This is the website of Abulsme Noibatno Itramne (also known as Sam Minter).
Posts here are rare these days. For current stuff, follow me on Mastodon
|
Since the New Hampshire primary, I have seen a flurry of Facebook posts and numerous articles all over the place, first by people suddenly shocked to learn of the existence of superdelegates, then folks complaining about the DNC “screwing over” Sanders, and then folks claiming superdelegates should be ignored because they don’t matter. All this in response to places, like my site electiongraphs.com, that point out Clinton’s huge lead due to superdelegate counts.
My current chart of raw delegate totals looks like this:
That is Clinton 461 to Sanders 50 including the latest updates that prompt this post.
The claim is that the Clinton lead here is a complete illusion. That counts that look like this are simply anti-Sanders propaganda meant to discourage Sanders voters, that looking at numbers like this is meaningless, and one should only look at “pledged” delegates, where the total right now would be Sanders 36, Clinton 32. That would indeed show a very different picture of the race. Sanders would actually be winning!
But really? Would that tell us where the race really is?
First of all, there isn’t any secret or underhanded manipulation. There are 32 delegates in New Hampshire. 24 of them were to be determined by the primary. Of these Sanders won 15 and Clinton won 9. Some of those are determined proportionately in each congressional district, some proportionately by the statewide results. Those numbers come straight from the votes. Just math.
Then there were also 8 “super delegates”. That is one Senator, one Representative, one Governor, and five members of the DNC. These are eight people who have names and have been known for ages. Six of those eight had declared they supported Clinton long before the primary. The other two haven’t publicly stated a preference yet. That means the delegate total in New Hampshire is now 15 to 15 with 2 not yet determined.
The Sanders folks might not like that, or like the existence of super delegates at all, but none of this is new or an unexpected manipulation or anything, it is simply how the race is structured. A lot of people are expressing shock because they are just now realizing this, but it is no surprise to anyone familiar with the Democratic primary process, and certainly Sanders’ campaign staff is quite familiar with it.
As we’ve chronicled here, Clinton started out with nearly 8% of the national convention delegates already in her column before the first votes were even cast in Iowa and New Hampshire. She has been working on cultivating those people since at least 2013, probably earlier. Sanders managed to get a handful of these folks, but not many. So he started out way behind, and had to get over 54% of the remaining delegates to be on a pace to catch up and win. A majority wasn’t good enough.
It isn’t a good spot to be in for him, and you might argue that super delegates shouldn’t exist. Yes, superdelegates were originally created and exist to give the party apparatus some buffer against a popularly supported insurgency, and that sounds like Sanders. But the superdelegates have existed for decades and they are a known part of the process.
So complaining about them at this point is along the same lines as complaining about the electoral college. The person complaining might be right that in some ways a straight popular vote would be more “fair”, but these were the rules of the game and anybody running knew the rules when they chose to play.
It wasn’t superdelegates specifically that made the difference, but remember that in 2008, Hillary Clinton won the popular vote in the Democratic primary process.* It was only the structure of the delegate race that won Obama the nomination. Because those were the rules of the game. That is still the case today. And that includes superdelegates.
OK, fine, but what about the fact that super-delegates can change their minds? Their votes aren’t locked in stone, and they certainly wouldn’t want to overturn the results if a candidate was winning all the contests and had a majority of the non-super delegates, right? So since they can change their minds and will probably support “the winner” we should ignore them, right?
Indeed, it is important to note that the super-delegates are real people and complete free agents. They can and do change their minds. And yes, superdelegates may be reluctant to be the deciding factor in a close race, but you can’t really say “the super delegates would do X”. It is 712 individual humans, who will each decide on their own what they want to do.
Many superdelegates have yet to express a preference at all and can probably be fairly easily swayed. They have not expressed an opinion for a reason, and may well go either way. Others have publicly supported Clinton, but it could be weak support and maybe they could be flipped if they were actually courted. Many others are actually actively working on the Clinton campaign in prominent state level roles. People in this last category are unlikely to change their votes unless or until Clinton herself drops out and asks them to. At the very extreme, Bill Clinton is a superdelegate. Is he going to vote for Sanders?
You really do have 712 different people, with 712 different reasons for their preferences, and 712 different degrees of how strongly they support their candidate.
If Sanders was serious about the superdelegates, he and his surrogates wouldn’t be pushing talking points about how superdelegates don’t matter and will vote for the popular winner so they should be ignored. Instead he would be heavily lobbying the currently uncommitted superdelegates to start backing him. They are a constituency to be courted. They are an IMPORTANT constituency to be courted.
If you were to see such movement, the first place you would detect it would probably be a higher percentage of currently uncommitted superdelegates raising their hands for Sanders when they do come out and reveal a preference.
Superdelegates who have already publicly stated they are supporting Clinton will probably not start actually flipping back to uncommitted or to Sanders until or unless things are actually looking very dire for Clinton.
If you do end up seeing Sanders winning more contests than he loses and building up a big lead in the regular non-super delegates, you may start to see both kinds of movement. After all, people like backing a winner. But if it happens, you will actually see that change happening in the delegate counts if you watch this site or one of the other places that include superdelegates in the totals. It will not be invisible.
There is no sign of any such movement yet.
Can Sanders win without flipping any Clinton superdelegates over to his side? Yes, of course he can. It is a high bar, but it is within the realm of possibility. He just needs to get about 55% of the remaining delegates.
Can Sanders win with even a smaller portion of the remaining delegates? Why yes, of course he can, if he starts convincing Clinton delegates to change their minds. But if that is going to happen, they have to actually be convinced to do so. Here is a list. Go start lobbying them. Try to change their minds. But nobody should be assuming that is automatic.
Yes, if Sanders ends up leading the non-superdelegate count at the end of the primary season and superdelegates are the only thing putting Clinton over the edge, there will be extreme pressure on the superdelegates to flip. But they still have to actually do so. And most likely, when they start waffling on their positions, they will actually say so, and we will know.
The delegate counts including superdelegates are NOT misleading. You just have to interpret them correctly… meaning that you know superdelegates are actual human beings that can change their minds, and you think about what situations might lead that to happen. But ignoring that they exist and just assuming that of course they will change their minds is extreme folly. It is sticking your head in the sand and ignoring reality. It is willfully blinding yourself in order to believe in an illusion instead. If and when the superdelegates start changing their minds, we will see it. Until or unless that happens, Clinton really does have a huge delegate lead that Sanders would have to overcome in order to win.
Meanwhile, as the Sanders camp has been complaining about superdelegates and how they don’t matter, superdelegate movement to Clinton has continued. Since the New Hampshire Primary results, I’ve added 44 additional superdelegates to Clinton’s column. Not all of these actually announced their support this last week, some did so months ago, but I just found the references more recently. Their support gets backdated in the charts and graphs to the date of the public source documenting their support. I also removed 2 Clinton superdelegates that had been accidentlly double counted. Net gain, Clinton +42.
There were no new Sanders superdelegates.
So where does that put us on “% of remaining delegates needed to win”?
Clinton now needs only 45.18% of the remaining delegates to win.
Sanders needs 54.84%. Right before the Iowa caucuses, my best estimates were that sanders needed 54.22%. (Including pre-Iowa superdelegate preferences I know about now that weren’t included at the time, it would have been 54.65%.) So despite his close second in Iowa, and his win in New Hampshire, Sanders is worse off in the delegate race than he was before Iowa. But not by much… he has ALMOST been holding things steady. He needs to do better to catch up and win, but Iowa and New Hampshire are only a small portion of the overall delegates, so there is still a lot of room to maneuver.
The next delegate race is Nevada.
There are 35 delegates up for grabs in Nevada that will be determined by the caucuses there. Plus 8 superdelegates. Of these, 3 have already said they are for Clinton, and 1 has said they are for Sanders. So don’t be surprised when you see those added in at places showing the state delegate count. They aren’t part of the 35 determined by the caucus though.
Of the 35, Sanders needs to get 20 in order to be on a pace to catch up with Clinton. (Again, assuming no superdelegates change their minds, if a lot do, this number becomes lower.) The RCP average has Nevada at Clinton 47.5%, Sanders 36.0%. That certainly wouldn’t get Sanders the 20 delegates he needs, but that average only includes one post-New Hampshire poll, and that new poll shows the race tied. The Pollster average discounts the pre-New Hampshire results entirely and shows the race at 45% to 45%… a tie.
Delegate rules are tricky. In Nevada some delegates are determined at the congressional district level, others at the state level, there is rounding involved, etc, etc. It is also a multi-stage caucus process, not a primary, so the “real” delegate totals won’t be known until May. But we will have estimates.
Remember though that it is not that Sanders needs 54.84% of the popular vote. It is that he needs 20 of the 35 delegates. That is the number to watch.
And superdelegates of course. :-)
A final thought: It is very possible to catch up from a deficit. But catching up is much harder if you pretend you aren’t actually behind.
* It has been pointed out to me that this would likely not have been the case if Obama had, like Clinton, chosen to compete in Michigan despite DNC sanctions… Clinton essentially ran unopposed in Michigan and this margin was enough to put her in the lead in the national popular vote. If you do not include Michigan, Obama lead the popular vote in 2008. This does not however change the point, which was that there are many ways that the delegate rules can lead to results that differ from the popular vote, of which the superdelegates are only one.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
[Edit 18:06 to add link to Wikipedia superdelegate list.]
[Edit 18:48 to add final thought.]
[Edit 2016-02-16 06:50 UTC to add asterisk note about Michigan.]
I mentioned yesterday that the delegate totals were provisional and the counting wasn’t done yet. The counting is complete now, and in addition there have been some clarifications to the delegate allocation rules.
First on the rules, the way in which rounding is done was clarified. Instead of just rounding at the end, the percentages are rounded to a whole number first, and then the delegates are rounded to whole numbers at the end. This is a bit of a silly way to do it, because you are introducing the rounding distortion twice instead of just once, but that’s how New Hampshire chose to do it.
Then a rule change that affects not just New Hampshire, but all subsequent states too. Frontloading HQ has the scoop. The bottom line is that national RNC party rules changes are overriding rules at the state level about what happens to the three “automatic delegates” in each state. These are the National Committeeman, the National Committeewoman, and the chairman of the state Republican Party who all get to be delegates “automatically” based on their positions. In some states they were bound by primary or caucus results. In some states they were required by the rules to stay neutral. But in others, they were free agents. In the past the ones that were free agents had been the Republican party’s equivalent of the Democratic superdelegates. But the RNC national rule change is requiring these three automatic delegates to always be “bound” by the same rules that cover at-large delegates in each state.
So, essentially, no more superdelegates on the Republican side. All delegates will be bound by primary or caucus results. (There is still an exception for delegates who are directly elected by their own name apparently, but that is rare.)
For New Hampshire, when you consider final vote totals, and both rule clarifications, this changes the results from the primary from:
11 Trump, 3 Kasich, 2 Bush, 2 Cruz, 2 Rubio, 3 TBD
to:
10 Trump, 4 Kasich, 3 Bush, 3 Cruz, 3 Rubio, 0 TBD
The updated raw delegate chart showing this change is here:
The net result here is Trump loses one delegate, Cruz, Rubio, Kasich and Bush all gain one.
Trump is still clearly in the lead, but it does change his New Hampshire results in an important way. 11 out of 20 delegates was a majority, and above the threshold he needed to improve his “% of remaining delegates needed to win” number. 10 out of 23 is not a majority, and was not enough to improve his position on that metric:
If you look at full movement from before New Hampshire to these updated results, you can see that ALL of the candidates now got worse, as none of them got the level of delegate support needed to be on track for an outright win. Trump specifically went from needing 50.37% of the remaining delegates, to now needing 50.43% of the remaining delegates. This is much less of a change than the others of course, but it still went up.
So the statement I made in yesterdays update that “if Trump continues to match New Hampshire, he wins” is no longer true. If Trump (and the rest) continue to match New Hampshire, we are in for a contested convention.
Of course, we won’t see New Hampshire cloned in the subsequent states. We will see continued developments in the campaigns, more candidates drop out, and a very dynamic race continue to play out.
And fundamentally, the one delegate difference in Trump’s total doesn’t change the over all picture all that much.
I won’t repeat the details of yesterday’s post walking through scenarios for South Carolina and Nevada. If you haven’t already, go read it.
The bottom line is that unless there is a major change before South Carolina, it is very plausible that Trump walks out of it with a significant delegate lead. Updating for the New Hampshire changes above, that could look something like:
67 Trump, 11 Cruz, 10 Rubio, 15 Others
Nevada is more proportional, but will still give Trump a further opportunity to increase his delegate lead. Absent a surprise loss in South Carolina, he will almost certainly have a majority of the delegates allocated by then.
At that point, only 5.4% of delegates will have been decided. Lots of delegates left to play with.
But that is when things speed up. If the non-Trumps are going to consolidate support by all but one or two dropping out, it really should happen by the end of February, because 25.8% of the delegates will get decided on March 1st. And then another 30% or so in the two weeks between that and March 15th. When we pass March 15th, 60% or so of the convention delegates will have been decided.
If the anti-Trumps consolidated before then, it is possible that one of them will be leading instead of Trump. If the anti-Trumps are still divided though, then if Trump has managed to get over 50%, he’s almost certainly going to be the nominee. If he is leading, but under 50%, then talk of a contested convention becomes real, not just news junkie fantasy.
Absent the oft-predicted but never realized Trump collapse, for any of the other Republicans to beat Trump, all but one (MAYBE two) need to drop out and drop out quickly, and then they need to focus all of their effort on Trump. But for the moment, all the anti-Trumps are still squabbling with each other, leaving Trump to consolidate a lead while they are for the most part continuing to ignore him.
That is just a formula to anoint Trump. At some point people will drop out, and the race will narrow. The question is does it happen in time for whoever is the surviving anti-Trump to have a real chance.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
[Edit 2016-02-14 06:18 UTC to add end note.]
New Hampshire’s delegate rules are such that the winner gets a big bonus. (The details of how this works were discussed here last week.) So Trump’s approximately 36% of the vote turns into 55% of the delegates. Counting isn’t actually done yet, but if current trends hold, the delegate haul out of New Hampshire will be:
Trump 11, Kasich 3, Bush 2, Cruz 2, Rubio 2.
Adding these to the previous results in Iowa, and we have these new totals: Trump 18, Cruz 10, Rubio 9, Kasich 4, Bush 3, Carson 3, Fiorina 1, Huckabee 1, Paul 1.
In the form of a graph of the total delegates, we see this:
Some groupings look apparent on this chart.
Trump is out by himself in the lead.
Cruz and Rubio make up a second tier, at about half of Trump’s level.
Kasich, Bush and Carson are the third tier. Way behind, but at least with multiple delegates.
Then finally Fiorina, Huckabee and Paul bringing up the rear with one delegate each. Two of these three have already dropped out. It will be surprising if the third does not do so soon.
How does this match what we’re hearing in the spin out of New Hampshire? Well, there is wide acknowledgment that Trump did have a nice win. But the majority of the attention still seems to have been on the race for second, and how that positions the anti-Trumps going forward.
Within that, relatively little attention was paid to Cruz, and he is still in 2nd. Rubio is being dismissed because he did badly in New Hampshire, but he is still in 3rd over all. Meanwhile lots of attention given to Kasich and Bush, but they are still stuck in the third tier with Carson with less than a handful of delegates.
Time to flip this and look at this in terms of my favorite metric, the “% of remaining needed to win”:
Trump exceeded the 50.37% of the delegates he needed to get to improve his position. Going forward he needs 50.33% of delegates to be on a pace to win. That is a relatively small change. With almost 98% of delegates still to be determined, there is a huge amount of flexibility available here to all the candidates.
Given that Trump got 55% of the delegates in New Hampshire, that means however that NONE of the other candidates got what they needed, or even close. Essentially, every one of the non-Trump candidates fell further behind today.
Now, Trump still doesn’t have a majority of delegates. To be on a pace to actually win, rather than just to have the most delegates going into the convention, he needs to do better than he has so far. But he is far closer to that mark than any of the other candidates. And if he continues matching his New Hampshire results, he will win.
The story here continues to be that while Trump hasn’t managed a majority in any contest so far, the opposition is so divided that Trump still easily wins (New Hampshire) or at the very least comes very close (Iowa). Does this continue? Does it get worse?
Time to look at the next contest on the Republican side, South Carolina.
There have not been any South Carolina polls since before Iowa, so new polls are eagerly anticipated, but at the moment the RCP average has Trump at 36.0%, Cruz at 19.7%, Rubio at 12.7%, Bush at 10.0% and then everybody else.
Now, that may change. Specifically if either Rubio or Bush drop out that will scramble things a bit. But that may not happen. There is a case to be made for both of them sticking in at least through South Carolina.
Trump probably wins this. But by how much? Time to look at the delegate rules in South Carolina.
From Greenpapers:
- 21 district delegates are to be allocated to presidential contenders based on the primary results in each of the state’s 7 congressional districts: each congressional district is assigned 3 National Convention delegates and the presidential contender receiving the greatest number of votes in that district will receive all 3 of that district’s National Convention delegates. [Rule 11(b)(4)]
- 26 (10 base at-large delegates plus 16 bonus delegates) at-large delegates are to be allocated to the presidential contender receiving the greatest number of votes in the primary statewide. [Rule 11(b)(5)]
- 3 party leaders, the National Committeeman, the National Committeewoman, and the chairman of the South Carolina’s Republican Party are pledged to the presidential contender receiving the greatest number of votes in the primary statewide.
If Trump wins by any margin, no matter how small, he gets 29 of the 50 delegates right up front. That already would give him 47 of the 100 delegates that would have been allocated in Iowa + New Hampshire + South Carolina.
Of course, in order to win the whole state, he would have had to have won some of the congressional districts too. Probably most of them. And so far in Iowa and New Hampshire, it doesn’t seem like there has been a lot of geographic lumpiness to Trump’s support. So he may well win ALL of the congressional districts. In that case Trump would walk out of South Carolina with all 50 of the available delegates. This is very possible.
In that scenario, after South Carolina, the overall delegate count would look like: Trump 68, Cruz 10, Rubio 9, Others 13
Trump would have a huge delegate lead. Even if he doesn’t win all 50, he will still have a very large delegate lead.
Nevada is three days after South Carolina. It is proportional, not semi-winner-take-all like South Carolina, but the latest polls still have Trump with a nice lead in Nevada. Those polls are old and out of date, but does anybody think two straight wins would hurt Trump’s position?
To “stop Trump” things have to move very quickly to a one on one with Trump vs Somebody. Maybe a three way race could do it, but might just be a recipe for a contested convention. But as long as we’re still looking at four or five candidates splitting the non-Trump vote, Trump wins.
But the non-Trumps still seem to be in a locked battle with each other for that anti-Trump one on one slot. And in doing so, they just let Trump win. The race will of course consolidate further. More candidates will drop out. Some may drop out before South Carolina. But it is unclear if that will be enough to let one of the others actually win in South Carolina.
Trump won New Hampshire. Trump looks like he will probably win the next two states. Things get more complicated once we get into March, but Trump would get there with a substantial lead.
Things are by no means settled yet. There is a long way to go. Even after we get past both South Carolina and Nevada 94.6% of delegates will still be up for grabs. There will still be time for the non-Trumps.
But if we only get consolidation down to 2 or three candidates after the March 1st contests, when 25.8% of the total delegates will be decided, the non-Trumps that are left may have a really tough uphill fight to catch up.
If consolidation doesn’t happen until after March 15th, 60% of the delegates will already have been decided, and it may just be too late. (Although if several non-Trumps are still strong at that point and Trump has the lead but not a majority, we’re looking at a contested convention, and that starts getting into completely different scenarios.)
But if the non-Trumps want to actually win, they have to stop fighting each other, some of them have to drop out, and they have to actually act like Trump is the frontrunner, and that their priority has to be bringing him down rather than each other. Otherwise they are done, and Trump ends up walking away with this.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
Lets jump right to it. In the last Democratic update we mentioned that to actually improve the “% of remaining needed to win” Sanders needed to get 54.36% of the delegates available in New Hampshire. That would be 14 out of 24 delegates that were available today. He got 15 out of 24, or 62.5% of yesterday’s delegates. So he did it!
In addition, since my last update, there were some superdelegate changes. The net change from those was +1 Clinton, +1 Sanders. There was also a net reduction of 1 available delegate, bringing the total number of Democratic delegates down to 4,763.
With all of that in place, here is the new chart for “% of remaining needed to win”:
The Sanders % Needed has indeed gone down, and he made Clinton’s go up!
Sanders’ % Needed specifically moved from 54.36% to… 54.30%.
This was at 54.22% before Iowa. 54.30% matches where this was right after Iowa.
So, Sanders did improve his situation, but only enough to undo the superdelegate changes that happened after Iowa, not enough to get back even to where he was before the voting started.
Another way to look at this is to just look at New Hampshire itself. Of delegates allocated by the Primary, Sanders got 15 delegates to Clinton’s 9. But New Hampshire also has 8 superdelegates, and 6 of these had already declared for Clinton.
So the actual New Hampshire total as of today is 15 Clinton, 15 Sanders, 2 TBD.
So even with his big win, which was indeed a big win, Sanders actually has only pulled even in delegates in the state.
Now, of course, superdelegates can change their mind. It has been said that if one candidate was clearly leading absent superdelegates, the superdelegates would not go against them, and would start to change their minds and flip to the candidate leading in non-super delegates. Maybe. It could happen. And if it does, that movement will be tracked here. Certainly no such movement has been seen yet.
In the mean time, next up for the Democrats is Nevada on February 20th. Nevada often gets relatively little attention compared to South Carolina, but Nevada will actually be the next place where we see if Sanders is able to take his New Hampshire results and build anything longer term out of it.
Nevada is a caucus state, so delegates won’t actually be allocated on the 20th, but there will be results that can be used to estimate convention delegates. Nevada has 43 delegates. Of those 8 are superdelegates. Clinton already has 3 superdelegates in her column, Sanders has 1. Nevada’s caucuses will determine 35 delegates. 23 will be allocated based on results in the 4 congressional districts, 12 will be allocated based on the state wide results.
To once again have a result that keeps him on pace to catch up to Clinton and win, Sanders needs to get 20 of the 35 delegates available on caucus day, assuming no additional superdelegate declarations or other adjustments between now and then.
At the moment, Nevada poll averages show Clinton leading 50.0% to 30.5%… but there has been no polling there since December. Expect a variety of new polls in the next week. Watch them carefully. For Sanders to be on track, these numbers have to flip… and then Sanders has to build an even larger lead.
In South Carolina, a week after Nevada, Clinton is even further ahead, although again polls there are stale and new ones taken post-Iowa post-New Hampshire are eagerly awaited.
Between now and Nevada, and probably between now and South Carolina, there will be almost non-stop talk about Sanders’ threat to Clinton, how Clinton is a lot weaker than expected, how vulnerable she is. A lot of this is true. To a degree.
But to evaluate if Sanders is actually a threat to Clinton eventually taking the nomination, as opposed to just being a temporary speed bump in her way, Sanders has to not only take the lead in Nevada, South Carolina and beyond, but be ahead by a significant margin, not just a little bit. This is a tall order. Maybe it is possible.
But if there is talk about how Sanders is “closing the gap” or “making it closer than expected” in Nevada or South Carolina, keep in mind that isn’t enough. He needs to actually win and win by a large enough margin to get more than 54.30% of the delegates to be on pace to catch up and win. Otherwise, the task in later states just becomes even harder.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
The Iowa results on the Republican side were officially certified on Wednesday, earlier than anybody expected, and with a surprise. Almost every source believed 27 delegates would be decided by the caucus results, but the remaining 3 “automatic delegates” who are party leaders, would attend the convention unbound and able to support whoever they wished. But it looks like those three delegates will be bound by the caucus results as well.
When you do the required math on 30 delegates instead of 27, everything remains the same, except Huckabee gets a delegate too.
Updating the chart showing the “% of remaining needed to win” graph:
And the raw numbers:
Huckabee and Paul have suspended their campaigns, but since these are bound delegates, they keep them unless the candidates officially release them. I haven’t seen a report of that yet, so won’t remove them from the list at this point. Not that it matters much for single delegates. This is a much more important factor when candidates drop out later in the race once they have already accumulated a significantly sized chunk of delegates.
Huckabee’s one delegate makes no difference to the analysis I posted Tuesday or the follow up from Wednesday.
We still have three candidates showing strength going into New Hampshire It is within the realm of possibility that New Hampshire will add a fourth to that list, but just as likely it will solidify the “three way race” picture. And as I outlined in the Wednesday post the specific New Hampshire delegate rules matter a lot. Specifically the rules essentially give extra bonus delegates to the winner, and don’t allow delegates at all for anyone coming in under 10%. This is a recipe to transform a plurality win to a healthy delegate majority.
So if you are watching the polls as they start to include post-Iowa data, watch carefully not only who is in the lead, but exactly who is above or below the 10% threshold, because that will make a huge difference in the delegate counts.
And an 8-7-7 split with only 1.21% of delegates determined is essentially a tie. And a tie that will soon be overwhelmed numerically by the results from New Hampshire and beyond. So don’t get sucked too deeply into the spin from all quarters taking about who is up and who is doomed. It is still too early for that. (Well, at least if you don’t drop out.)
Five days until New Hampshire.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
I’ve promised not to post every day there is an update of a superdelegate here or there. Instead, I’ll generally include them in the next post after an actual primary or caucus. However, if enough accumulate, I’ll go ahead and do a post just on superdelegates. There have been 10 superdelegate changes since my last post on the Democrats. That’s as good a threshold as any, so here is a quick update.
First of all, I probably should have made this change before that last post, but I didn’t… O’Malley dropped out on caucus night shortly after the results started coming in, as it was immediately clear that as expected he would not have a good night. If O’Malley had earned actual pledged delegates, they would remain in his column until such time as O’Malley formally released them. But these superdelegates are always free agents, and with O’Malley out, there is no reason to believe their previous endorsements of O’Malley still stand. So I zeroed out O’Malley’s three superdelegates. Goodbye O’Malley.
In addition, news reports showed 5 new Clinton superdelegates in Illinois, and 2 new Clinton superdelegates in Iowa.
So, net Change: Clinton +7, O’Malley -3.
New “% of remaining needed to win chart”:
And the updated data chart:
As of my post-Iowa summary post, Sanders needed 54.30% of the remaining delegates to catch up. That is now up to 54.36%.
In terms of New Hampshire specifically, this doesn’t matter of course. Either way Sanders needs 14 of the 24 delegates that will be allocated based on the primary in order to be on a pace to catch Clinton.
Five days until New Hampshire.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
Yesterday’s update was preliminary, and I ended up using the delegate estimates provided by the AP which were used at the New York Times, Washington Post, and other places. My favorite source, Green Papers, had a different set of numbers. When I tried to calculate delegate numbers myself, I came up with yet another set of numbers. I ended up going with the AP numbers.
A little more than a day later and Green Papers updated, now matching what I had calculated myself. So I’m updating the estimates accordingly.
This moves us from:
8 Cruz, 7 Trump, 7 Rubio, 3 Carson, 1 Bush, 1 Paul
to
7 Cruz, 7 Trump, 6 Rubio, 3 Carson, 1 Bush, 1 Fiorina, 1 Kasich, 1 Paul.
The main change here is one less delegate each to Cruz and Rubio, with those delegates picked up at the bottom end by Fiorina and Kasich.
But wait! There is more! The Washington Post and Politico both add two additional delegates. The 27 allocated by the caucus results were already accounted for, so these would be two of the three “party leaders” who end up as automatic delegates. No indication of which party leader did what, but they add one additional delegate each to Cruz and Rubio.
The new Iowa totals become:
8 Cruz, 7 Rubio, 7 Trump, 3 Carson, 1 Bush, 1 Fiorina, 1 Kasich, 1 Paul.
Between these two changes, the net is simply adding the delegates for Fiorina and Kasich. This means 29 of the 30 Iowa delegates are now accounted for.
Note that even though Paul has “suspended his campaign”, he keeps his delegate unless he explicitly releases it.
The raw delegate count chart now looks like this:
Just a little uptick at the end for Fiorina and Kasich to join Bush and Paul at the 1 delegate mark.
Flipping this over to look at “% of remaining needed to win”:
On this chart, lower is better. When you get down to 0% you win. If you get to 100% you are eliminated. Everybody is still heading upward, not downward, but it is still very early. The whole field here ranges from needing 50.31% of the remaining delegates, to needing 50.59% of the remaining delegates.
Looking specifically at New Hampshire, there are 20 delegates at stake. We are so early, and the % needed numbers so close, that this is easy. The number needed to be on track to win is 11 delegates. Anything less, and the candidate is actually going to be in a harder position in the next contest rather than an easier one.
Let’s game out New Hampshire based on the current RCP averages. Yes, yes, this is based on polling before New Hampshire. Things will change in the next six days. But this is just to model how it works.
The rules via The Green Papers:
- 20 National Convention delegates are to be bound proportionally to presidential contenders based on the primary vote statewide.
- A 10% threshold is required in order for a presidential contender to be allocated National Convention delegates.
- Allocate delegates based on the 20 Г— candidate’s vote Г· total statewide vote. Round to the nearest whole number.
- Any delegate positions that remain open are awarded to the candidate with the highest statewide vote total.
Assume each candidate gets exactly their poll average today. Only the candidates over 10% matter, so with RCP as of right now that is:
33.4% Trump, 12.2% Cruz, 10.8% Rubio, 10.4% Kasich
Converting this to delegates with the rounding rules above, you get:
7 Trump, 2 Cruz, 2 Kasich, 2 Rubio
This is only 13, so there are 7 delegates left. According to the last rule, these all go to the leader, so we end up with:
14 Trump, 2 Cruz, 2 Kasich, 2 Rubio
And therefore Trump would get 70% of the delegates despite only getting 33.7% of the vote. Trump’s line on the “% needed to win” would go down, while everybody else would go up.
After this hypothetical New Hampshire result, the overall race would be:
21 Trump, 10 Cruz, 9 Rubio, 3 Carson, 3 Kasich, 1 Bush, 1 Paul, 1 Fiorina.
Trump would still only have 42.9% of the delegates, which is not a majority, but Cruz would be way back at 20.4% and you’d have Trump starting to pull away from the pack. Still only 1.98% of delegates allocated at that point though, with lots of room for further changes.
The RCP average on February 3rd is NOT going to look like the actual results on Primary night. This is just an example.
The points to take away:
- The detailed delegate rules matter, you don’t necessarily have to win a popular vote majority to win a delegate majority.
- We’re still so very early, one contest can have a major effect on the delegate totals, and each state is very different, so don’t put much stock in extrapolating Iowa to future contests.
Of course New Hampshire will be here soon enough, and we’ll do this with the real results.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
Unlike the Democrats, the Republicans didn’t have years of superdelegates endorsing one candidate or another, so Iowa really was a clean slate. The very first delegates earned in the race for the 2016 Republican nomination. The headlines are of course that Trump underperformed the poll-based expectations letting Cruz slide to a win, while Rubio outperformed his expectations, coming in a close third. This is all true.
But it is also true that only a tiny fraction of the delegates have been allocated so far. None of the candidates got a majority in Iowa either. So nobody is actually yet on a track to overall victory. Cruz may be slightly ahead, but Trump and Rubio are right at his heels. Cruz is not in a dominating position. It really is a three person race.
Looking at the raw delegate graph quickly…
…you immediately notice that we just have a really boring graph with a few straight lines. And Rubio’s line and Trump’s line completely overlap too. As do Bush and Paul’s lines. That’s because this is our first real data point after everyone was just at zero.
All of the graphs are like this. So just this once, lets skip the graphs and look at the summary table instead:
I’ll note here that Green Papers has slightly different numbers. But the New York Times, Washington Post, and CNN all agree with the estimates above, so for the moment that is what we’ll go with here.
It is clear that Cruz/Rubio/Trump are in a separate class here than Carson/Bush/Paul. But even these three would need to nearly double their performance to actually be on a track to win, because, well, you have to be getting more than half of the delegates to win outright. 25% to 30% won’t get you there. So at some point, somebody has to start going over the 50% mark if there is going to be a winner.
There is plenty of time for that of course. Nobody is predicting a contested convention quite yet, as much fun as that would be.
We probably won’t see anyone over 50% after New Hampshire either, but how close we get may depend on the viability thresholds. In New Hampshire on the Republican side you need to get 10% of the popular vote to be eligible for delegates. Although this will definitely change in response to the Iowa results, at the moment RCP has Trump at 33.2%, Cruz at 11.5%, Kasich at 11.5%, Bush at 10.3%, and Rubio at 9.5%. The rest are pretty far from 10%. But that is four candidates all within a couple percent of 10%.
Depending on which of those ends up right above or right below the 10% threshold, you could see very different delegate results coming out of New Hampshire. Most likely, we’ll have three or four candidates over 10% and a mix of delegates, but in the extreme case, if all but Trump drops under 10%, then Trump gets all the delegates. So a wide range of things is possible based on really small movements from the current positions in the polls.
Final points here though to the people saying Iowa proves that Trump is done…
- Trump was pretty close in Iowa. A one delegate difference means almost nothing.
- Although the Iowa results may diminish it, Trump starts with a big lead in New Hampshire.
- Only 1.09% of the Republican delegates have been determined so far.
We have a three way race right now. It is not inconceivable that a 4th candidate could do well enough in New Hampshire to join that group. Things are just as up in the air and chaotic on the Republican side as they were before Iowa. Anything could still happen.
Stay tuned.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
The Iowa caucuses have finally come and gone, and we have the first “earned” delegate results of the nomination race. Sorta. No actual Democratic delegates were actually determined. It was just the first step in a multi stage process. But as of now, estimates can be made. The estimates as per The Green Papers are 23 delegates for Clinton and 21 for Sanders. There were also 4 Iowa superdelegates already for Clinton.
So the Iowa totals as of this post are: 27 Clinton, 21 Sanders, 4 superdelegates TBD.
Even if all 4 remaining superdelegates went for Sanders, which is unlikely, Clinton still wins Iowa 27 to 25.
This despite the predominant reporting of tonight’s results as a tie or near-tie for Sanders.
In addition, since my last post on the Democrats in other states there have been 3 additional super delegates who came out for Clinton, and 2 who came out for Sanders.
So, where does that put us? First, the raw delegate totals:
You can see the “kink” in the chart that is the result of Clinton getting a much smaller percentage of the Iowa delegates than her absolute domination in the superdelegate race. This of course means her percentage of the delegates so far takes a bit of a swoon:
Clinton is down to only 91.55% of the delegates allocated so far! That’s the lowest she has been so far!
OK, that is down, but actually it is still pretty dominant.
But what does all this do to the critical “% of remaining delegates needed to win”?
When I made the last blog update, Sanders needed 54.22% of remaining delegates to win. Right before tonight’s new Iowa results, that had moved to 54.23%. Adding in the Iowa results it is now… 54.30%. So up a little, but really not by much. Sanders did not do well enough to bring this percentage down. It has to actually go down for Sanders to be on a pace to catch up and win. But he at least did well enough to keep it from getting very much worse than it was.
Looking at the New Hampshire polling, Sanders is currently at 55.5% in the RCP average and 55.2% in the Pollster average. Either way, that is a smidgen above 54.30%. So even if he gets no boost in New Hampshire from his near tie with Clinton in Iowa, Sanders may meet the bar he needs in order to improve his numbers on this metric. He doesn’t have much leeway though. If he drops at all, then he essentially will fall even further behind after New Hampshire.
The states beyond New Hampshire of course look more hostile to Sanders based on early polling. To make any real race out of this, something would need to happen that would cause Clinton to not just do a little worse, but a lot worse. Otherwise, the rest of this Democratic race is just going to be a process of Clinton mopping up, no matter what sort of media hubbub results after a potential Sanders win in New Hampshire.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
The Election Graphs delegate tracker is live! In the last few hours I have posted an intro to this effort, a tour of the new features, and a FAQ. But now it is time to get into the meat of things and start talking about results and what they mean.
While the Republican Party does have unbound delegates that will be able to make up their own minds independently, most of those people won’t be officially chosen until the primary/caucus season is well underway, and regardless, I haven’t yet found a good source of information tracking which of these folks is publicly supporting which Republican candidate. So the flow of delegate information for the Republicans won’t start until we get the results of the Iowa caucuses next week.
But on the Democratic side… most of the “superdelegates” are already known (although some may change before the convention). The superdelegates are Governors, Senators, Representatives, members of the Democratic National Committee, and “Distinguished Party Leaders”. As Democratic activists, these people very often are quite public about who they support. And they show up in news reports when they do that. And there are places tracking it.
Since this is the first analysis post of the primary season, I’m going to walk through the charts I use on my site, building up to the main one I believe is the one worth concentrating on going forward. If you just want to get straight to the punchline of how this race is going, skip to the last chart in the post and read from there.
Since June of 2013, slowly but surely, the superdelegates have been publicly choosing sides…
See the red line? That is Clinton. See the Green line? That is Sanders. Clinton is absolutely dominant here.
By my best count, 389 of the 713 superdelegates have expressed a preference, and 375 of those have come out for Clinton. Sanders has 11. O’Malley 3.
Now, the chart above is by date. It looks all jumpy because the delegates revealed their preferences at various random times, with a few big jumps when things happened, like Clinton officially announcing, or when AP did a survey of all the superdelegates. That can be interesting, but actually distorts the picture of how the race is coming along. Time to do a quick switch and show how this looks when we look at the % of delegates allocated instead of the date:
Suddenly, we have nearly a straight line. The lumpiness caused by the uneven distribution of delegate preference announcements over time is gone and we get a much clearer look at what is happening. We see that over 8% of the delegates to the convention have already expressed a preference. Clinton has been consistently racking these up as they come in. She is dominating in this race. But by how much? Lets switch to looking at the % of delegates each candidate has…
Once again we see that Clinton is dominant, but a few more details start to be seen. For instance we see that so far at least, Clinton’s weakest point was between the 4% and 5% marks when Biden, Sanders and O’Malley all had a few delegates in their columns and Clinton had about half the delegates she has today. But Biden dropped out, and Clinton kept on getting delegates at a pace not matched proportionately by Sanders or O’Malley, so Clinton’s dominant percentage increased again for awhile.
Very recently though, at the tail end of this chart, you can see that Sanders has been collecting delegates at a fast enough pace that his percentage has actually been going up, while Clinton’s has been decreasing. The Sanders percentage is still small, but it is going up. So where does that put Sanders? How about his surge? Can he win?
Time to look at the chart I consider to be the most important one to look at for understanding the race. Instead of looking at the percentage of delegates each candidate already has, we shift to looking at the percentage of the remaining delegates that the candidate would need to get in order to win the nomination:
As Clinton racks up delegates, it becomes harder and harder to catch up. With the lead she already has based on superdelegates, in order to win, Sanders doesn’t need to get 50% of the remaining delegates, he needs to get 54.22% of the remaining delegates.
Specifically looking at Iowa, 44 delegates are at stake.
To be clear, no Democratic delegates will actually be earned on the night of the Iowa caucus. It is the first step of a multi-stage process that won’t end until June. But after the precinct caucus results next week, we’ll be able to make estimates based on the initial vote results and what we know about the specific delegate allocation rules in Iowa.
With 44 delegates up for grabs, given the 54.22% number we calculated, for Sanders to actually put himself in a better position after Iowa than before, Sanders doesn’t need 22 or 23 delegates, he needs 24.
Now, if Sanders gets more delegates than Clinton, the media and press coverage about the Sanders surge and the risk to Clinton will be overwhelmingly loud. Nobody other than this site will be talking about how he won, but didn’t get to 24, so he is actually worse off. Everything will be about the Sanders win and his “momentum”.
And there is some fairness to this. If Sanders wins Iowa’s popular vote, no matter what the delegate estimates turn out to be, he will get a lot of positive attention. And people will talk about how Clinton is a lot weaker than she had seemed. That talk may improve Sanders’ position in New Hampshire or in other states further down the road. Perhaps it would boost him enough to compensate for the increase in the “% of remaining needed to win” that would result from falling short of 54.22% in Iowa. Getting a narrow win that doesn’t get enough delegates to actually improve the overall position in the short term may actually still have a positive impact in the medium to long run. Spin matters.
The polls are close in Iowa, either Clinton or Sanders could easily take the state. But the numbers to watch are not the popular vote totals. You shouldn’t even be looking simply at who is estimated to get the most delegates. The question is if the person who gets the most delegates gets enough to be on pace to win.
For someone other than Clinton to win, they have to start by catching up. This already means that they have to get more than 54% of delegates as we go forward. If they fail to reach that number, with each subsequent contest they will have to do even better to catch up.
If they do reach that bar though, each subsequent contest becomes easier. That’s the way it works. But the target is not getting the most delegates in each contest. It is getting more than the number you need to be on pace to win. Right now Sanders needs 54.22% of the delegates. Clinton only needs 45.90% of the delegates. She doesn’t even need to get a majority since she has already banked so many delegates before voting even starts.
So far, in the “invisible primary”, Clinton has scored an incredible 96.4% of the delegates. This will not continue. The Democratic leaders that make up the superdelegates have a very different perspective and background than Iowa caucus goers, or New Hampshire primary voters. Sanders will do better than the 2.8% of delegates he has managed so far. The question is how much better and if it puts him on track to actually catch up and win, or if it quickly becomes apparent that he is just playing the role of the protest candidate almost everyone originally expected him to be.
If Sanders doesn’t get to that 54.22% mark in Iowa, we may know pretty quickly that he is falling short, although if he wins in the popular vote, there will still be a crazy media circus for weeks.
If he exceeds that mark though, then the media howling might actually be warranted. At least until South Carolina. South Carolina is a very different state than either Iowa and New Hampshire, and the picture might change dramatically again after that.
But the next few weeks will be exciting almost any way the results go. And we haven’t even talked about the Republicans this time since there aren’t any delegates to look at quite yet. There will certainly be quite a bit of excitement on that side as well.
Tune in here for all the twists and turns as this race shifts into high gear!
Just under a week until Iowa. Hold on tight.
Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added.
[Edit 06:45 UTC to fix wording]
[Edit 08:09 UTC to update graphs]
[Edit 08:15 UTC to fix typo]
[Edit 2016-02-02 06:03 UTC to change “Delegate Race” in title to “Democrats” in preparation for separating posts on the two races]
|
|