Okay guys, I’ve put this in Facebook posts and tweets and everything else so I guess I’m required to do it here. Two things to note, I am not *really* trained in statistics. I can do multidimensional statistics and I’m fairly good in R but I’m not really a statistician. But these are not really difficult tests. I didn’t need to bust out R for this; you can check my math in Excel and a lot of this you can probably do in your head, but we’re going to go through it bit by bit. I’m getting my numbers from Wikipedia and from Google. For some of these I did check and for example Real Clear Politics has differing vote counts and in some cases delegate counts than other sources. Some of this is because data might be pulled from exit polls rather than the actual results or it may have been compiled in a certain point in time and not checked or corrected since. That said, Google generally provides the most up to date results and … well you’ll see. The numbers are such that if I miss even fifty voters in a state it’s not going to change things.
This story starts in 2016. Well, technically it starts well before even 2008 but we’re only going back to 2016 for the time being, and before anyone runs to the comment section and accuses me of rehashing an old battle understand that Sanders in 2016 came much closer than he did in 2020. And that’s actually crucially important to understanding the results here.
So we do have a few key differences between 2016 and 2020 we need to address. I’m not going to touch on the whole incredibly misogynist comments and race bating, I’m just talking schedules and rules for now. So in 2016 the Democratic Party counted 4051 pledged delegates and 712 unpledged or “super delegates.” Pledged delegates were awarded proportionally for wins in primaries and caucuses and super delegates are “Party Leaders and Elected Officials,” (PLEO).
I need to briefly explain what a PLEO is. PLEOs are Democratic Governors, Senators, I believe members of the House, distinguished former elected officials like Presidents, and some party leadership and leaders of affiliated groups. This is not arbitrary and to get a leadership position within the Democratic Party you must be elected at each and every level. Within each state’s or affiliated organisation's structure they will hold elections to determine their leadership and SOME of these leaders also get to be super delegates. These are not members of some shadowy cabal; they are people who have been elected and who have worked hard within the Party often in voluntary positions for years. In some cases super delegates will be people you yourself voted for.
But anyways, super delegates get to decide for themselves who they’ll vote for unlike pledged delegates who have to honor their proportional allocation. The Republicans do a winner takes all thing rather than a proportional thing and maybe some day when I’m bored I’ll run the numbers on that but it might be responsible in part for Trump’s ascent.
In 2020 there are 3979 pledged delegates and 771 super delegates. So the proportion of super delegates in 2016 was 14.95% and in 2020 it’s 16.23%. The way it worked in 2016 is once a candidate got over half the total delegates, so 2383 they became the presumptive nominee and that would be confirmed by affirmation after a sort of ceremonial vote at the Democratic National Convention. This sort of thing is performative, but it’s something we humans need, so I acknowledge its inherent silliness, but let’s move on. Theoretically, if no candidate reached the magical 2383 number the first vote would likely be a bust unless supers shifted to the leader, but they’d sort it out in subsequent votes.
In 2020 the rules have been changed such that the magic number is only half of the unpledged votes (so 1991) and supers don’t get to vote until the second vote. So if a leader emerges it’s the same deal what with the presumptive nominee thing, but if one does not then it will just take longer to get to a nominee. Ironically, Sanders and his supporters pushed for this rule change and I put it like that because in no world does this help him out and it’s arguably hurting his chances this time around.
In 2016 if you just go by DNC rules Sanders did not have a chance the second he lost Iowa. Supers will already know a the Presidential candidates and knowing both Hillary Clinton (I’ll be referring to her by her first name or HRC so as to avoid confusion with her husband) and Sanders about five hundred of them chose Hillary. So when she won Iowa she already had about 543 delegates to Sanders’ 60. Iowa’s massively important because it showed — up until 2020 — whether or not a front-runner actually had it and could actually win the votes of the general public. So generally whoever was liked by the party would show up with their massive delegate counts and if they won Iowa great, and if they didn’t, well that’s a problem and they’d have to immediately take stock. That’s not applicable in this race because we are not counting supers until the Democratic National Convention.
The Very Recent History that Feels Like Eons Ago
We will throughout this article have to occasionally refer back to 2008. Barack Obama did not start with the super delegate lead. But he was well known and well regarded within the Democratic Party so as best I can figure out he had about 150 super delegates to start out with. Hillary Clinton did start out with more than Obama in 2008, but again, they were a lot better balanced. She did concede before it was confirmed she’d lost the popular vote as a result of her super delegates going over to Obama, but in popular votes Obama and Hillary were neck and neck until June 3 just before she conceded the race giving her formal endorsement of Obama on June 7. Obama’s proof of concept was in winning Iowa. He was the underdog and by winning Iowa he proved himself a contender. He then went on to win a bunch of other races swapping back and forth with Hillary as they contended for the nomination. 2008 was a close race, and again, we’ll be returning to it.
2016 though was a blowout. Sanders hates the Democratic Party and as a result, the Democratic Party isn’t too keen on him either. This is why Hillary started out with so many super delegates and Sanders with so few. Elected Democrats had been on the receiving end of his foolishness and weren’t inclined to support him whereas Hillary had always worked hard for them and the Party and so was in their eyes the better candidate. And okay, I get that that might make some of you mad but a lot of you are angrily saying Democrats have somehow insulted you and then expect you to vote for them. That’s basically how Democrats feel about Bernie Sanders. So, people are entitled to their opinion and most elected Democrats just don’t like him.
The Simple Numbers in 2016
So let’s talk numbers and get into this a bit. Just so you know what I’m doing I can’t really find a timeline for when exactly supers endorsed Hillary or Sanders in 2016, so in November of 2015 Hillary had 359 supers to Sanders’ 8 and I’m assuming that as Hillary had by May 520 supers and Sanders had 39 that those were about their numbers throughout the race. I appreciate that’s a big assumption but they ended with 570.5 and 42.5 respectively so I imagine I’m close. By the way, that half delegate … I’m not really going to get into it but Democrats Abroad basically do half delegates. So that’s probably what that’s about. I’m actually a member of Democrats Abroad and I don’t fully understand it myself, but that’s what that is. I think.
If in 2016 the presumptive nominee needed to get to 2383 with all possible delegates than HRC clinched it on June 7. That was actually the day she was declared the presumptive nominee. If the presumptive nominee needed to get to 2026 (half of the pledged rounded to the nearest whole number) with only pledged votes then HRC clinched it also on June 7. She closed out the race on June 14 with 2205 pledged delegates and 570.5 super delegates. Sanders lagged behind her with only 1846 pledged delegates and 42.5 super delegates. HRC was also way ahead in the popular vote with 16,904,874 to Sanders 13,222,946.
(In 2016 Washington and Nebraska held caucuses and non-binding primaries and in both cases Sanders won the caucuses and HRC won the primaries. If we include the votes cast in those non-binding primaries then the totals come to 17,368,027 votes for Hillary and 13,642,983 for Sanders.)
At no point was Sanders in any position in 2016 to win the nomination. Even if we ignore the super delegates, he lost the pledged delegates never breaching 1900 let alone 2026 and he lost the popular vote by more than Trump lost by. Sanders won only 42.64%-43.99% of the vote and 45.57% of the pledged delegates while HRC took 54.51%-56.01% of the vote and 54.43% of the pledged delegates. So in this instance super delegates did not matter. Hillary did not need them to win.
And it actually gets much worse for Sanders in 2016. If we do actually count supers then Sanders path to the nomination was pretty untenable. He started out way behind and lost more and more ground with each race. After March 15, 2016, counting just unpledged delegates he needed over 57% of the remaining delegates to win. And counting supers he needed over 73%. He was in a bad position because of the supers, but Obama proved that you can start with fewer supers and still win. The issue for Sanders in 2016 is simply that he was not able to get the sort of grassroots support that either Obama or Hillary did.
In 2016 Sanders overall average in all primaries and caucuses was 47.29% with a standard deviation of 0.1633. Now, HRC’s average was only 50.71% with a standard deviation of 0.1611, so she was by no means unbeatable, it’s just that Sanders failed to beat her.
In 2016 for Sanders to have won after March 15th his performance would have had to have been statistically significantly better than what it actually was and statistically significantly better than it had been before March 15th. Basically, after March 15th if Sanders had actually pulled off a win it would have set off everyone’s alarm bells because he almost definitely would have had to cheat to do it. I know 57% may not sound like a lot, but in a national election in a country as big as the United States, it really is.
The Simple Numbers Compared to the Present
But I’m actually only using 2016 as a comparison here. The big difference comes in 2020. As of now, late March 2020, Sanders is losing pretty hardcore. We’re dealing with a different set of rules this time around so Iowa was not quite as important as it was in 2016 or prior, but Sanders is still losing. By a lot. According to Google, Joe Biden has 1201 delegates to Sanders 896. My count has them at 1199 and 894 respectively and Wikipedia is … well it’s around there. But in every single report Sanders is about 300 delegates behind.
If we look at just the delegate counts from 2016 this is about comparable. Taking just a straight count of Hillary’s and Sanders’ pledged delegates as of March 15, 2016 you get 1170 and 861 respectively. So similar to now, Sanders was trailing by 309 pledged delegates. If you rearrange the 2016 calendar to match the 2020 calendar Hillary would have 1298 delegates to Sanders 1102, so the spread would be 196.
This would suggest that the 2020 calendar is better for Sanders than the 2016 calendar so let’s have a look at Sanders’ relative performance in both 2016 and 2020.
It’s pretty clear that 2016 Bernie Sanders put on a much stronger performance than 2020 Bernie Sanders. Before we process anything note that the total number of delegates and the vote count for Sanders is higher in 2016. We’ll talk in a minute about the delegate count but the number of votes cast for him in the primary going down is a huge red flag. There are fewer caucuses in 2020 than there were in 2016 and Sanders is supposedly the guy bringing in new voters. 2018 was a banner year for Democrats, so if Sanders were headed towards the nomination he should have a hefty lead in votes counted buoyed by his own increased popularity, new voter turnout, and the increase in primaries.
But we’re seeing the exact opposite happening.
And it’s significant. If you compare Sanders’ 2016 and 2020 share and vote count in primary contests it’s statistically significant to 0.05. So if we’re just looking at voter enthusiasm for Bernie Sanders between 2016 and now, it’s pretty clearly decreased. Demonstrably in 2016 he did lose by quite a lot, but he’s now doing even worse. This is not recoverable.
Some of you may be tempted to blame the relative popularity or lack of popularity of either Hillary Clinton or Joe Biden so … okay let’s look at how they did side by side.
Overall, Hillary Clinton and Joe Biden put in relatively comparable performances. Biden’s not doing quite as well as Hillary did but it’s only statistically significant in the share of votes. As far as the delegates and vote count goes, they’re not all that different.
If one were trying to explain why Sanders is doing so poorly in 2020 the most obvious thing to point to would be that there were more candidates in this for longer than in 2016 which necessarily decreases the share of votes along with the delegate count and total number of votes cast. Okay. Fair enough. But that would effect Biden too. Again, Biden’s not doing as well as HRC did, but the number of votes cast for him and the number of delegates now pledged to him is not significantly different from those HRC had.
The 2020 Race Right Now
At this point the sane thing to do for Sanders and really everyone would be for Sanders to withdraw from the race. These numbers in no way indicate that there is any of that verve or enthusiasm we’ve heard so much about. I’m constantly loudly informed that Sanders has grassroots support that just hasn’t materialised yet.
The problem with that is the evidence cited to support the argument for Sanders’ grassroots support is his fundraising capabilities and the fact that he seems to have taken on quite a few micro-donations. Okay. Again, cool. But there’s two problems with that.
One: Sanders’ campaign specifically asked for multiple small donations from the same donors, meaning the number of donations is not at all indicative of the amount of votes he has or will receive. This was meant to create the impression that he’s on Obama’s path to the nomination, but Obama started with more support from elected democrats and wasn’t asking rich donors to cosplay poor. Obama had actual grassroots support. He didn’t have to fake it.
So, this indicates that it doesn’t particularly matter what kind of donations you’re receiving or if you’re taking any at all: money is not indicative or predictive of support.
But okay, let’s just ignore all that and talk about what a path to the nomination for Sanders would look like at this point. Twitter oft reminds me that there’s still something like 40% of the delegates available. So let’s examine that claim.
The above is my little chart of how Biden and Sanders are doing in this primary. Biden may not be doing as well as Hillary Clinton did in 2016, but he’s basically sweeping the map. At 19 contests won thusfar he’s won more than double what Sanders has at 8. Hillary Clinton in 2016 ended up with 34 to Bernie Sanders’ 23, so Biden has not yet surpassed her, but he’s on track to do so.
But I keep on promising to talk about delegates so let’s talk about delegates. If we add those two totals together we get 2093 and if Google is to believed I need to add another four to that bringing us to 2097. “BUT WAIT,” you may declare, “delegates were awarded to candidates who have since dropped out and those delegates have not yet been reallocated.” And you’d be right. So the number of delegates which have been decided is 2305 meaning the remaining pledged delegate total is 1674. That is indeed about 42% of the total number of pledged delegates. Well done, everyone!
But the problem is that is not enough for Sanders. Remember, our magic number in this case is 1991. Biden needs an additional 792 or 47% of the remaining delegates to get to that number while Sanders needs 1097 or 65.5% of the remaining delegates. That’s actually a greater percentage than he needed at this time in 2016 of just unpledged delegates, and that was still super untenable.
To be entirely honest Biden will have to consistently perform better than he has to get to 1991, but given that his only remaining competition is Sanders that shouldn’t be too hard. Biden is in a very similar position to Hillary in that he could consistently lose all of the remaining contests to Sanders and still win the nomination because Sanders is that far behind. Both Biden’s and Sanders’ shares are kind of artificially low because there was no clear front-runner until South Carolina and super Tuesday, but Biden is actually in a better position than Sanders in all ways.
It’s no secret I’m no fan of Sanders. And I’ve even written pieces about how I did initially have fear that he wasn’t bluffing. But there’s nothing there. He demonstrably does not have the support he says he does. He did raise an impressive amount of money and people who understand campaign finance better than I are having a go at that too, but as I said, money does not equal success.
Sanders cannot win. I, personally, don’t want him to win, but when I say that, I’m just talking about the numbers. He does not have any of the support or yes, popularity to win. This is never going to happen for him. In this article I’m not really interested in why although I have opined on that before and likely will in the future. But numerically it’s just not possible. It didn’t happen in 2016 when the stars aligned for him and now — to mix the metaphor — his political star is fading.
Some of Sanders’ supporters have stopped tweeting #votebluenomatterwho and are now demanding the Democratic party simply name Sanders the nominee for reasons. So firstly, that’s ridiculous. The Democratic party is nothing if not democratic. The clue is in the name. So if the “Party elite,” decided to name Sanders the nominee in either 2016 or 2020 there would be a mass defection. Sanders’ supporters who go for this say that if Sanders doesn’t win they won’t vote or they’ll vote third party. That’s awful and it’s on Sanders to follow Hillary Clinton’s 2008 example and keep that from happening, but let’s assume he doesn’t and they’re not bluffing. That is effectively why HRC isn’t President now. And HRC did win the election against Donald Trump.
But Sanders’ support has seriously slipped. (I apologise: I’m rather enjoying alliteration at the moment.) Even if all of Sanders supporters refused to vote for anyone but him in the general there’s simply not enough of them remaining to sufficiently turn the tide. But the other thing, and this is crucial, is Sanders is a rather divisive character. The reason Biden’s support coalesced in South Carolina and thereafter is because most Democratic or liberal voters did not want Sanders as the nominee. That’s democracy, sure, but a lot of people supported for instance Klobuchar, but threw their vote to Biden simply because they didn’t want Sanders.
That means even if Sanders were nominated by the “Establishment,” most Democratic voters would not vote for him. I understand that there exist polls from both 2016 and 2020 that say Sanders wins in a head to head against Trump, but polls and votes are very different things. We have hard concrete numbers now from people who have voted and they indicate that not only is Sanders not popular, but he’s even less popular than in 2016 when he lost. It would be a tactical disaster for the Democratic Party to hand him the nomination based on the threats of a privileged few in his own dwindling base.
But my last thought for this post is that with the COVID-19 pandemic raging it’s really crucial that Sanders withdraw from the race. He got very angry about that idea when asked a few days ago, but the fact of the matter is he will gain nothing by remaining in the race partially because of the numbers and partially because he cannot campaign with this going on. He was specifically angry because he said he was dealing with too much but if that’s the case then he really shouldn’t remain in the race. Either way, Joe Biden and Michael Bloomberg have both dedicated money and resources to help slow the spread of this virus and “flatten the curve.” For Bloomberg, that’s his own deal, but Sanders distracting Biden at this point particularly when Sanders hasn’t really made an effort or been a leader during this crisis is inappropriate of him.