Here is a graph showing polling for Bernie Sanders and Joe Biden. See below for some important details.
The numbers used for this graph come from 38 national polls asking for voter preference about a varying number of candidates. There is a large variation across the polls in how many answered something other than a particular candidate (like “none”). These two factors cause useless and distracting variation in the actual percentage value given to a candidate for a given poll. You can imagine that if a certain candidate gets 23% of the “votes” in a given poll, that number could change a lot if non-answers were excluded, or the total number of candidates was different. An imperfect but still improved way to calculate the percent value for a given candidate is, then, to only look at a subset of the candidates across all the polls, and recalculate the percentage of polling for each candidate using only those numbers. That is what this graph shows, for these candidates only:
Why that particular list? Well, I noticed that if you look across all the polls, one minor candidate (minor in terms of percent in the collection of polls) seemed to vary from the middle of the middle tier to the bottom of the middle tier, but was never in the lowest lowest tier, and also, was polled from early on: Klobuchar. So, I took the RCP average at about the time of the debates, and applied the Klobuchar Factor. If you were below Klobuchar, you were out of consideration. Since then, the candidates have moved around a bit, and a present day Klobuchar Factor would produce a different list. But I don’t really care, because I just needed to have a cutoff somewhere.
The regression analysis suggests that about 56% of the variance seen in each canidates’ polls is explained by time (i.e., there is a pretty robust trend where time matters). I’ve extended the regression line out 20 days into the future, which would be the end of July.
So, getting back to the story of these two candidates. I want to consider each candidate separately. The reason they are both in the same graph, and blog post, is because they are the two candidates with the highest number across the entire data set, so the graph makes sense for their scale, and the process is cleaner of we separate out candidates by scale.
The story of Joe Biden is this: He started off high, around 50%, and ended up much weaker, closer to 30% with some of the most recent polls showing 25%. He halved, almost. Or at least, looking at the extended projection, he is in the process of measuring out his polling half-life, as it were. He was probably artificially high partly due to name recognition, and lost ground as other candidates gained. He also started out in a different sort of artificial high, as a well known and widely loved guy where policy had not been vetted, and has lost among Democrats in that way as well. But this is Biden, and this is how he has performed in his earlier presidential campaigns. Biden watchers are not surprised. Biden watchers will not be surprised if he isn’t really a factor in this campaign by the end of the year.
The story of Bernie Sanders is interesting. His numbers show the second lowest amount of variance, scaled by magnitude, of all the candidates. He started of around 20%. He is still around 20%. Bernie is not moving up, Bernie is not moving down. Well, maybe a tiny bit down. What he seems to be doing, really, is slowing down just a bit as Elizabeth Warren is passing him, much like a car going 45mph slows down a bit when a faster car is passing them on the highway. Though that is of course a bad analogy because the intentionality of events is very different.
In short, Biden is gliding to a campaign ending landing, while Sanders is flat-lining. The latter observation is, I think, the most significant. It tells us something, maybe, about Sanders campaign. His base is unmoving. This is expected, I think. I just hope that should Sanders not get the nomination nod, that base sees fit to support the nominee in 2020, all of them, different than what happened in 2016.
… if you are doing what a lot of people are doing on the Internet. Being wrong!!!!
The Russian organized and operated trolls that will attempt to ruin the 2020 election will sow divisions among Democrats so that the process of selecting the best candidate to go up against Trump will be so badly damaged that they can’t win.
How will they do this? By declaring particular candidates as not electable. By declaring that this or that candidate’s positions are entirely different than they actually are, in a way that makes potential supporters turn away. By causing friction among those who are otherwise allies or friends so that social networking communities are ripped asunder, and so on.
The thing is, most apparent Trumpo-Russian trolls are not actually Trumpo-Russian trolls. Rather, they are you, or others like you, who have fallen into this pattern. Time will tell if this pattern has been promulgated in small or large part as an arm of the Russian attack on our democracy, or if people are just acting this way because it is human nature. But it does not matter. Employing these and similar tactics in our public conversation about our candidates looks and works the same, and has the same effect, whether the act is bought and paid for by the Republican-Trump-Putin axis, or whether it happens all by itself.
Don’t be confused for a Putin Troll. Being like a Putin Troll is the same exact thing as actually being a Russian troll.
All the bad things people say
You can’t fairly judge a candidate based on what people on Facebook or Twitter tell you. Such comments are more often than not inaccurate, often purposefully so.
Claim: Candidate X thinks America is not ready for healthcare for all! Next!
Truth: Candidate X makes a clear statement that we need universal single payer healthcare. The same candidate then lists several possible steps to get there.
Claim: Candidate X is the only candidate that can beat Trump.
Truth: Most people can’t even name most of the candidates, and there has not been a single debate. There are candidates that haven’t even declared yet. There is simply no way to say who can beat whom. As a matter of fact, there is a pretty darn good chance Trump isn’t going to be the guy to beat anyway. He’ll be pushed out or removed or in some other way unavailable.
Please consider this strategy:
The election is so early that not all the candidates have even declared,and most are in fact unknown with respect to position or abilities, regardless of what you may think. So:
1) Wait to declare a candidate you prefer the best. If you like one candidate above the others, do go ahead and say nice things about that individual, but please do not write off the other candidates or attack people who have a different opinion.
2) Wait to write off individual candidates that you really don’t like. There is nothing wrong with having such an opinion, but for now, please do what your mother tried to teach you: If you have nothing good to say about someone, keep your stupid mouth shut for now (I’m sure she was thinking it that way, though she may have used other words).
3) Don’t repeat the trollish comments you hear. They are not hard to identify. A very smart and thoughtful friend of mine did this recently, the first example above is based on that. A candidate was attacked by a troll on twitter. The attack was very inaccruate. My friend simply repeated the attack. Don’t do that, makes you look like an idiot, and it amplifies the trollish message.
4) Don’t BELIEVE the trollish comments you hear. In the case mentioned above in Number 3, virtually no one seems to have responded to the recycled attack by questioning it. Make up your own damn mind with facts you have obtained from good sources and verified. It isn’t that hard. It is your responsibility, your job, to do this.
5) Remember where we are. We are at present BEFORE the beginning. This is not the time to weed out candidates. Take your time. Remember, there is a Democratic debate (probably two) in June. Wait until at least the debate to start weeding out candidates, and even then, be fucking civilized about it, not trollish. Please.
6) Please make the distinction between the process of selecting a nominee and running for president. There are important differences at many levels. A full third, in my estimation, of the embarrassingly stupid things people said during the 2018 race came out of ignorance of the difference.
7) Part of your message, your public opinion, should always be how you will support the nominee no matter what. Note that you can’t really say that now if you also say “I will never vote for Candidate X no matter what.” So stop saying the latter, always include the former. As part of this, please do not let the perfect stand in the way of the pretty darn good.
8) Do not complain about the system of selecting a nominee unless you are willing to spend at least a little time helping to select the nominee other than just showing up like a drone on Primary day. Stand up and do something. You are needed.
Since 1968, about 17 candidates ran in Democratic primary races and earned enough votes (above about 20% all told) to count as having been contenders.
Of those, one was murdered, one was shot but lived, one was eliminated from competition by GOP dirty tricks, and one left the race because of insufficient support but would probably have been exposed as having two families (that would have been a scandal) had he stayed in the race.
Putting this another way, there is about a 24% chance that a Democrat running in a primary will be taken out of the race for extrinsic reasons.
Given the stakes, i.e., becoming the most powerful person of the 7 billion on Earth, one would probably stay in the race if one is in second place.
I should note that the gunning down of candidates has not happened in a while, and those early events caused a significant increase in security. Dirty tricks are still a possibility, and we may have seen that in this year’s race, but if so, they were against Clinton, not Sanders. Scandalous behavior wiping out a candidate is unlikely this year as well. Clinton has been more heavily vetted than any candidate in history, and unless Sanders’ tax returns turn out to actually be interesting (we’ll probably never know), he seems fairly scandal free.
But, the odds is the odds, and since the modern system of primaries emerged, which could be dated to 1968, your opponent has only a 76% chance of survival even if you do nothing.
There was a lot of talk about California, and a lot of back and forth, but in the end I stuck with my original model to predict the outcome of that race. See the table above for the results, but the bottom line is that I predicted that Clinton would get 57 percent of the votes and Sanders 43 percent. It turns out that Clinton got 57 percent and Sanders got 43 percent.
Excuse me for a moment while I bask in the bright light of being-right-ness.
Thank you. Now, on to the details.
First, a quick, note on the numbers and methods. All my percents (for prediction and as reported for the outcome) are the proportions of each candidate’s take of the two candidates, so “other” or “The Lizard People” or anything other than Clinton or Sanders are taken out of consideration. In some cases this will cause the numbers to look different than those reported by the press. The awarded delegates I provide here are from the Washington Post, and often do not add up to my predicted proportionate amount. This is because the process of awarding delegates is complicated and bizarre. Eventually the numbers of proportionate delegates will settle to be very close to those you would get form using the percentage of votes for each candidates.
The outcome of yesterday’s primaries was pretty much as expected, but not exactly. Polls and my model both seemed to predict that Clinton would win New Jersey by a large margin, California by a good amount, likely New Mexico, and that Sanders would take Montana and the Dakotas.
Clinton ended up doing better in New Jersey than expected, but in the case of landslides, the final numbers are often a bit off probably because of some fundamental behavior of variance. California was as expected, as was Montana. Sanders did much better in New Mexico (a closed primary, by the way) than expected, but still did not win.
The Dakotas are the enigma. The expectation was that Sanders would do very well in both states, better in South than North. It turns out that South Dakota totally reversed, with Clinton winning by four percent. In North Dakota, Sanders wiped Clinton out, not only winning by a large amount as expected, but trouncing clinton with what must be one of the highest margins all season.
With respect to my model (detailed here), I think we are looking at sample size and a few other things. I was within a fraction of a percent in the largest state, and the smallest states were the oddest. But, I also suspect different campaign efforts by the different candidates played a role. Also, when we talk about openness of the primary (or caucus) it is important to note that not all contests have corresponding Republican contests going on at the same time. That may be a big factor in the Dakotas.
In the end, there are two big winners today. Hillary Clinton had a resounding victory in the largest state, and did very well across the board otherwise. This comes hours after the press deciding to declare her the Winner-Apparent based on math, and it verifies that math. Sanders has continuously said he would fight to the convention, attempting to overthrow the process using super delegates. He seems to have not noticed that the entire Democratic Party is mad at him, even former Sanders supporters, and the super delegates’ job is actually to make an effort to maintain the spirit of the process when something goes wrong. Sanders is the thing that is going wrong at the moment — with his effort to reverse the democratic process — so there is zero chance that the Supers are going to come to his aid.
The second winner is, of course, Science by Spreadsheet. I’ve been running spreadsheets on elections since spreadsheets were invented, and this is the best cycle I’ve had. I’m pretty sure my model out performed all the other models. Perhaps I will summarize all that in another post at some point.
Can’t wait to get started on the electoral map.
I should mention that DC still has a primary to go, and it will go overwhelmingly for Clinton.
First, Sanders lost Nevada because Hillary Clinton won the caucus.
Then, the Sanders campaign put their ground game into effect, in an effort to overtake Clinton during the nearly-unique-to-Nevada process that allows for changes in pledged delegates at later caucuses. But he didn’t get enough delegates to achieve that. The Sanders campaign does get credit for getting more delegates than they had before, of course.
Then, at the State Convention, Sanders had enough delegates in place to gain a couple of more delegates and possibly tie with Clinton in the end. But the organizers for the Sanders campaign failed to ensure that all the delegates to that convention were totally on board with what they needed to do in order to be credentialed at the event. Of the thousands of delegates at the convention, a handful of Clinton delegates failed to be credentialed (an expected number) but something over 60, initially, of the Sanders delegates were not legit, so they could not participate. A few of those managed to get credentialed by clarifying their information, but most did not. I’m not 100% certain of this, but I think that had they all been credentialed there would have been enough Sanders delegates to win one more delegate.
The fourth failure is complex, and mainly philosophical. First, Sanders supporters around the country are complaining a lot about how the Democratic Party process is an insiders game and ignored the will of the people. This is odd, considering that the will of the people leans strongly towards Clinton. In any event, the Sanders campaign playing the ground game in the middle of the Nevada process was inside politics. This sort of inside politics is perfectly normal, legal, expected, and what you have to do if you want to win. But by complaining loudly about the Clinton campaign doing this sort of thing, and then doing this, Sanders lost a moral high ground. The fact that this particular moral high ground does not exist to begin with means that this is merely an annoyance, but it is annoying.
Another part of the fourth failure is the cacophony of Sanders voices complaining about getting screwed in Las Vegas (I’m sure they weren’t the only ones that evening). This is a problem because it engenders bad feelings among democrats, but the accusation is based on nothing. What really happened is that the Sanders campaign tried to grab a couple of more delegates, but owing, I think, to too many people involved being ignorant of how the system works, failing to do as well as they might have. This same ignorance has led to unfounded complaints about what happened at the Nevada convention. This whole thing, this fourth way of losing, has given people whoa are getting tired of the Sanders followers more of a reason to call for Sanders to drop out of a race he has already lost. That loss may be more important than the small number of delegates that the Nevada Sanders campaign organizers failed to get.
A few points for those not fully aware. First, the numbers of delegates at the convention is very large, and the number of delegates who were not credentialed is very small, a fraction of a percent. Second, yes, this convention was chaotic, but guess what: they are all, always, chaotic. What happened at this convention was mostly pretty normal, though the Sanders antics did make the event run way more over time than usual. Another thing that was a bit unusual was that they were allowed to go over time by three hours. That is fairly remarkable. Usually, a extended event (caucus or campaign) shuts down much sooner.
Both campaigns had people involved in counting delegates, credentialing delegates, and running the meeting. This was not a case of Sanders people all on the floor and Clinton people running the show. Rather, both Clinton and Sanders people were involved in all aspects of this convention, and both Sanders and Clinton people, for the most part, acted properly during the event. I think it was just a small number of Sanders people who were causing all the extra raucous, and later complaining about it.
When considering the events in Nevada, remember that no two caucuse systems are alike, and the Nevada system is probably much less like the others than any. People are calling for a revision of the Nevada system, to not allow so much changing around of delegate pledges after the initial causus (though that has nothing to do with what happened Saturday), but actually, this system is better for the campaign process and for democracy. First, candidates have to demonstrate that they are willing to do more than just show up for a few days of stumping and buy a few ads. They have to be involved at the state and local level the whole time. Second, if there is a shift in the opinion of party mebmers as to who should get the state’s delegates, this allows for that adjustment. In this case, the adjustment mainly indicated a shift towards Clinton and away from Sanders. Thus, the Sanders people are a bit upset. Understandable, but just part of the process.
By and large, a lot of Democrats (both Sanders and Clinton supporters) are deciding to love or hate a the process, or a particular part of the process, based on whether their candidate won or lost. Please stop. In fact, estopp. You signed up for the game, this is the game, these are the rules. Feel free to suggest changes in the rules, but you can’t issue a complaint when the rules are followed but you didn’t get your way.
Between now and the end of the primary season, I expect Sanders to pick up more delegates than Clinton, in total, by a very small margin.
On Tuesday, April 26th, there will be primaries in Connecticut, Delaware, Maryland, Pennsylvania and Rhode Island. That’s 384 pledged delegates at stake.
Polls put Clinton ahead in all these states, but not all the polls are current and not all the Clinton leads are strong.
I noticed some very strong reactions in the comments section from people apparently (but not very clearly) accusing me of making up numbers to make it look like Sanders will win some races (esp California?), with the presumption that I’m a Sanders supporter.
Those of you who have been following my writings on the campaign will know that for the first several weeks of the primary season, until very recently, I did not support one or the other candidate. I like them both. And, if you like either of them, and you know anything at all about American politics, you’ll like the other as well, though of course you are entitled to have a strong preference. Either way, it is impossible to like one of these two candidates and not prefer the other over either Trump or Cruz (or any Republican who ran this year). If you do like any of those Republicans over one or the other Democratic candidate, please note that most people looking at you will be thinking “WTF”?
Anyway, the analysis I use to make these predictions is something that I have been developing and refining since the very first days of the primary season, and it is a dispassionate unbiased statistical prediction, and has nothing whatsoever to do with which candidate I support.
If you are making an assumption that I support, say, Sanders, and that is why I wrote this post, then I’m pretty sure that you’ve not read the post. Why do I say that? Read the whole post and find out!
My model, as you know, has been doing a pretty good job at predicting outcomes in this year’s Democratic primary process. And, that poll says that Clinton will win three states, Delaware, Maryland, and Pennsylvania, and garner a total of about 221 delegates, and Sanders will win two states, Connecticut and Rhode Island, getting a total of 163 delegates.
Note that my expected spread in Pennsylvania is actually very close. Clinton is firmly ahead in the polls, my model says she’ll squeak by, and my model has done better than polls in many instances. Who knows, maybe Sanders will win there?
Delaware and Rhode Island are really close, and could go either way. On a related note, there is supposed to be a new poll for Rhode Island coming any second now (there is no current polling there) so that will be interesting.
The table at the top of the post shows my projections for Tuesday as well as through the rest of the race. Note that starting Tuesday and running to the end of the primary season, Bernie Sanders is expected to get more delegates than Hillary Clinton, but only 10 more. This a very small number, and the final count could go either way. It would, of course, take Sanders winning a much larger number to catch up to Clinton in pledged delegates. Sanders is behind by 237 delegates.
In order for Sanders to close the gap with Clinton, he would have to do 17% better than my model projects from here on out.
That does not sound like a lot, but there are two things to consider. First, my model has been very accurate. It has been closer to a few percent off over time, and I don’t expect it to suddenly stop working at that point. Second, to the extent that my model is wrong, it tends to under predict Sanders in caucuses and open states, esp. open caucuses. All the remaining contests are primaries, and most of them are closed or semi-closed.
Note also that my model conflicts with the polls and common knowledge in California, where I say Sanders will win, and everybody else (except Sanders, I assume) says Clinton will win. Also, note how some of these contests are very close, really too close to call especially Indiana, and Kentucky.
I don’t know yet, but as soon as I do, I’ll post that below.
With 98.5% of the delegates counted, Clinton won 57.9% of the vote, Sanders 42.1%. This puts Clinton at 139 delegates, very close to my prediction of 137.
If that holds, this is pretty much of a shellacking for Sanders. Sanders out spent Clinton on ad buys, has campaigned heavily, and has set the expectations as a definitive win. This is Sanders home state (of birth, not representation). Yet he seems to have definitively lost. This will put Sanders even more behind in the delegate count.
The bigger the loss for Sanders, the bigger the steaming pile of bull substance will be put forth by that presumably-tiny-and-hillary-has-them-too-yadayada Sanders supporters, with claims that the election was unfair, stolen, etc. And that will probably turn off even more people undecided between the candidate, and Sanders will do even worse in future contests. I’ve predicted that he will win in California, and I’ll stick to that story until my own analysis suggests otherwise, but it won’t be enough to offset his current deep-second position, Clinton’s increasing lead, and all that.
You know what they say. It ain’t over until the big green lady with the torch sings.
And she just did.
ORIGINAL POST FULL OF INSIGHT AND WONDER:
Meanwhile, some background on a key aspect of today’s Democratic primary in New York.
This is a closed primary in a state where you have to be registered in a political party by some time in October in order to participate.
But, is this what will happen? Clinton does better than Sanders in southern states, and New York is not a southern state. In fact, Clinton tends to win all of the southern states, and while Sanders wins more non-southern states than Clinton, he certainly does not win all of them. See this for more details on the southern effect.
New York is a big state, and Clinton tends to do better in big states, as shown here.
There have been nine closed caucus states, and Sanders has won seven of them, with a tie in one. There have been three open caucus contests, and Sanders has won in three of them. There have been 7 closed primary contests and Sanders has won in one of them, with a tie in one. There have been 13 open primaries, and Sanders has won three. So, he does better in caucus states, but tends to lose in primary states, and possibly least well in closed primary states, which is what New York has.
Now, here is the interesting thing, recently pointed out by Rachel Maddow. Sanders, and the Sanders campaign, is not making any attempt at all to control expectations in New York. Clinton has a better claim to favorite daughter status. New York is relatively diverse, and Clinton does better in diverse states. Clinton tends to win closed primaries. The polls show Clinton ahead. My own projection, not based on polls, has Clinton winning. But Sanders keeps up with the “we will win here” mantra, which is not the advisable approach if you are not going to win. You can win and lose at the same time by setting up the expectation that you will lose by, say, 15% and then you go ahead and lose by only 9%. That’s a win(ish) in the primary process. But Sanders is not doing that.
Here’s Maddow’s thing:
Sanders’ evidence is that he tends to come from behind, and over perform. And in my own modeling, that has tended to happen. All those times I was right about the outcome of a contest and the great FiveThirtyEight was wrong, it was a Sanders over performance, pretty much.
But, for all the reasons stated above, I don’t expect this to happen in New York. If it does, that will be very significant, and we may have to rethink the whole primary process this year!
Anyway, just for fun and because I thought you might find it interesting, I rand some numbers. I simply took the last several polls in several states, and recalculated the percentage for Clinton and Sanders such that the percents attributed add up to 100%, and then added to the top of the list the actual performance in the contest. From this I made a graph, with the moment of winning on the left side. If Sanders tends to jump up and win the contest, this will be seen by a line tracking (backwards) along below 50% then suddenly, for the actual voting, jump above 50%.
I did not do this for all the contests because there simply isn’t enough poling data. Indeed, Sanders tends to win in open and/or caucus states, and pollster don’t even bother polling in those states because they are so crazy. And, he tends to win in small states, and pollsters tend to not poll in small states. Which, if you think about it, should give you pause in considering Sanders’ claim. He does better than expected when the expectation is based on nearly zero or otherwise crappy data.
Anyway, I non-systematically picked a bunch of states and made a bunch of graphs and shoved them all onto one graphic:
Sanders did the Bernie Blast to the top in Minnesota, but we had almost no info in Minnesota. He seemed to do it in Michigan, but if you look at the polling over time, it was not utterly unexpected. He might have done it a little, but not enough to win, in Arizona, but note that this is the state that had all that voter repression.
I indicated on each graph the nominal category of contest so you can gaze at these results and draw your own conclusions.
As I’ve said numerous times, each contest is a test of a particular hypothesis or model about how the primary season is going. If Clinton wins New York and wins it by about 10-15%, the NY primary does not change the fact that she will win the race, but come in just under the required number of pledged delegates to lock without super delegates. If she does way better, that changes our expectations for the rest of the primary season.
If, on the other hand, Sanders wins, that will be huge and require a major revision of our thinking.
Polls close at 9:00 PM Eastern in New York. If urban and NYC districts are counted early, and upstate later, because they use clay pots and send the results in by pony or something, then we should see Clinton surge then Sanders slowly slog towards catching up (or not).
As you know, I’ve been applying a model to predict the outcome of each of the Democratic Primary contests, and have done pretty well at predicting results.
All of the future contests are primaries, not caucuses. It turns out that the two modes have very different patterns. Many have suggested that this has to do with how the process works, and somehow caucuses, or open contests, favor Sanders, who has won several. However, it also turns out that caucusing is a northern thing (and Sanders does somewhat better in the north, or more accurately perhaps, rarely wins in the south). Caucusing is also a white thing, apparently. Caucuses happen in non-southern mostly white states, and these are states that Sanders can (but does not always) win.
Since the remainder of the contests are primaries, I used my simple ethnic-based model, which predicts the outcome of the various contests based on the estimated percentage of African American voters. I used only data from previous primaries to develop a simple linear model. This model applied to all of the future contests, starting with New York, tells us that Clinton will win in New York.
After that, Sanders wins in several smaller and mostly norther states, but also, California . Clinton wins in Pennsylvania and New Jersey, which are relatively large. If this plays out as predicted, between now and the end of the primary season, Hillary Clinton will pick up about 795 delegates and Sanders will pick up about 778 delegates.
How many delegates does each candidate have so far? Clinton has approximately 1310 and Sanders approximately 1094. (This is approximate because in some states it is actually a little hard to count because of the nature of the system.)
Here is a table showing all of my projections from here on out. I’ll probably redo the model a few more times, especially if anything unexpected happens, so stay tuned.
A while back Vox produced a tax modeler that would tell you how your taxes would change with Sanders plan. It raised most people’s taxes by a few thousand dollars. But the modeling was misleading because the same plan would probably reduce health care costs for those same individuals.
I pointed that out back at the time but most of the response to me pointed out was the ridiculous recitation of completely wrong information (from both sides of the debate) so I dropped it because it really doesn’t matter. President Sanders or President Clinton would not produce any tax plans. Not their job.
Anyway, The Nation, which has gone all in with Bernie as I’m sure it should, came out with a counter modeler for taxes. so the Vox piece is here, The Nation’s piece is here.
I predicted who would win the Wisconsin primary, although my prediction suggested that Sanders would do better than he did. (He underperformed.) I predicted the outcome of the Wyoming primary exactly.
These are the most recent two in a long series of mostly correct predictions of which Democratic candidate will win each of the contests in this long presidential primary season. My predictions of which candidate would win have been mostly accurate, but also, fairly accurate with respect to how many delegates each candidate would pick up.
Several primaries back, for several primaries in a row, Sanders did somewhat better than my predictions suggested, indicating that the model I was using to make these predictions possibly underestimated that candidate’s long term performance. However, that stopped happening, and Sanders went back to performing pretty much as I expected him to perform, or not as well.
This verifies the fact that Hillary Clinton will finish this presidential primary season in the lead. Yes of course, one never knows. But at some point one has to presume, even if there is a small chance that a numerically nearly impossible outcome will emerge. And, if this turns out to be wrong, since I am tracking every delegate, I’ll be among the first to know and acknowledge, and shift strategy as needed. But at the moment I feel very comfortable working with the assumption that the primary season will end with Clinton having about 2,000 pledged delegates, and Sanders having between 1700 and 1800 pledged delegates.
If the unpledged delegates simply track this outcome, this will give Clinton the nomination on the first ballot.
I have been using a similar model for making these predictions all along, but refining the model (how it works) and adding data (with each contest’s outcome). I have tried several times to develop a version of the model that would put the consistently second place candidate, Sanders, at an advantage, biasing the model with assumptions about a possible improved performance. Not once did that alternative version of my predictive model put Sanders in the lead at the end of the primary season, though he got close once.
Simply put, barring an unlikely surprise, Hillary Clinton will be the Democratic Party’s nominee for president in 2016. That is the important thing.
Now the troubling thing.
From the start of this primary season, I was happy with either candidate, and vowed to support whichever candidate is nominated. Most sane people intend to support the winner, because the alternative is rather horrible. Most people did pick a candidate earlier in the process, but I refused to. Of course, every time I questioned a criticism of Clinton, some Sanders supporters “accused” me (as though that was a legitimate accusation of wrong doing) of supporting Clinton. When I would critique a criticism of Sanders, the reverse would generally happen. Many people were simply not allowing me to be supportive of both. Also, I wasn’t undecided. I had decided that both were excellent candidates, in their own ways.
But it goes beyond that. During this primary season, I’ve witnessed, again and again, people who had previously shown signs of high level functioning and impressive intelligence saying many utterly stupid things. I’ve closely monitored and been involved in many presidential elections, and I note that this often happens to some degree, but this year, this has been happening wholesale and to an extreme. I will not give you examples. If you are a reasonable person who has been paying attention, you don’t need me to give you examples because you know exactly what I am talking about. If you are one of the folks who has been quick to make utterly illogical or fact free arguments about every aspect of this race, often reaching far into the land of conspiracy theory, then you don’t know what I’m talking about but you will sense, somehow, that this paragraph is deeply insulting to you. Feel free to make defensive comments below. I will ignore them. And, I have nothing else to say about this. This departure from reason is, of course, the troubling thing.
The dangerous thing overlaps with the troubling thing.
Weeks ago it started to look like a small number of supporters of Bernie Sanders, in the event that Hillary Clinton was nominated, were going to either write in Sanders, vote for Trump or Cruz, or not vote at all. This did not surprise me, because a good number of the Sanders supporters where I live, in the shadow of Michele Bachmann’s congressional district, are fairly right wing. This may not make sense if you see Sanders as a progressive, very left candidate (which he is) but have a non-nuanced view of politics, but it is both true and understandable. The same thing happened with the Paul mini-Dynasty. I will not spend any time here outlining how this happens.
Over time, however, this “small percentage” has grown, and polling indicates that something close to 20% of declared Sanders supporters are what has become known as “Bernie or Busters” of various political stripes, but all holding the same dangerous view. These folks will not support anyone but Sanders, or will turn on the Democratic party if Sanders is not nominated.
Parallel to this phenomenon we see myriad other destructive practices by Sanders supporters, and by destructive I mean destructive to the political process and to the Democratic candidacy. Given that Clinton is going to get the nomination, it is a significant problem that so many Sander supporters are trying so hard to damage her.
These trends, of “Bernie or Busters” or of taking Clinton as seemingly equivalent to Satan, are a problem not only because of their immediate effects, but because the Sanders campaign accepts and exploits these activities and attitudes. It is no longer possible to point to the two or three times that Bernie Sanders scolded someone for this attitude and claim he is taking care of this, and it is no longer possible to give the Sanders campaign the benefit of doubt, suggesting that they just don’t know about what is going on. Campaigns know these things. Sanders knows about these things.
To this we add the clearly emerging pattern of the Clinton campaign working down ballot, to elect a blue, or at least, bluer Congress (and to help Democrats in other ways), while Sanders does very little in this area (he has done some things, but not much). Sanders’ strategy of having the masses show up in DC to shame the GOP Congress into not being nefarious haters was never going to work. Clinton’s strategy, and the strategy of the rest of the Democratic Party, to take back Congress, can work if we follow through. The numbers show that we actually could do it this year, if we don’t throw away the opportunity. Sanders appears to be throwing away the opportunity, Clinton is not.
So that’s the dangerous part. We need to approach the general election with a candidate and supporters who are going to do what is needed. The Sanders campaign has become a danger.
Several days ago I posted this on my Facebook page:
Among the reactions to this meme were assertions that somehow it is wrong for candidates to help each other (see comments above about taking back Congress .. Franken’s election is exactly how the Democrats retook the Senate for a couple of years). Among the reactions was a call to find someone to primary Franken. These are insane reactions. These are the reactions of deluded cultists, not political activists.
And, these reactions were among the small number of final straws that had fallen upon this particular camel’s back. I decided to take a break from the Facebook conversation about this election for a few days, and I blacked out my profile pics, without comment, as a form of protest. To underscore the protest, I began posting nothing but cat pictures. A handful of my Facebook friends understood and commiserated. A good number of Sanders supporters seemed to quiet down (except one or two), probably realizing that I was fed up.
And now, I’m back. But guess what. I’m not going to argue about Sanders, or Sanders vs. Clinton. The Sanders campaign is done. If this had all gone somewhat differently, I’d still be talking about Sanders, points he’s making, interesting things about his campaign, but the cost of doing that is too high. The Sanders campaign, owing mainly to the personality cultists and the Bernie or Busters, which are probably in total about a third of his supporters, have ruined the campaign, and made it not worth talking about. The Sander campaign, sadly, has become less interesting, more annoying, and just as predictable, as a bunch of cat pictures.
This is not to say that Sanders contribution has not been great. It has been very significant, and this souring of his campaign detracts from that only modestly. But that part is done. We’ve heard Bernie, we’ve listened, he’s influenced the process.
But from hear on out, if you are a Bernie supporter, talk to the paw.
We are in the Primary Doldrums. For the last several days and the next several days, there is not too much happening, big gaps between the action. Wisconsin is important, and it is Tuesday, Then Wyoming by itself, then New York by itself, then a sort of Super Tuesday with several states.
As you know I’ve created a multivariable model that has a good record of predicting primary and caucus outcomes in the contests between Hillary Clinton and Bernie Sanders. For the rest of the primary season, this is what it looks like.
I used yellow highlighting to indicate who is expected to win the most delegates on each primary/caucus day. Sanders will do well in Wisconsin, tie (or maybe even better) in Wyoming, do well in Indiana, and on balance, do well on June 7th when there will be six contests at once including Pennsylvania. But while Sanders may win the day on three (or four) days, Clinton will win the day on five. In total, Clinton is predicted to take 886 delegates, and Sanders 790.
This is the distribution of cumulative delegates starting with now and moving across this range of primary dates, showing the evolution of the difference between the two candidates throughout.
On balance, Clinton will, according to this model, will widen her lead over Sanders. If Sanders does better than projected this gap will narrow, but he’ll have to do very well to close the gap.
But there is some controversy over what kind of bird it is.
There is no doubt whatsoever that the bird was a finch. However, what kind? Most likely a house finch, because they are common, and the most likely to live in a big auditorium thingie and not be fearful of people.
Bernie Sanders has either stated or implied two features that make up his strategy to win the Democratic nomination to be the party’s candidate for President this November.
Implied, sort of stated: Convince so-called “Superdelegates” (properly called “uncommitted delegates”) in states where he has won to vote for him, even if he is in second. That is a good idea, and if the two candidates are close, it could happen. However, when I run the numbers, giving Bernie “his” uncommitted delegates and Hillary “her” uncommitted delegates, it is pretty much a wash. The uncommitted delegates are not perfectly evenly distributed across the various voting units (states and such) but they are evenly enough distributed that not much happens. Not that this can’t come into play when Spooky Delegate Math is applied, but there isn’t much there.
Stated, the other part of the strategy: Get more votes. The idea here is that the second half of the primary season (counted in terms of numbers of delegates awarded over time), which started on March 22nd, is more favorable to Sanders than it is to Clinton.
Earlier work I did showed that this strategy has only a small chance of working, because Clinton will in fact win plenty of delegates during this second half of the season, and she has plenty of delegates under her belt now. Bernie just can’t catch up. See this post for details.
As I have said many times, each primary or caucus, or each day on which there are a number of contests at once, is a test of one or more hypotheses. One hypothesis at stake last Tuesday was the accuracy of the model noted above. The various iterations and updates of my models for predicting primary outcomes have been very accurate all season, and I accurately predicted the outcome of Tuesday’s primary in terms of wins. I predicted that Hillary would win Arizona, and Bernie would win Idaho and Utah, and they did.
However, the magnitude of the predictions was off. Hillary won fewer votes than expected in Arizona and Bernie did way better in Utah and Idaho than predicted. (Also, the role of crossover voting was reduced as a likely factor in these elections, because Bernie did so well in Arizona with no crossovers.)
The difference in magnitude was so great that the seemingly assured Clinton victory in delegate count was turned on its head, and Sanders got more delegates than Clinton.
Is that a wakeup call? Or is it random variation?
Well, let’s assume for a minute that this is not random, and that this small set of contests tells us that the model is fundamentally wrong(ish). One thing I could do to fix that is to add the new data into the multivariable model and recalculate, but the number of new data points is insufficient to make a difference.
Another thing I could do is to assume that there is change over time in voting behavior, and add a variable for time. There are two reasons to not do that. One is that the more variables you add, the more accurately the model can predict the past (i.e, predict the value of the variables that are used to make the model), but not necessarily the future. The second reason is that if time is in fact a variable, simply adding it now would not work because of imbalance over time in sample size for the relevant variable.
So, what to do? Well, a third possibility is to fudge the data. Let us take a chance and provisionally assume that Arizona, Utah, and Idaho indicate that from here on in the expected outcomes based on my model are off by a certain amount, and then adjust future states to reflect that.
I quickly add that I’ve done this before … fudging the model to see if a Sanders claim about future outcomes might change the numbers … and each time that new hypothesis was falsified by subsequent primaries. But, why not try it again? The numbers from yesterday’s contests are startling enough to make it, actually, necessary, if one wants to remain honest about what is happening on the ground.
I have felt all along, and still feel, and most people agree with this, that there are two kinds of states, those that tend to favor Bernie and those that tend to favor Hillary. Also, the variables used in the multivariable analysis may have asymmetries across the nearly-even-state boundary of bias. (In fact I’m pretty sure they do.) So, let’s consider Arizona as a Clinton-favoring state in which she underperformed a certain amount that we estimate by comparing the expected results with the actual results. Let us also assume that Utah and Idaho are Sanders-favoring states in which he over performed by an amount that we can similarly estimate.
This is conservative because the estimates are based on the differences between the candidates, not the absolute magnitude of their delegate takes in each contest.
In this revision, then, I put Clinton’s expected future performance in Clinton favoring states as a 30% reduction in the spread, and Sanders’s expected future performance in Sanders favoring states as a 300% increase in spread. (Notice the asymmetry emerges here.)
Those sound like really different numbers, but they are not. The typical predicted Sanders win is small, so the total number of extra delegates Sanders ends up with is pretty similar in the two kinds of states.
When I do this, Clinton still wins. See the chart at the top of the post. But, there are three very important things to note.
First, this is too close to call. If this Sanders II strategy works out over the next few contests, and we believe it is the New Normal for this primary season, then it will simply be impossible to say who will win. The outcome here is very close, and had I used just slightly different numbers, I could have come up with an equally close outcome with Sanders winning.
Second, it is possible, depending on what happens with uncommitted delegates, that if the race is this close, there could be a brokered convention. I actually think this is unlikely, because in order to have that happen you probably need three or more candidates staying in it until the end, so a bunch of delegates are bound to vote for someone other than the two front runners. But, I’ve not looked at the numbers and the data and the rules closely enough to be sure. Consider it something to look into.
Third, the role of the big states now emerges as more important than it was before. The really big states, including New York, Pennsylvania, California, and New Jersey, were actually all very close in the model, and frankly, I can’t tell if they are Sanders vs. Clinton favored states. This is for a good reason. These states are so large that they are internally fairly diverse, and also, not easily affected by odd rules in the primary or caucus process the way some other states are. The apparent bimodality of states in general applies mainly to the smaller states.
Putting this another way, the larger the state, the closer to the national average response we see, and the national preference between the two candidates is similar. Smaller states stray away from the mean, larger states regress towards the mean. Like this:
So what does this mean? This means that larger states are not going to break strongly for either candidate. But, it also remains true that there are a lot of delegates in these states. So, this could mean that a strategy that effectively focuses on the big states, or one or two of them, could push that state over to one side or another.
I can make you this promise. Both campaigns are currently having this conversation and there will be intense campaigning in the big states. It is possible, maybe probable, that the candidates will watch each other doing this and end up differentiating, with the different states being focused on by different candidates. But, there are also states neither will give up. I suspect New York and California will be fought over heavily, while Clinton may give way to Sanders in Pennsylvania and Sanders may give way to Clinton in New Jersey.
The cycle over the last several weeks has been to see Sanders as possibly moving closer to Clinton, but then, failing to do so. But this week, he did. And, this is the first week in a series of contests where elements of the stated or implied Sanders strategy are supposed to come into play. And maybe they did. Or maybe not.
Frustratingly, the next several states are not going to be too informative. Washington is big, and Sanders will probably make big gains there. My main model, which I will continue to assume is the most accurate projection until proven otherwise, has Sanders getting ten more delegates there than Clinton. The revised Sanders II concept, in contrast, has him getting 30 more delegates than Clinton. That will be a test of the Sanders II hypothesis.
Then, eventually, comes New York, where we will see a test of the Too Big To Fail In State strategies. My model has Clinton winning in New York by just a few delegates, and the Sander II model says pretty much the same (remember, it is conservative, addressing only the gap). If New York is close to a draw, as predicted, then we will be left wondering. If Sanders takes 20 or more more delegates than Clinton in New York, then we will be left in wonderment.
Following that is Little Big Tuesday, with several small states and Pennsylvania. That should also be close to a draw, according to my primary model, with Clinton winning a few more delegates than Sanders. But the Sanders II model has Sanders winning not just a few more, but many more delegates.
According to the Sanders II model, at the end of the day on Tuesday, April 26h, after Pennsylvania, a ca 320 delegate lead by Clinton will be cut to a 190 delegate lead. According to the main model, the one I still trust until proven otherwise (perhaps over the next few weeks), the Clinton lead will still be over 300.
So, that’s my story, and I’m sticking to it. Both of them. For now.
This post was written in two parts, pre-primary and post-primary. To see the result and a discussion of what they mean, skip down to the last part of the post, where I’ll discuss why Tuesday’s results may mean that Sanders could win the primary.
As already discussed, Clinton is likely to win the Democratic nomination. Sanders is too far behind to catch up without extraordinary results, as outlined here. However, it is also true that Sanders is likely to win a majority of contests from here on out, while at the same time, Clinton is likely to win many (if not most?) of the actual delegates.
Here, I’ll review what my recently upgraded predictive model indicates for today’s primaries in Arizona, Idaho, and Utah. Also I’ll provide a list of states and delegate counts for the upcoming primaries (including today’s) that would have to be realized for Sanders to catch up to Clinton. At the end of the post, you’ll find results of today’s primaries, and some discussion, when available.
First, the expected outcome of today’s primaries based on this model:
Clinton is expected to win big in Arizona, while Sanders is expected to squeak by in Idaho and win handily in Utah. The total delegate count for the day would be 82:49, Clinton:Sanders, so if this model is accurate, Clinton will win the day. As I’ve noted before, this model tends to under-predict Sanders’ wins when he does win, so the delegate count could be closer.
Or, this could be totally wrong and Sanders does much better, which would require me to go back to the drawing board. Which, of course, I’ll do.
In order for Sanders to catch up to Clinton, he’ll have to do much better than he’s done, even given the fact that he is favored in a lot of upcoming states. If we take all the upcoming states together and simply give Sanders even wins across the states sufficient to tie Clinton on the last day of contests, then he’ll need to win Arizona 44:31, Idaho 14:9, and Utah 19:14.
Here’s a chart of the outcomes across all states for Sanders and Clinton to finish the primary season in a tie.
This is, of course, totally unrealistic. Sanders would likely do much better in some places, and just OK in others. But this chart serves as a basis of comparison for future races.
Every primary or caucus is a test of a hypothesis. The hypothesis that Clinton will do what I suggested she will do here is being tested by today’s contests. If Clinton gets somewhere around 75 to 89 delegates, the hypothesis is not rejected. If Sanders manages to perform much better than 49 delegates, say, over 62 or so, then the hypothesis has to be rejected (I’m not being formal here with rejection levels) and the possibility of him catching up has to be re-evaluated. If, of course, Sanders gets fewer than 40 or so delegates today, than he will have an even steeper uphill battle for the rest of the primary season.
I’ll add more information and commentary below after we get results!
OK, it is early the next morning and we have results, but the results are not entirely complete. Because of oddness in the way delegates are assigned, it is often the case that the votes are counted, the primary results published, but the delegate allocation incomplete. Texas took forever, it seems, to post its actual delegate count, for example. All three states that had contests yesterday have incomplete delegate counts, even though we know how people voted. Proportional representation applies in all three states but things are not so simple.
For example, in Arizona, a certain number of delegates are eventually (in April) selected at the congressional district level, a certain number are at large, a certain number are linked to party officialdom, and a certain number are linked to constitutional office. Delegates are selected at different times (most at the convention in April). There is a right of review (by the candidates) of some of these delegates. Some delegates are committed to vote a certain way on the first ballot at the national convention, some are uncommitted. There are threshold effects whereby certain delegates may not be assigned if the threshold is not met in the preference ballot (I think … this part confuses me). A delegate is both a number (i.e., 10 delegates for Mary and 10 Delegates for Sam) and a person (Joe Bleaugh will go the National Convention as a delegate). There are lists of delegates (as in number as well as personage) and the exact number of “delegates” that might be on that list depends on … all of the above.
And that is the simple version of it. The rules are 18 pages long. Arizona is not unusual. Anyway, Arizona has 75 pledged delegates, of which 63 are counted as pledged (though they don’t exist yet as people) now, but the rest will be eventually. In the end, the allocation will be close to proportional, but because of the precinct and district level math, and other things, the exact number for each candidate probably can’t be known at this time.
So, given all that, I’m going to go out on a limb and say that in Arizona, Clinton will have 44 pledged delegates, and Sanders will have 31. This is not the same number you will see reported, because several delegates are listed as “available” for reasons cited above.
Using the same method, Idaho will award 5 delegates to Clinton and a whopping 18 delegates to Sanders. This is close to the reported amount, but off by one. I’ll attribute that to the Washington Post’s rounding error.
Meanwhile, Utah has reported delegates nice and clean like, straight shooters that they are, and we have 18 assigned to Sanders and 5 assigned to clinton.
My model predicted that Sanders would win Utah and Idaho, and he did. My model predicted (along with everyone else in the country) that Clinton would win Arizona.
However, the numbers are different than expected. Sanders did much better than my model suggested and better than mainstream media expected.
My model had predicted that Clinton would walk away from yesterday’s contests with 82 delegates to Sanders’ 49 delegates. Instead, depending on rounding and other factors, Clinton will have 54 and Sanders 67.
This means that Sanders is walking away from Tuesday’s contests with more delegates than Clinton instead of the other way around.
I’ve stated several times that Sanders has to average 60% of the take for the rest of the contest in order to tie clinton. He didn’t do that this time, he only got 55%. But that is 55% on a day when the largest contest, Arizona, was expected to go very favorably towards Clinton. In other words, because of the variation across primaries noted above in the discussion of “what Sanders needs to do to tie Clinton,” Sanders may have actually done what he needs to do this week.
You see, Sanders is expected to get about 48% of the votes here on in, with Clinton at about 52%. He needs to achieve a seemingly unlikely 60%. Last night, he got 55%. My model suggested that last night he’d get less than the overall expected, by a tiny a mount (46.7%) but he got much more.
Sanders is expected to win many of the upcoming states. Hawaii and Alaska are next, and I have no idea what will happen in Hawaii. But he will likely win Alaska, then Washington, then Wisconsin, then Wyoming and, I think, New York. My current prediction is that he’ll take about 57% of the delegates trough that period, but if he performs better during that time than expected at the same level as yesterday, he could easily exceed the required 60% return and move significantly toward catching up to Clinton.
All I can say is that I like both candidates a lot and will be happy with either one. If you are supporting either of these candidates, I hope you keep in mind that it is still possible that the other candidate, the one you don’t support, whoever that is, may win. Vote blue no matter who!
Almost exactly 50% of the votes have been cast in the Democratic Party primary and caucus process. I’ve been updating a model to predict primary and caucus results all along, and the model has done fairly well. The most recent update, however, was a bit off. That update involved separating states into two groups, southern vs northern, then calculating different sets of likely voting patterns by ethnicity for those two groups, and integrating that with estimates of ethnic distribution (“white, black, hispanic”) among Democratic voters by state.
What I did not do in those models was to incorporate the effect of whether or not a primary or caucus is open, closed, or somewhere in between.
Now that we have had quite a few primaries and caucuses, it is possible to move to a somewhat more sophisticated model, because there is (probably) enough data.
I ran a multi-variable regression analysis that coded primary openness (0=closed, 1=semi open, 2=open) and whether or not a state is southern or not southern, then included the percent of each ethnic group by state.
The result indicated that the percent of a voting group (by state) that is hispanic did not influence the result. In doing the analysis I looked only at states, and excluded Vermont and New Hampshire because of the strong favorite son effect. The resulting model, naturally, predicts the number of delegates that have already been awarded to each candidate, in total, precisely, for the simple reason that the model is based on that number. Within the data set, the R-squared value is 0.83, which is pretty good. This means, roughly, that 83% of the variation in voting (by percent who voted for each candidate) is explained by those variables. The following table shows the actual delegates won vs. the delegates predicted by the model.
Also indicated is the spread between the two candidates in percent. The spread starts off a bit wonky because there are only a few contests, but then settles in to about 20% and remains at that level. Not shown is an analysis of the degree to which Sanders performed relative to expectations. If that number changed a lot, showing a trend, this would be important for predicting the future. The first half of the contests show Sanders under performing, according to this model, by 2%, and the last half have him over performing by 2%. So there may be a very low level “surge,” but not enough to make any real difference in the outcome.
So, what does the future look like? There are several states coming up where Sanders is likely to do well. But is it enough to make it likely for him to overtake Clinton? With a 20% spread and half the votes counted, Sanders would have to take an average of 60% of the delegates from here on. That is very unlikely.
The following table shows the primary and caucus outcomes through the present, followed by the predicted delegate commitments for the rest of the primary season. The percent spread between the candidates is indicated, and it does indeed drop over time, though slowly, reaching a minimum of 8% for the last few races.
The total number of delegates required to lock the nomination is 2,383. There are 717 uncommitted delegates (aka “Super Delegates”). If we assume that all of those uncommitted delegates will simply vote for the majority candidate, then the number of delegates required to have a likely lock on the nomination is 1669. This is not a fully supportable assumption because some of the uncommitted delegates may chose a different path, but it is a reasonable approximation.
The part of the table above marked in yellow indicates the approximate point in time when the leading candidate, Clinton, will get somewhere around 1669 delegates. So, if this model is reasonably accurate, Clinton will achieve a lock about mid May.
The next set of primaries, next week, are Arizona, Idaho, and Utah. In my view, these are somewhat hard to predict. Polls suggest a weak Sanders win in Idaho and a weak Clinton win in Utah. My model predicts a strong Clinton win in Arizona, and Sanders victories in Idaho and Utah. The total number of delegates at stake next week is small (131 in total). In order for Sanders to signal that he can overtake Clinton, he would have to win about 79 delegates in total. If he falls short of that, the rest of the road is more uphill. If he does better than that, then he may be seriously in the running.
Sanders is also expected to do well in the next several races (Alaska, Hawaii, Washington, Wisconsin, and Wyoming) according to my model. However, I don’t actually expect my model to work at all in Hawaii. My model suggests that he may well achieve over 55% of the vote in those primaries, but again, he will have to have already achieved 60% (unlikely) on the 22nd for this to start to accumulate to a catch-up number.
Following Wyoming is New York State followed by Super Tuesday III, six states with 631 delegates. My model suggests he will get less than half of these delegates, though he will do well in Pennsylvania and lose by not much in New York. I’m also predicting that he will win in California, in June, but not by much.
Between now and the end of the race, there are 1946 uncommitted delegates to fight for. Of these, the top five states account for a whopping 1138 delegates. These states are Washington, New York, Pennsylvania, California, and New Jersey. I predict he will come close to even with Clinton or win most of these states (but Clinton will do very well in New Jersey), but in order for Sanders to overtake Clinton by focusing on these states, he’ll have to do VERY well in all or most of them.
This model uses everything that happened before (mostly) to predict everything that will happen in the future. The first half of this series of events is over (in terms of delegate counts) and there is no evidence of any dynamic change occurring at the moment. This model does an excellent job at retrodicting the prior races, but it might slightly underestimate Sanders performance, since for the last half of the retrodicted contests Sanders outperforms the model by an average of 2%. However, in order for him to catch up to Clinton, he has to outperform the model by 10%.
The graphic at the top of the post is the predicted delegate counts for the entire primary season. The already-held contests are represented as predictions instead of actual because the final number (today’s delegate count) is the same for both predicted and actual. There is a slight narrowing of the gap (see table above) but not enough to change the outcome of Clinton achieving a lock on the Democratic Party nomination in May.