shows:

1) Biden has had a steady decline, and the rate of that decline may have increased after the first and so far only debate, but he is still number one.

2) Sanders has had consistent, immutable, results the whole time, never changing. It is like there is a certain number of people who support him, and they are not budging, nor are they gaining allies.

3) Warren started to rise in the polls well before the debates. This seems to have corresponded with intensification of her campaign, and her issue oriented displays of knowing things and having plans. Most experienced candidates and campaigners will tell you that is a bad approach. For Elizabeth Warren, it may have moved her into second place.

4) Harris was steady in her just barely 10% status — remarkably flat in fact — until the debate, when she suddenly rose almost meteorically, but not beyond the first cluster.

Is Warren’s rise more stable and issues and candidate based, therefore long lived, while Harris’s rise is a temporary bump from going after Biden in the debates? Is Biden’s downward trend going to continue at its newly accelerated rate or will it flatten out a bit, as hinted in these numbers?

To find out the answers to these and other questions, stay tuned!

But seriously, the next cluster of polls will be available in less than a week from now, most likely. The current pattern requires that the average for Biden be 35% or lower. Warren needs to be a strong second with over 25%. Sanders, while looking very flat, is actually down at his lowest rate in this sequence at present. Sanders should drop below 20%. Harris is likely to stabilize at around 20 or drop back to below 20. Or, she will rise to the mid 20s at the expense of Biden, mostly.

In evaluating these projections, remember how they are calculate. The poll numbers you see will all be lower than those mentioned here because of this. I don’t have full confidence in these projections, but when I say it all out loud, it seems right.

Of course, the debate will shake all this up and possibly add or remove individuals from this top three spot.

This moves Biden from bottom to middle tier for me, which makes me feel better about the fact that he is crushing everyone else in early polls.

California Convention. Since California a) has more electoral votes and more national party delegates than any other state, and b) is a Super Tuesday state now, all of the sudden for the first time in memory, the California Convention received additional special attention outside of California.

And, candidates were sorted. Have a look:

Yay Warren! Yay Sanders! Yay Buttigieg! Yay Harris! Boo Hefferlooper, Boo that other guy!

Perhaps California Democrats are not the same as other Democrats, but in fact, they aren’t different. The outliers in the Party of Kennedy and Wellstone are the right wingers found here and there in Old Dixie or or the High Plains, and a few machine cities or country states in Appalachia or the south. I think we saw some of the herd thinned out in California.

Head to heads. In a recent Quinnipiac poll held in Texas, Biden beat Trump in the head to head, but Trump beat all the other tested candidates. In Michigan, Biden and Sanders trounced trump in the head to head, and Warren, Harris, and Buttigieg did fine. Who cares. Trump was going to win Texas anyway, since Texas is populated with so many god fearing evangelicals who love them their transgressors.

Warren. Warren remains a weak third, but consistent in that spot. In the frontline primary states (New Hampshire, Iowa, Nevada, and South Carolina) it is typically Biden and Sanders in first (strong) and second place. In the latest North Carolina poll (which is not South Carolina, but still, has a lot of African American voters and it is near South Carolina) that held true, but Warren pulled a very strong third (39-22-15). But generally, Warren, while usually in third place, does not break single digits and is statistically in the same bed as Harris and Buttigieg.

Yang, Gabbard, Ryan and Inslee are number one candidates. And by that, I mean, if you round up their numbers, the get to 1%. I don’t see a way up for them, even though this is very early in the race. Klobuchar, Booker, and Castro are consistently in the wings, the one digit 1-3 point wings, and there are things about them that might make them factors later on. They seem to be keeping their powder dry. O’Rourke and Buttigieg could possibly be described as candidates that peaked but then sort of guttered. They are still in the race, but at the moment they were supposed to ride into town on their dark horse, the horse was doing something else that day.

Until proven otherwise, it feels like a race between Biden, Sanders, Warren, and Harris, with Warren and Harris ready to move ahead at any moment, though the Buttigieg-O’Rourke-Booker faction looms small in the background.

In other words, I have no faith in the idea that it is a totally open race. It is a race between twenty-whatever people in which a maximum of five are for real, and we know who the top two or three are and the next two or three will come from a small set of the remainders.

I also have no faith in the order of the leaders. Biden has a history of guttering. I don’t see Sander support moving because of Sanders, but rather, because he absorbs support from other candidates. If ever there was a primary season where an early adoption of a veep is tempting, it is this one. A wavering Biden could be surpassed by a suddenly formed team of two of the top non-front runners, as long as one of them is Sanders. I hasten to add this piece of classic advice about vice presidents: Don’t do that. No talk about the vice president until the convention.

(Hickenlooper and Delaney need new campaign managers. Or just don’t bother.)

And it turns out that this was the Russians. The people doing this were not the Russians. Rather, the Russians, either working for Trump or working in parallel with him (we shall eventually find out) were isolating vulnerable individuals, individual who fit a certain discernible pattern of attributes, and psychologically manipulating them to rage against Hillary.

Looking back, it is now pretty clear that this vitriolic wall of hate, rising up unexpectedly and looking a certain way, was an anomaly. Yes, yes, there was all this Hillary Hate before, but that was from a different demographic, had a different look, and a different feel. This new thing was a different thing, and looking back, one can see it clearly.

And now, it is back. The statements being made that, when you run them down, are not based on reality. The vitriol, almost threatening way it is put. The logical conclusion of the rhetoric being self defeating and damaging to the Democratic Party and progressive idea.

But this time it is about Bernie. I will bet all my bagels and muffins against all your donuts that the Russians are hacking us again, but this time, instead of manipulating Bernie-favoring people to hate bigger and better on Hillary, they are manipulating Hillary supporters to bash Bernie.

It is the same level of vitriol, the same badly sourced accusations, the same direct link between accusation and a final solution of leaving or tearing apart the Democratic party, etc. It appeared in the social networking world instantly, and it is suddenly everywhere. Have you seen it?

I’ll give yo an example. Perez gets up in front of an audience and all the Bernie Bots boo him. That is what is being said. What really happened is that Perez got Bernie to do a talk in a place where Bernie was very popular, to shore up Democratic support, and a lot of Bernie people cheered the heck out of Bernie. There was not moment of booing Perez, though there may have been one moment of mixed hemming and hawing. The actual non-false, real thing that happened is that Perez gave a speech that had the audience cheering and on their feet, then Sanders gave a speech that had the audience cheering and on their feet. If you don’t believe me, you can watch it yourself, below.

But this was converted into a false accusation that Bernie was trying to ruin the Democratic Party. The vitriol is intense. He’s a socialist, run him out. He’s an independent, run him out. He lies all the time. Etc.

The result of this falsehood laced vitriol is to split anti-Republican and anti-Trump forces and to throw the grassroots of the Democratic Party in to chaos.

OK, I do not know that this is Russian Hacking. But it looks exactly like the Russian anti-Hillary hacking. It has the same form, the same technique, some of the same rhetoric, and is exploiting a similar set of vulnerable individuals. If this is not Russian hacking then, indeed, you can have all the muffins and bagels in the land for yourself.

Tom Perez Was Not Booed; that is a pernicious lie

How many Democratic events have you been to? This was a pretty typical one, but slight bit more raucous. There is a heckler yelling something in the beginning, I have no idea what. Hecklers could be paid operatives, or just crazy people, or just people who are not fully in control of themselves.

The so called “booing of Perez,” which was not booing of Perez, happens at just after 6:18. Perez is not on the stage, therefore he is not booed. The speaker asks the audience if they are there to hear about the future of the Democratic Party and the new chair. This is an audience of people who came to hear Sanders. It is a light hearted, fun, charged up rally. If you say to them, “you are here to see ______” where the blank is filled in with anything at all that is not the keynote speaker, they will boo. So, some booed, some cheered, it was pretty ambiguous, and most importantly, UTTERLY MEANINGLESS. It is this moment of alleged but not actual booing of Perez that is among the items being used to bash Bernie and create this unnecessary division.

Now, watch Perz at just after 32:00 . He gives a great speach. He is cheered and loved. there is not Perez hate here.

The people who are being manipulated by this latest round of psychological warfare are unlikely to be convinced that they are wrong. Assuming this is manipulation, the psychology is immune, laced with paranoia and preformed hatred of disagreement. What needs to happen is this: People need to realize that the hacking that happened before can happen again and, probably, is happening again now. The other thing that needs to happen is that the individuals who are doing this hacking, who can’t find their center, their rational self, and slough it off, need to be isolated. Don’t engage, don’t follow, just … well, just do this:

Because that is what your friend’s facebook page looks like.

A candidate is the presumed nominee when she or he obtains the required number of pledged delegates to be at 50% plus a fraction in the total pledged delegate count. This is because a candidate must have a true majority to win the nomination when the delegates are all counted up at the convention, and the pledged delegates are required to cast their lot with the candidate they are pledged to, assuming that candidate exists at the time of the convention.

Hillary Clinton and Donald Trump have not reached that bar. Therefore, neither is the presumed nominee for their party.

But then there are the unpledged delegates. Unpledged delegates can vote for whomever they like at the convention, and therefore, anything can happen. However, it is the practice among unpledged delegates to “endorse” or otherwise show support for a particular candidate. News agencies may use that statement of support to place that unpledged delegate in the column for a particular candidate.

Using this form of math, Trump did not reach a true majority of delegates a couple of weeks ago for two reasons. First, the Republicans have very few truly unpledged delegates. (The Republican and Democratic systems are not parallel or comparable, but the Republicans do have a certain number of delegates who can do what they want when the convention rolls around.) But then, one day, a bunch of unpledged delegates from one of the Dakotas made a statement. They said that they would definitely cast their ballot for Trump in the Convention. This was just enough to put Trump over the top, by adding together the pledged delegates that were pledged to him, and this small number of “unpledged” but now “pledged-ish” delegates.

That is still not clenching the nomination, because even though those Dakota delegates went beyond support or endorsement, to the level of actually promising to vote for Trump, they really still don’t have to vote for him.

But, the press took this as an event, and decided to go with it, and Trump became the actual nominee.

That may seem like a digression in a post about the Democratic primary, but it is relevant because the press has this thing they do where they balance or equalize. Therefore, even though the systems are not truly comparable or parallel, and the event in the Dakotas was actually meaningless, the press did in fact go with the “Trump is the presumed nominee” thing, and therefore, one should expect, even in the absence of a logical underpinning to the argument, the press to do the same in the Democratic party. That is only a small part of the story, but it is part of the story.

I should reiterate that unpledged delegates (in the Democratic party, unofficially called “Super Delegates”) are unpledged even when they pledge. That is a simple fact. But, there are nuances. For example, I know one Super Delegate that on principle will not declare for a candidate until the convention. But I also know that this individual liked Bernie Sanders. I suspect that this means that under some conditions, this delegate would vote for Sanders, but maybe not. I know another Super Delegate who has endorsed Clinton, and another who has endorsed Sanders, publicly. However, I do not assume that either one of them will absolutely vote for that candidate. An endorsement is not a pledge. If Bernie Sanders is found sitting in a hot tub full of fruit jello with the leader of North Korea on a yacht owned by the Koch Brothers, making a deal to trade nuclear warheads, that the delegate that endorsed Sanders will not cast a ballot for Sanders at the convention. But the pledged delegates from the same state will be forced to by the rules. (This is why we have Super Delegates. This is also why we can expect the Republicans to add a higher percentage of unpledged delegates when they rewrite the rules for the next primary season.)

The Dakota delegates, however, did something different. They did not endorse, or show support, but they pledged. However, their pledged is, in fact, legally irrelevant.

And now, we come to 2008. It could be said not too inaccurately that a point in time came during the 2008 nomination battle between now President Obama and Hillary Clinton, when it became apparent that Obama was going to win, the press said so, and Clinton took two days or so off and came back into the ring no longer fighting Obama, but now as part of his tag team.

And, it could be said not too inaccurately that this same moment came in the present election about now. Staring a few days ago, various members of the press began to note that this moment was upon us, and to imply that it would be unfair to Hillary to have given this moment to Obama in 2008, but not give it to Clinton now. I think the belief 48 hours ago might have been that this moment would definitely be on us by the end of the voting process in today’s primaries, but then another thing about the press came into play. The press has to treat everybody and every event like they are all identical blue Smurfs but they also have to do things first, to beat out their rivals, to scoop. In fact, this “moment of clinch” could have been after the Puerto Rico primary, or even earlier. And it was absolutely going to happen after Tuesday. So, AP jumped out of the gate and made it happen Monday, and this is now the True Reality.

So, let us review.

Hillary Clinton is the presumed nominee because she has almost enough pledged delegates plus a gazillion unpledged delegates.

However, part of the impetus for declaring this is that Trump got that courtesy two weeks ago.

But, Trump was the only person running in that race, and Clinton still has an opponent.

Still, numerically, Clinton can’t not get the nomination because she has many hundreds of Super Delegates and Sanders has only a few dozen.

On the other hand, Super Delegates are unpledged. UNpledged. We argue all along that they should not be counted. Then suddenly we count them. Is that fair?

One could say, however, that it is fair. At some point it becomes fair because the numbers become so tilted. If the hundreds of Super Delegates that have endorsed Hillary decided to randomize their preference using a coin biased in favor of Sanders, there would still be more than enough to put Clinton over the top.

It is only fair to Clinton that she gets the same treatment as Obama.

It is only fair to Sanders that she not.

And on and on it goes.

So, is there a good reason that Hillary Clinton is now regarded as the Democratic Party nominee for the office of the President?

Yes.

And no.

A good part of the reason that both answers are valid is because the press has painted themselves into a corner located between a rock and a hard spot and have only a Hobbson’s choice. That is a bad reason. Another reason is fairness. That is a good but not overwhelmingly good reason. There is no reason that the process one year needs to be the same as other years, since presidential election years are so different in so many ways. Another reason is math. While we wish to keep the Super Delegate count separate and let the pledged delegates do their job, at some point the Super Delegates should probably be considered as a factor, if not counted precisely. (See this, “Fixing The Super Delegate Problem,” for an alternative way of doing this whole thing.) That is probably reasonable and fair. If the numbers are big enough. But there is no objective criterion for when the numbers are big enough. So maybe not so fair.

So here is where the honest conversation part comes in. There really is no considered, informed, honest position on this that ignores the complexity and dismisses other opinions out of hand.

I hope you read this post on Tuesday, June 7th, because starting the next day it is not going to matter too much.

Why predict primaries and caucuses?

Predicting primaries and caucuses is annoying to some people. Why not just let people vote? Polls predict primaries and caucuses, and people get annoyed at polls.

But there are good reasons to make these predictions. Campaign managers might want to have some idea of what to expect, in order to better deploy resources, or to control expectations. But why would a voter who is not involved in a campaign care?

I had a very particular reason for working on this project, of predicting primaries and, ultimately, the course of the Democratic race for the Democratic nomination as a whole. When this campaign started, there were several candidates, and they all had positive and negative features. Very early in the process, all but two candidates dropped out, and I found myself liking both of them, though for different reason. I would have been happy supporting either Hillary Clinton or Bernie Sanders.

Personally I believe that it is good to vote, during a primary, for the person you like best in direct comparison among the other candidates. But at some point, it may be wise to support the one you feel is most likely to win. There are two closely related reasons to do this, and I think most observers of the current campaign can easily understand them. One is to help build momentum for the candidate that is going to win anyway. The other is to limit the damage that is inevitable during a primary campaign as the candidates fight it out.

So, early on in the process, I decided to see if I could produce a reliable method to predict the final outcome of the primary process, in order to know if and when I should get behind one of the candidates. That is the main reason I did this. In order for this method to meet this and other goals, it had to be more accurate than polls.

There are other reasons. One is that it is fun. I’ve been doing this in primaries and general election campaigns for quite a few elections. I like data, I like analyzing data, I like politics, I like trying to understand what is going on in a given political scenario. So, obviously, I’m going to do this.

Another reason is to test the idea that the voters are changing their minds over time. In order to do this one might use all the primaries and caucuses to date to predict future primaries and caucuses, and then, if the predictions go out of whack, you can probably figure that something new is going on. This relates to overall feelings among the electorate as sampled by each state, but it also relates specifically to ideas about why a particular state reacted to the campaigns the way it did.

An example of this came up recently when Bernie Sanders won in West Virginia. My model had predicted a Sanders win there, and the actual vote count was very close to the prediction. Since that prediction was based on voter behavior across the country to date, I was confident that nothing unusual happened in West Virginia. But, something unusual should have happened there, according to some conceptions of this campaign.

The economy of West Virginia is based largely on coal mining, and there are lot of Democrats there. (Democrats in local elections; they tend to vote for Republicans in the general.) So, it was thought that the voters would pick a candidate based on a perceived position on climate change and coal. Clinton went so far as to pander to the West Virginians with a rather mealy mouthed comment about how we could still keep mining coal as long as we figured out a way to have it not harm the environment. That was the Clinton campaign doing something about the coal mining vote. Others thought that a Sanders win there would indicate that he somehow managed to get a strong climate change message across to coal miners. That idea is a bit weak because when it comes down to it, Clinton and Sanders are not different enough on climate change to be distinguished by most voters, let alone coal supporting voters. In any event, the win there by Sanders was touted as a special case of a certain candidate bringing a certain message to certain voters. But, he then lost in the next coal mining state over, Kentucky, and in both states the percentage of voters that picked Clinton and Sanders was almost exactly what my model predicted, and that model was not based on climate change, coal, or perceptions or strategies related to these things, but rather, on what voters had been doing all along.

So, nothing interesting actually happened in West Virginia. Or, two interesting things happened that cancelled each other out perfectly. Which is not likely.

In short, the closeness of my model to actual results, and the lack of significant outliers in the overall pattern (see below), seems to indicate that the voters have been behaving the same way during the entire primary season, by and large. This is a bit surprising when considered in light of the assumption that Sanders would take some time to get his message across, and pick up steam (or, I suppose, drive people over to Clinton) over time. That did not happen. Democratic voters became aware of Sanders and what he represents right away, and probably already had a sense of Clinton, and that has not changed measurably since Iowa.

How does this model work?

For the first few weeks of this campaign I used one model, then switched to an entirely different one. Then I stuck with the second model until now, but with a major refinement that I introduce today. The reason for using different models has to do with the availability of data.

All the models use the same basic assumption. Simply put, what happened will continue to happen. This is why I sometimes refer to this approach a a “status quo model.” I don’t use polling data at all, but rather, I assume that whatever voters were doing in states already done, their compatriots will do in states not yet done. But, I also break the voters down into major ethnic groups based on census data. So, for each state, I have data dividing the voting populous into White, Black, Hispanic and Asian. These racial categories are, of course, bogus in many ways (click on the “race and racism” category in the sidebar if you want to explore that). But as far as American voters go, these categories tend to be meaningful.

The fist version of the model used exit polling (ok, so I did use that kind of polling for a while) to estimate the percentage of black voters who would prefer Sanders vs. Clinton. I used the simple fact that in non-favorite son states that were nearly all white Clinton and Sanders essentially tied to estimate the ratio of preferences for white votes at about even. I ignored Hispanic and Asian voters because the data were unavailable or unclear.

This model simply simulated voters’ behavior (in the simplest way, no randomization or multiple iterations or anything like that). I also used some guesses (sort of based on data) of the ethnic mix for Democrats specifically in so doing. That somewhat clumsy model worked well for the first several primaries, but then, after Super Tuesday there were (sort of) enough data points to use a different, superior method.

This method simply regressed the outcome of the primary (in terms of one candidate’s percentage of the vote) against the available ethnic variables by state. Early on, the percentage of Hispanic or Asian did not factor in as meaningful at all, and White and Black together or White on its own did not work too well. What gave the best results was simply the precent of African Americans per state.

“Best results,” by the way, is simply measured as the r-squared value of the regression analysis, which can be thought of as the percentage of variation (in voting) explained by variation in the independent variable(s) of ethnicity.

Primaries vs. Caucuses and Open vs. Closed

Many things have been said about how each of the two candidates do in various kinds of contests. We heard many say that Sanders does better in Caucuses, or that Clinton does better in closed primaries. During the middle of the primary season, I tested that idea and found it wanting. Yes, Sanders does well in caucuses, but the ethnic model predicts Sanders’ performance much better than the caucus-no caucus difference. It turns out that caucusing is a white people thing. There are no high diversity states where caucusing happens. It is not the caucus, but rather the Caucasian, that gives Sanders the edge.

This graph shows how Sanders vs. Clinton over-performed in caucuses vs. primaries.

The value plotted is the residual of each contest in relation to the model, or how far off a theoretical straight line approximating the pattern of results each contest was. Two things are apparent. One is that caucuses are less predictable than primaries. The other is that while Sanders did over-perform in several caucuses, this was not a fixed pattern.

This graph shows the residuals divided on the basis of whether the contest was open (so people could switch parties, or engage as an independent) vs closed (more restricted).

Open contests were more variable than closed contests, but it is not clear that either candidate did generally better in one or the other.

After many primaries and caucuses were finished, there became enough data to use the kind of contest as a factor in conducting the regression analysis. There are a lot of ways to do this, but I chose the simplified brute force method because it actually gives cleaner, and more understandable, results.

I simply divided the sample into the kind of contest, and then ran a multivariable regression analysis with each group, with the percent of Sanders plus Clinton votes cast for Clinton as the dependent variable, and the percentage of each of the four ethnic categories as the independent variables. There are some combinations of caucus-primary and open-closed/semi-open/semi-closed that are too infrequent to allow this. For those contests, I simply developed a regression model based on all the data to use to make a prediction in each of those states. The results, shown below, use this method of developing the most accurate possible model.

How does this sort of model actually make a prediction?

The actual method is simple, and most of you either know this or don’t care, but for those who would like a refresher or do care…

The regression model, using multiple variables, produces a series of coefficients and an intercept. You will remember from High School algebra that the formula for a line is

Y = mX + b

X is the independent variable, along the x axis, and Y is what you are trying to predict. m is the slope of the line (a higher positive number is a steeply upward sloping line, for example) and b is the point where the line crosses the Y axis.

For multiple variables, the formula looks like this:

Y = m1(X1) + m2(X2) + … mn(Xn) + b

Here, each coefficient (m1, m2, up to mn) is a different number that you multiply by each corresponding variable (percent White, Black, etc.) and then you add on the intercept value (b). So, the regression gives the “m’s” and the ethnic data gives the “X’s” and you don’t forget the “b” and you can calculate Y (percent of voters casting a vote for Clinton) for any given state.

So, enough already, who is going to win what primary when?

Not so fast, I have more to say about my wonderful model.

How have the public opinion polls done in predicting the contests?

Everybody hates polls, but like train wrecks, you can’t look away from them.

Actually, I love polls, because they are data, and they are data about what people are thinking. The idea that polls are inaccurate, misleading, or otherwise bogus is an unsubstantiated and generally false meme. Naturally, there are bad polls, biased polls, and so on, but for the most part polls are carried out by professionals who know what they are doing, and I promise that those professionals are aware of the things you feel make polls wrong, such as the shift from landlines to cell phones.

Anyway, polls can be expected to be reasonable predictors of election outcomes, but just how good are they?

Looking at a number of races today, excluding only a few because there were no polls, I got the Real Clear Politics web site averages for polls across the states, transformed those numbers to get a percentage of the Sanders + Clinton vote that went to Clinton, and plotted that with the similarly transformed data from the actual primaries and caucuses. The r-squared value is 0.52443, which is not terrible, and the graphic shows that there is a clear correlation between the two numbers, though the spread is rather messy.

The ethnic status quo model outperformed polls

My model is actually many models, as mentioned. I have a separate regression model for each of several kinds of primary, including Closed Caucuses, Closed Primaries, Semi-Closed Primaries, and Open primaries. I did not create separate models for the much rarer Semi-Open Primary, Semi-Open Caucus or Open Caucus style contests, as each of these categories had only one or a few states. Rather, the model used to calculate values for these states is derived from all the data, so addressing specific quirkiness of each kind of contest is sacrificed for large sample size.

I also generated models that included White, Black, Hispanic, and Asian; each of these separately; and various combinations of them. As noted above, the best single predictor was Black. Hispanic and Asian were very poor predictors. White was OK but not as good as Black. But, combining all the variables worked best. That is not what usually happens when throwing together variables. It is more like mixing water colors, you end up with muddy grayish brown most of the time. But this worked because, I think, diversity matters but in different ways when it comes in different flavors.

When the total data set was analyzed with the all-ethnicity model, that worked well. But when the major categories of contest type was analyzed with the all-ethnicity model, some of the data really popped, producing some very nice r-squared values. Closed caucuses can not be predicted well at all (r-squared = 0.2577) while Open Caucuses perform very well (over 0.90, but there are only a few). The most helpful and useful results, though, were for the closed primary, open primary, and Semi=closed primary, which had R-squared values of 0.69, 0.61, and 0.74, respectively.

What this means is that the percentage of the major ethnic groups across states, which varies, explains between about 61 and 74% of the variation in what percentage of voters or caucusers chose Clinton vs. Sanders.

Polls did not do as well, “explaining” only about half the variation.

So, the following graph is based on all that. This is a composite of the several different models (same basic model recalculate separately for some of the major categories of contest), using nominal ethnic categories. The model retrodicts, in this case, the percentage of the vote that would be given to Clinton across races. Notice that this works very well. The few outliers both above and below the line are mainly caucuses, but the are also mainly smaller states, which may be a factor.

Who will win the California, New Jersey, Montana, New Mexico, North Dakota, South Dakota, and D.C. primaries?

Clinton will win the California, New Jersey, New Mexico and D.C. Primaries. Sanders will win the Montana, North Dakota, and South Dakota primaries. According to this model.

The distribution of votes and delegates will be as shown here:

This will leave Sanders 576 pledged delegates short of a lock on the convention, and Clinton 212 pledged delegates short of a lock on the convention. If Super Delegates do what Sanders has asked them to do, to respect the will of the voters in their own states, then the final count will be Sanders with 2131 delegates, and Clinton with 2560 delegates. Clinton would then have enough delegates to take the nomination on the first ballot.

In the end, Clinton will win the nomination on the first ballot, and she will win it with more delegates than Obama did in 2008, most likely.

Every state is special, and some are more special than others. West Virginia has 29 pledged delegates, but not all of them were assigned today. I assume they will be assigned later. Thus, the slight difference in numbers between what I predicted and what happened.

A key message here is this. Clinton and Sanders did exactly as well in the West Virginia primary as my model predicted, and that has been very close to exactly true for most primary races all along.

There is something important about this likely win that I want to point out.

According to many (but not all) of those pushing the candidates on climate change, Sanders is THE man when it comes to climate, and Clinton will be throwing the planet under the bus the moment she is elected. The degree of contrast between Sanders and Clinton in the minds of many leading climate activists is an overstatement, and highly inaccurate. Both candidates have stated that we need to come to the point where we keep the fossil Carbon in the ground, but Sanders is the only candidate who has come out with a weakened version of this, where we only try to attain 80% non-fossil fuel use by the middle of the century. Sanders wants to make fracking illegal, which he is unlikely to be able to do, while Clinton wants to regulate it into near nothingness, which could be achieved in a year or so. On the other hand, Clinton did say to a coal miner the other day that we have to find a way to keep mining coal as long as it does not add CO2 to the atmosphere. Sure, let’s do that! But first, we’ll have to change physics, because we GET energy from releasing fossil Carbon, but we have to USE energy to re-attach the Carbon to something solid.

In other words, neither candidate is where they need to be on climate change, and we will have to work to make that situation change. But, if you ask most people, they will probably tell you that Sanders is the climate change guy, and Clinton not so much. But we also know that addressing Climate Change means, essentially, shutting down the main industry in West Virginia, which is coal mining.

So why is Sanders beating Clinton in West Virginia?

I’ll put the results of the primary below later this evening or tomorrow morning. I’ll be busy during the time the returns are coming in, visiting with my friend Emo Phillips, but I’ll update the post at a later time.