The Delegate Fight: 2016

[Erc]

On the Democratic side in Wisconsin, the breakdown appears to be 48 - 38 in favor of Sanders.

This is precisely the margin Sanders needs in order to have a shot of catching up to Clinton in pledged delegates...the problem is that he now needs to replicate it in all the remaining states, including New York and Puerto Rico.

If Sanders is only hitting his average target in one of the best primary states left for him, it's not looking good for his hopes of winning the nomination without dirty tricks or a preposterous superdelegate super-miracle.

For comparison, his equivalent target in New York is a 140-116 victory.  Good luck with that.

[Erc]

New York (D): April 19

Overview
291 Delegates (6.10% of total)
Closed Primary
163 District
54 At-Large
30 PLEO At-Large
44 Superdelegates

Details

Groups of 54 and 30 delegates are apportioned based on the statewide vote.  The CD delegates are apportioned based on the results in each CD: 5 in CDs 11,19,22,23; 6 in CDs 1,2,4,5,6,8,9,10,12,13,15,16,17,18,21,24,25,27; 7 in CDs 3,7,14,20,26.

Superdelegates

Clinton (39): Bill Clinton, Gov. Andrew Cuomo, Sen. Chuck Schumer, Sen. Kirsten Gillibrand, Reps. Steve Israel, Kathleen Rice, Gregory Meeks, Grace Meng, Nydia Velázquez, Hakeem Jeffries, Yvette Clarke, Jerrold Nadler, Carolyn Maloney, Charlie Rangel, Joseph Crowley, José Serrano, Eliot Engel, Nita Lowey, Sean Maloney, Paul Tonko, Louise Slaughter, and Brian Higgins, Jay Jacobs, Sarah Kovner, Barbarlee Diamonstein Spielvogel, Herman Farrell, Stephanie Miner, Maria Cuomo Cole, Vice Chair Sheila Comar, Jennifer Cunningham, Emily Giske, Dennis Riera, Gerry Sweeney, Robert Zimmerman, Randi Weingarten, Ralph Dawson, Stuart Appelbaum, Hector Figueroa, Laphonza Butler

Uncommitted (5): Ex-Senate Majority Leader George Mitchell, Vivian Cook, Democratic National Convention CEO Leah Daughtry, Vacant (was Sheldon Silver), Vacant Chair (was David Paterson)

Useful Links
The Green Papers: NY
NY Delegate Selection Plan

[Erc]

Per request, posted my current projections for the remaining states on the tumblr.

Reposting them here.  Comments and critiques are very welcome.

Colorado: 28 Cruz - 3 Uncommitted
Results out of Colorado so far suggest a Cruz sweep, though a couple extra Uncommitted (beyond the 3 RNC members) is possible.

Wyoming: 14 Cruz - 3 Uncommitted
A Cruz sweep seems all but certain here, given the results in the County Conventions.  In fact, I’m so sure that I’ve already incorporated this projection into my current totals, though an extra uncommitted or two is always possible.

New York: Trump 84  - Cruz 4 - Kasich 7
Trump should break 50% statewide and in most CDs.  A fair amount of uncertainty here, as so much depends on the few registered Republicans in New York City.

Connecticut: Trump 18 - Cruz 3 - Kasich 7
Trump has a solid win, but not a sweep.  One of Kasich’s better targets of the remaining states.

Delaware: Trump 16
Unclear, but Trump seems favored to win this WTA state.

Maryland: Trump 29 - Cruz 3 - Kasich 6
Another unclear one, but Trump should benefit from a divided field. Cruz maybe picks up the panhandle CD, while Kasich takes a couple of DC metro CDs.

Pennsylvania: Trump 37 - Cruz 7 - Kasich 7 - Uncommitted 20
Only 17 pledged delegates technically at stake here, but a lot of the unpledged delegates have agreed to support the vote winner in their CD.  Trump seems likely to prevail at the moment over a divided field.

Rhode Island: Trump 10 - Cruz 3 - Kasich 6
Trump dominates, but proportionality keeps his delegate margins down.

Indiana: Trump 9 - Cruz 48
Cruz victory, though not a sweep.  If Trump can pull off the upset here, he likely clinches the nomination.

Nebraska: Cruz 36
Cruz wins.

West Virginia: Trump 34
There’s a possibility Trump doesn’t sweep here due to loophole primary issues, but it shouldn’t make a difference of more than one or two delegates.

Oregon: Trump 11 - Cruz 12 - Kasich 5
Proportional, with Cruz favored.

Washington: Trump 20 - Cruz 17 - Kasich 7
Proportional; unclear, but giving the slight edge to Trump.

California: Trump 121 - Cruz 45 - Kasich 6
The big Kahuna.  This can make or break the nomination.  This figure is for a sizeable (but not landslide) Trump victory, with Trump dominance in heavily Democratic districts.

Montana: Cruz 27
A Trump upset here would make a huge difference.

New Jersey: Trump 51
(Insert YUGE pun of choice)

New Mexico: Trump 10 - Cruz 10 - Kasich 4
Proportional.

South Dakota: Cruz 29
Seems like Cruz country.

Totals (including all other states):
Trump 1208
Cruz 813
Kasich 200
Other 141
Uncommitted 110

If those are the actual results, it gets hard to stop Trump on the first ballot.  He’d only need 29 more delegates, and the remaining Pennsylvania and insular unbound delegates should be enough to almost get him there.  Should be a really interesting convention!

[Erc]

Kasich's Prospects

Obviously, reaching 1237 on the first ballot is impossible for him, as, at this point, is even qualifying under Rule 40.

Continuing his campaign actively does net him more delegates (perhaps helping to deprive Trump of some), lets him continue any delegate selection efforts in the states that have already voted (though there is no indication that such efforts exist), and keeps him around as a potentially viable candidate after the first ballot.

In the meantime, though, what are his prospects?  Does his staying in the race help or hurt Trump?  Let's go through this state by state.

Colorado & Wyoming

He's a non-factor here.  He had zero delegate candidates in Wyoming on the County level, and does not have a full slate of delegates in Colorado.

New York

The game here is to keep Trump under 50% statewide and in as many CDs as possible.  Given that, Kasich staying in can only help, especially since he's a better fit for much of the state than Cruz.  It's possible a divided field could help Trump win a CD or two in odd places, but it's worth the risk.

Connecticut

At-Large is proportional, but CDs are winner take all.  Kasich is a much better fit than Cruz here (and actually has some small shot at winning the state); if anything, it's Cruz who should back down here.

Delaware

Hard to tell if either of the not Trumps have a chance here, to be honest.  Kasich not competing here likely wouldn't make a difference.

Maryland

Divided field helps Trump here, but I don't think Trump would lose the 1v1 fight here vs Cruz.  This is one of the friendlier states for Kasich in the remaining schedule, and he may be able to pick up a CD or two.

Pennsylvania

This one's a mess.  Polling is divided as to who the leading non-Trump candidate would be here, which means Trump likely wins the state.  I think Kasich has to make the play here, though, especially as I don't think Trump would lose to Cruz here.

Rhode Island

Proportional; no reason to stay out.

--May--

Indiana

Very naively a target for Kasich, but he should stay out and make sure Cruz beats Trump to a pulp.

West Virginia

No hope anyway.

Oregon & Washington

Proportional, so might as well make a play.

--June--

California

He could target a few selected CDs in the Bay Area, but should otherwise steer clear.

Montana & South Dakota

Stay away and let Cruz win these.  Seriously.

New Mexico

Proportional, so no harm in making a play here.

New Jersey

Abandon all hope.

TL;DR: Kasich should seriously campaign in all of the remaining April primaries.  After that, camp out in Oregon, Washington, New Mexico, and maybe the Bay Area and make no serious effort anywhere else, barring a major shift in the race.

[Erc]

I feel like a lot of the PA delegates might say right now that they will vote for the state's winner to get elected, but they wouldn't really do that.  Also, I think you're WAY too optimistic for Trump in California.  I think Kasich is falling more than you account for as well.  I just projected it on 538 and got 1122 for Trump.

Entirely possible on all counts.  PA and CA are the two major sources of uncertainty here.

We'll see what happens later this month!

In Pennsylvania, I see no real reason not to take them at their word; it would be quite something if Cruz were to essentially "steal" a state Trump won.  Not that that will necessarily stop anyone (Cruz will win Louisiana on delegates, after all), but going back on their word rather than just rigging the selection process a bit is a big step.

As for California, Trump's prospects might diminish if Kasich does indeed tank.  It should also be noted that Trump might still win the majority of delegates in CA even if he loses the state, if he dominates among the disproportionately-represented Democratic districts.  The figure now is a complete estimate; I may do a more detailed CA analysis later and come up with a more reasonable figure.

Also, when I have to make judgment calls, I'm going to make them in favor of Trump; it's best in my mind to have a clear view of the worst case scenario.

[Erc]

I've added a Delegate Selection Calendar post to the first page.

Over 600 delegates have been chosen so far; we're just getting into the delegate selection season as we speak.

Pretty much every Saturday from now until May 21 should feature a State Convention, and there are CD conventions liberally sprinkled through the rest of the calendar.

Note that, due to RNC deadlines, the last delegate selection process is also on June 7 (the CA & NJ primaries); many later states actually have their delegate selection before their delegate allocation!

[Erc]

Thinking ahead to multiple convention ballots, the governors of states and state party chairmen could have decent amounts of sway in their states' delegate slates. Here's who Governors have endorsed that may go against Trump:

Cruz:
Walker (WI)
Herbert (UT)
Haley (SC)
Bryant (MS)
Abbott (TX)

Rubio:
Martinez (NM)
Haslam (TN)
Hutchinson (AR)
Brownback (KS)

Kasich:
Otter (ID)
Bentley (AL)

Note that, unlike on the Democratic side, Republican governors aren't guaranteed to be in the delegation (although it is relatively frequent; I know Kasich and Phil Bryant were there in 2012, for just two examples).

So far, the governors that have been chosen as delegates are Haslam (TN), Dalrymple (ND), Daugaard (SD), Calvo (GU), and Torres (MP).  That said, relatively few states have finished their selection processes yet; the only states that have finished and do not have their Republican governor in their delegation are Alabama (good thing too, because that would have been embarrassing) and Ohio (for obvious reasons).

[Erc]

There are way fewer delegates really at stake in New York than people think.  I can't see Trump falling below 70 there, though of course the difference between 70 and 90 is nothing to be sneezed at.

California, obviously, is the main one.  Breaking even is not good enough, while a McCain-style landslide would most likely clinch him the nomination.

Pennsylvania is bigger than most people think, due to the promises many delegates have made to bind themselves to the winner of the district.  And the 17 At-Large delegates are nothing to sneeze at, either.  Even so, barring a win in Indiana or a knockout blow in California, Trump is going to need a fair share of those unpledged delegates in Pennsylvania; if he were to actually lose there, I don't see a real viable path forward for him.

Obviously, New Jersey is also a must-win, but if he's losing New Jersey he's obviously already lost the nomination.

[Erc]

You say that Trump will do well in heavily Dem CD's in CA, but the latest poll, has LA County as Cruz's strongest area by far, with a pretty substantial lead.

Really?  Well, that's a bad sign for Trump.  Hope they are polling actual registered Republicans.

Maryland: Trump 29 - Cruz 3 - Kasich 6
Another unclear one, but Trump should benefit from a divided field. Cruz maybe picks up the panhandle CD, while Kasich takes a couple of DC metro CDs.

Chuck Todd's team seemed to think Trump will do badly in MD because of how badly he did in DC. He did win VA though so I'm not sure what to make of it.

Trump did get destroyed in DC, as well as in Arlington (albeit VA was an open primary, which actually hurt Trump there, in a reversal of the usual pattern) and to a lesser extent Fairfax.  Hard to extrapolate though...good sign for Kasich in Montgomery County, but hard to tell beyond that.

Baltimore and environs seem like they'd be good for Trump, given the BLM activity in the last year.

[Erc]

So, let's talk about California a little bit.

Field Poll subamples suggests Cruz doing well in LA County and the Central Valley, running close in the Bay Area (small subsample), but getting demolished in "Other Southern California."

So, "Other Southern California."  This seems to be San Luis Obispo, Santa Barbara, Ventura, Kern, San Bernardino, Orange, Riverside, San Diego, and Imperial.

Normally, I'd imagine San Diego and Orange to be prime Cruz country.  Does this mean Trump is demolishing everywhere else, or do I have the wrong read on these counties?

[Erc]

(No sane person would include Kern in "other Southern California," but it says it's 9 counties, so it has to be.  Kern could easily be either strong Trump or strong Cruz, I'm honestly not sure.  My folks are ancestrally from Tulare County; I'm curious to see how that neck of the woods turns out.)

Bay Area:

CDs 2, 3, 5, 11, 12, 13, 14, 15, 17, 18, 19.

LA County:

CDs 33, 30, 29, 28, 27, 32, 34, 37, 43, 40, 44, 38, 47, 25.

Other Southern California:

24, 23, 8, 26, 31, 35, 39, 41, 42, 45, 46, 48, 49, 36, 50, 51, 52, 53

Central Valley/Sierras:

1, 4, 6, 7, 9, 10, 16, 21, 22

Even if Cruz just wins the 24 CDs in LA and "Central Valley/Sierras", it's probably good enough to deny Trump the nomination.  Obviously, there's going to be regional variation, but that number is a good benchmark.  100-72 is not a big enough win for Trump, though of course he'll try to spin it as such.

And, all in all, this does stress that Kasich shouldn't be making a play here beyond the City of San Francisco itself.

[Erc]

Kasich and Rule 40

It seems abundantly clear that Kasich, barring a complete shakeup of the race and insular shenanigans, isn't going to qualify under Rule 40.

Both the Trump and Cruz campaigns have expressed an unwillingness to change Rule 40, so it seems unlikely that it will.  If this is indeed the case, what happens to Kasich's delegates?

Actually, not much.  The ones that are bound to him are still, for the most part, bound to him even if he isn't placed into nomination.  They will be obliged to vote for him on the first ballot, but the Secretary of the Convention will not tally their votes.  If they try to vote for someone else, that support will also not be recognized by the Secretary of the Convention.

There are a few exceptions to this.

New Hampshire (4 delegates): They are released if he is not a candidate before the convention; that said, Kasich picked these delegates and they may still vote for him anyway.

Vermont (8 delegates): These are released if he doesn't have his name placed into nomination.  These delegates are chosen by the State Convention on May 21, so Kasich doesn't necessarily have their loyalty.

Indiana, Montana, and South Dakota: In the event Kasich wins any delegates in these states (would require winning the state outright in the latter two cases, or any CDs in the former), they will also be released before the first ballot.

[Erc]

ERC, I am running the numbers and I don't see trump getting below 1237 if he gets all 95 NY delegates, which is huge possibility. In fact, he could lose California and MD and still get 1237 if all 95 NY dels go to him.

My current projections have him winning both California and MD (though neither in landslide fashion) and 84/95 delegates in New York, and still falling short by over 50 delegates.

Are you giving Trump Indiana or Montana or something?

[Erc]

Reposting this from another thread:

Master thread someone is keeping for the Pennsylvania delegates:

https://docs.google.com/spreadsheets/d/1NfzyoWfwqjrYbc5Xqb9lyerxztNxLuZDzwPBarYZJFA/edit#gid=0

[Erc]

The Master PA sheet tells us, for the top 3 delegates on the ballot in each district:

Winner of District: 24
Uncommitted: 10
Cruz: 9
Trump: 5
Unknown: 4
Kasich: 1
Winner of State: 1

Note that the 1 Kasich delegate is uncontested, so Kasich will probably get at least 1 delegate out of Pennsylvania regardless.

[Erc]

Master Delegate Spreadsheet

I've started a viewable delegate spreadsheet, which features live updates of:

1) Summary delegate chart on the front page.

2) A list of all 2472 delegates

3) Trump Tetris

4) RNC Delegate Map

It can be found here.  Hope this helps!

[Erc]

There are a few exceptions to this.

New Hampshire (4 delegates): They are released if he is not a candidate before the convention; that said, Kasich picked these delegates and they may still vote for him anyway.

Vermont (8 delegates): These are released if he doesn't have his name placed into nomination.  These delegates are chosen by the State Convention on May 21, so Kasich doesn't necessarily have their loyalty.

Indiana, Montana, and South Dakota: In the event Kasich wins any delegates in these states (would require winning the state outright in the latter two cases, or any CDs in the former), they will also be released before the first ballot.

Isn’t Rule 40 actually kind of a problem for Cruz then?  Here’s my logic:

I’m assuming that Cruz’s strategy is to win on the second ballot.  Winning on the first ballot seems like it’ll most likely be impossible for him, since there are enough Kasich and Rubio delegates who are pledged to vote for them on the first ballot regardless of whether their names are placed into nomination or not.  As you said, those delegates will vote for Kasich or Rubio, and the Secretary of the Convention just won’t tally their votes.  But that’s OK for Cruz.  His goal on the first ballot is just to make sure that Trump doesn’t get a majority.  A delegate voting for Kasich or Rubio is just as good as one voting for Cruz in terms of denying Trump a majority.

So then, Cruz should be happy to see delegates pledged to Kasich and Rubio on the first ballot.  If any of them become free agents, then some could defect to Trump, and it could be enough to get him to 1237.  Therefore, if Rule 40 means that Kasich delegates in IN or VT (or Rubio delegates in other states) get released and can vote for anyone, then Cruz has some additional delegates who he needs to worry about—who he needs to make sure don’t vote for Trump.

I'd agree with that, personally.  He's got two worries to balance here.

1) Trump winning on the first ballot.  Allowing Rubio and Kasich to be placed on the ballot ties up around 43 delegates (depending on interpretation of certain state rules), that you don't have to worry about defecting to Trump (assuming neither Rubio nor Kasich withdraw or release their delegates).

2) Kasich (or Rubio) winning on some later ballot.  If Kasich doesn't technically appear on the first ballot, his delegates may be demoralized on the first ballot (or, perhaps, confused about the rules) and not vote for him on later ballots, instead defecting to Cruz on later ballots.  If Cruz intends to campaign to change (or more properly, reinterpret) the rules to prohibit further names from being placed in nomination after the first ballot, then this is a larger factor.

Of course, Cruz has a lot of time here to figure out his convention strategy.  If Trump is above 1200, 1) becomes a larger concern.  Otherwise, it's not and he can feel free to shut Kasich out if he likes.

[Erc]

That said, the bulk of the delegates in question (32/43) are from Oklahoma and Minnesota, where he will presumably control the convention process and can handpick delegates.  (Already, he has one of the Rubio delegates from Oklahoma locked down).  Vermont is another question, of course, and if it comes down to those 8 delegates he may want to ensure Kasich is on the first ballot.

[Erc]

If the rules of the convention are that only Trump or Cruz could be nominated could the convention become deadlocked with neither Trump or Cruz able to reach 1237?

In principle, sure.  Of course, a majority of the convention could always vote to change the rules if a majority of delegates actually favor a compromise candidate.

Outright Cruz victory on the 2nd or 3rd ballot seems more likely to me, though.

[Erc]

Excellent! I think that in your summary sheet, cell L60 is wrong. Shouldn't it be the sum from L42 to L57?

Old, duplicate cell; deleted now.  Thanks for the catch!

Very cool Erc.  Just one extra thing that I was thinking of that would be useful: In another post you just made, about how Cruz loyalists controlled the delegate selection in states like MN and OK, that kind of thing would give us a clue as to who some of the "uncommitted" might actually support, right?

That is, as I understand it, you count 155 delegates (so far) not pledged to any candidate, but 33 of them have endorsed one of the candidates, so that leaves 122 for whom we're nominally uncertain about their loyalties.  But can you break down which of them were chosen in states where the delegate selection was controlled by Cruz loyalists (or loyalists of some other candidate)?  Do we have that kind of info, to help us decode the likely loyalty of those 122 delegates?

The hard information I have is buried in the huge list of all the delegates.  Could be useful to break the unpledged delegates out into their own spreadsheet.

My guesses as to how Oklahoma and Minnesota will play out are based on (1) the results so far in OK [which are admittedly limited] and (2) my anecdotal evidence from the conventions I've been to here in MN.  Not all that much to go on, but I'll make a note of it.

[Erc]

http://www.desmoinesregister.com/story/news/politics/2016/04/07/gop-delegate-fight-coming-iowa/82746800/

Trump senior adviser Tana Goertz said the front-runner’s campaign is executing a “caucus-to-convention strategy” aimed at getting as many Trump backers to Cleveland as possible, although she declined to share details on how that effort was playing out in Iowa.

“We’re doing everything that every other campaign would do and more — I mean, we do work for Donald Trump,” Goertz said. “... We’re doing everything that any other campaign would be doing that’s ethically acceptable.”

As Patrick Svitek pointed out on twitter, that's a pretty sketchy quote.

Maybe it's trying to be a slam at Cruz and general ratery?  If so, didn't really come off that well.

In other news, Cruz picks up 3 more delegates in the CO-5 convention.  If he sweeps another CD, I'm calling the At-Large delegates for him as well (though we'll find out for sure tomorrow).

[Erc]

Cruz picks up 6 more delegates in Colorado.  With only one CD outstanding, I'm calling the At-Large delegation (another 13 delegates) for him as well.

One interesting thing to keep an eye on going forward...

Many states that can bind delegates on multiple ballots provide an automatic release mechanism if the candidate falls below X% of the votes on some ballot. In states where they exist, these range from 10% (California) to 35% (multiple states).

35% of the delegates is 866 delegates.  Trump will reach that number, though it's not certain that Cruz will; he will likely need a win in California or a significant number of unpledged delegates to get there.  Both are eminently possible, but neither are guaranteed.

[Erc]

Michigan chooses its delegates today and tomorrow at their State Convention.

Unlike many other states, they are taking a hard line against "ninja delegates"; it seems they are letting the campaigns vet their prospective delegates.

This is apparently due to a brouhaha during the 1988 campaign; these people have long memories.

That said, at least one Kasich delegate has said they will vote Cruz on the second ballot.

Generally, the Cruz folks seem to be better organized but are not officially trying to poach anyone else's delegates.  His slate of 14 CD delegates can be found on Twitter.  Everyone on his slate has won so far (I've found results for 5/14 CDs so far).

[Erc]

Erc, I'm wanting to make a map, and I was wondering if you could help me do it. 

I want to have a map showing who has won the delegates in a given congressional district so far.  By that, who they really support, not who they are bound to on the first ballot.

All of this would be for republicans of course.

My spreadsheet would probably be the place to go for that, though it's still very incomplete (delegate selection process is still early, and I haven't done much effort to look for people's true commitments).

Also, note that some states don't do the delegate selection process on a CD basis at all (New Hampshire, certainly, and perhaps some others).  And of course there's the whole set of At-Large and RNC delegates.  There are only around 1321 CD delegates, only a bit over half of the total.

[Erc]

Why are Kasich's VT, NH delegates released?

Rule 40 of the RNC requires a candidate to command the support of eight state delegations before having his name placed into nomination.  So far, he's at one, Ohio.  While it is technically possible that he will qualify (as there are well over seven states remaining, as well as some insular territories/North Dakota that he could in principle win), it seems highly unlikely at this point.

It also seems unlikely at this time that the rule will be changed; both Trump and Cruz have effectively spoken out against a rules change.

As a result, I feel it's safe to project that Kasich will not be placed into nomination on the first ballot.  This doesn't affect all that much; the vast majority of his delegates are still bound to him and will have to vote for him on the first ballot (though those votes will not be tallied by the Secretary of the Convention).  There are two exceptions.  In New Hampshire, delegates are pledged as long as their candidate is a "candidate before [the] convention"; while this isn't 100% clear, I'm interpreting this as meaning they are released in the event they are not placed into nomination.  It's a bit academic anyway, as Kasich got to pick his delegates in New Hampshire, so they will likely vote for him on the first ballot anyway.  In Vermont, delegates are released if they are not "placed into nomination," which is precisely the case in question.  Vermont's delegates are chosen at their state convention on May 21, and will not necessarily be loyal to Kasich.  As a result, they really will be free agents on the first ballot.

Of course, this is all assuming that there aren't any major rules changes before the first ballot; if Cruz realizes that he may desperately need Vermont's delegates to be bound to prevent a first-ballot Trump win, we may see Rule 40 relaxed.