Nate Silver on 12/10: 20% chance of a brokered convention

[Torie]

Now wait a minute. Cruz gets the nomination. Hillary is elected POTUS. The Pubs lose the Senate, and lose 20 seats in the House, so their margin is down to 10 congresspersons to pick off. The filibuster is nixed. So now we are down to 10, and the job gets done! Or wait a minute. Many Dems have been bought off by the banks - including Hillary. I guess the Pub establishment is un-punishable then. Well we still have the estate tax. It goes up to 50% if the banks are reined in. If not, it goes up to 70%. Maybe that little compromise will appeal to them. What do you think?

If Hilary Clinton is president in 2018, there's no way the Democrats take over either chamber.  The GOP knows this.  And if the GOP participates in sweeping regulations AND a tax increase, every single representative responsible will be primaried to hell and gone.  The GOP knows this.

The TP wing won't care if Cruz loses in the general.  They'll just double down.  "Cruz wasn't conservative enough to excite the vast, silent majority of right-thinking Americans" will be the narrative.

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

[Beefalow and the Consumer]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

[emailking]

State polls have consistently placed Trump in the lead.  A huge lead.  For months. 

That's totally irrelevant to his point which was that early polls are a bad indicator. They don't become a good indicator because someone is leading big for a long time.

But he's forgotten that he's dealing with very small sample sizes here

Really? As many times as he has caveated sample size in the past, you really think he's plumb forgotten about them right now?

[Slander and/or Libel]

State polls have consistently placed Trump in the lead.  A huge lead.  For months. 

That's totally irrelevant to his point which was that early polls are a bad indicator. They don't become a good indicator because someone is leading big for a long time.

But he's forgotten that he's dealing with very small sample sizes here

Really? As many times as he has caveated sample size in the past, you really think he's plumb forgotten about them right now?

He talks about it a lot, sure. But his confidence in Trump's pending disintegration would indicate that he hasn't taken it fully to heart.

[Attorney General, Senator-Elect, & Former PPT Dwarven Dragon]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

Florida or Ohio. Of course, the Republicans can counter that by picking up Reid's seat in Nevada. CO/WA probably only flip in a wave.

[Torie]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

Oh, it is quite possible under a ticket led by Cruz or Trump. First, Murray is safe no matter who the Pub POTUS candidate is, and Bennet would be close to safe in this scenario. Heck he does not yet have a creditable opponent.  And then there is  ...  drum roll please - New Hampshire! I will leave NH to TNVolunteer to help you with that one. He specializes in NH, and while he is idiosyncratic on the matter, he good enough for government work as it pertains here. And then there is Florida, which is marked as a tossup seat at present. Assuming Grayson is not nominated, that one goes down. So now we are up to 51. Hillary's VP is not even needed.  That is about where the music stops. Burr is not going down in NC. The Dem trend there will be muted anyway. It is not part of the "when the roof falls in" zone. Oh, Ohio actually is where the music actually stops for the Dems, at 52 seats. I don't want to think about that one. Portman should be exempt from the punishment regime. He's special.

[MT Treasurer]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

1. Murray isn't going to lose, forget about it.
2. Ayotte is more likely to lose than not.
3. Katie McGinty isn't going to defeat Toomey.
4. Florida is a pure Tossup.
5. Nevada will be competitive as well.

[Torie]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

1. Murray isn't going to lose, forget about it.
2. Ayotte is more likely to lose than not.
3. Katie McGinty isn't going to defeat Toomey.
4. Florida is a pure Tossup.
5. Nevada will be competitive as well.

This idle chatter presupposes a Pub ticket led by Trump or Cruz.

[MT Treasurer]

Ah, I was thinking about 2016 myself. And the House will not flip. Thus my comment about a 20 seat loss. That is possible - maybe.

The Senate won't flip in 2016 either, unless something really drastic changes.  If Feingold, Duckworth, and Sestak/McGinty all win, that gets them to 49.  They still need to pick up another seat (from where?) and hope that Michael Bennet and Patti Murray can hold on.

1. Murray isn't going to lose, forget about it.
2. Ayotte is more likely to lose than not.
3. Katie McGinty isn't going to defeat Toomey.
4. Florida is a pure Tossup.
5. Nevada will be competitive as well.

This idle chatter presupposes a Pub ticket led by Trump or Cruz.

I think Portman and Toomey are smart enough to distance themselves from Trump/Cruz in a way that doesn't alienate their base but also helps them with swing/crossover voters. And Nevada is well known for bucking the trend, too (2000: GOP, 2004: DEM, 2006: GOP, 2010: DEM, 2012: GOP). Coattails are overrated, these individual Senate races are very hard to predict. Unless Trump (whose goal is to take down the entire party with him) is the nominee, Republicans don't have that much to fear. And even then, Democrats won't pick up more than 6 seats in the Senate and their new majority wouldn't last longer than 2 years.

[tallguy23]

Portman is safer than Toomey. Ayotte will have a hard time beating Hassan.

I really think Duckworth will unseat Kirk. She just checks so many boxes (female, veteran, disabled) and is a great candidate.

[Mr. Morden]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

[Mr. Morden]

Anyway, I think 20% is too high, but whatever the number is, I would put it higher this year than it has been for a long time.  And this is largely because of the wildcard of Trump.  I'm not sure to what extent "momentum" is going to be determinative of his support this time around, either positive or negative.  If he wins early, does his support build much higher?  If he loses early, does the bottom fall out?  I'm not sure.  He may be a candidate who is more impervious to momentum, if voters have a polarizing opinion of him.

I mean, consider 2004.  I don't think that, in the end, the Dem. primary voters felt all that strongly about which candidate was going to be their guy.  And so, Kerry unexpectedly wins Iowa, and you get the mother of all bandwagons.  But it doesn't always work like that.  In the 2008 Dem. race, even though both Clinton and Obama were popular among primary voters, opinions of which of them should be president were pretty polarized among different demographic groups within the Democratic electorate.  So "momentum" got trumped (ha ha) by demographics.  Does that also happen with Trump this time around?  Maybe the field consolidates but Trump is unmoved from 30 or 35% or whatever, and he ends up splitting victories and delegates with two other candidates.

Or maybe not.  I guess we'll find out soon enough.

[mencken]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

Surely a candidate who has polled a plurality with ~30% of the voters for six consecutive months makes the party's decision a bit harder to enforce? And should not the consistent failure of any of the establishment-friendly candidates to gain traction make it less likely that they will prevail in the end? (In other words, could Trump or Cruz simply beat out the clock?)

[Mr. Morden]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

Surely a candidate who has polled a plurality with ~30% of the voters for six consecutive months makes the party's decision a bit harder to enforce? And should not the consistent failure of any of the establishment-friendly candidates to gain traction make it less likely that they will prevail in the end? (In other words, could Trump or Cruz simply beat out the clock?)

I'm not saying that I'm fully on board with the "party decides" crowd.  Just explaining that they haven't (yet) been proved wrong, because no one's voted yet.  Whether they're right to ignore the polling, I don't know.  I'm just saying that if they say "don't listen to the polls, Trump won't win", you can't then say "but look at the polls, Trump will win".  That doesn't make sense.  The polls being a certain way doesn't in itself prove that the polls will be predictive, when the thing that we're trying to predict hasn't happened yet.

[Attorney General, Senator-Elect, & Former PPT Dwarven Dragon]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

Surely a candidate who has polled a plurality with ~30% of the voters for six consecutive months makes the party's decision a bit harder to enforce? And should not the consistent failure of any of the establishment-friendly candidates to gain traction make it less likely that they will prevail in the end? (In other words, could Trump or Cruz simply beat out the clock?)

But at the convention, you need a majority of delegates to be nominated. You can't do it with 30 or 40 percent of the delegates.

Assuming that neither Trump nor Cruz implodes, the establishment is so opposed to Trump/Cruz that they will take every measure to make sure Rubio/Christie/Kasich stays in the race till the end along with Trump and Cruz, to prevent either one of them from getting a majority of delegates. At the convention, delegates are released from pledges to support a certain candidate after the first ballot, and the expectation would be that Cruz/Trump delegates would quickly switch to supporting the "establishment survivor". Of course, if this backfires and results in Cruz delegates voting en masse for Trump to get him over the top (or vice versa), then the establishment has to live with a nominee they hate.

[mencken]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

Surely a candidate who has polled a plurality with ~30% of the voters for six consecutive months makes the party's decision a bit harder to enforce? And should not the consistent failure of any of the establishment-friendly candidates to gain traction make it less likely that they will prevail in the end? (In other words, could Trump or Cruz simply beat out the clock?)

But at the convention, you need a majority of delegates to be nominated. You can't do it with 30 or 40 percent of the delegates.

Assuming that neither Trump nor Cruz implodes, the establishment is so opposed to Trump/Cruz that they will take every measure to make sure Rubio/Christie/Kasich stays in the race till the end along with Trump and Cruz, to prevent either one of them from getting a majority of delegates. At the convention, delegates are released from pledges to support a certain candidate after the first ballot, and the expectation would be that Cruz/Trump delegates would quickly switch to supporting the "establishment survivor". Of course, if this backfires and results in Cruz delegates voting en masse for Trump to get him over the top (or vice versa), then the establishment has to live with a nominee they hate.

For one, that is Trump alone that pulls a consistent ~30%. All anti-establishment candidates combined have been in the 50-60% range throughout that time. Second, eventually the contests are going to get to be winner-take-all, at least at the district level. The establishment candidates will take on heavy losses unless one of their candidates can rise to a level where they can consistently beat both Trump and Cruz (as of right now, Rubio would be third in a three-way contest).

[Attorney General, Senator-Elect, & Former PPT Dwarven Dragon]

Technically, the establishment survivor doesn't need to get many delegates in May/June - all that needs to happen is Trump and Cruz split them fairly evenly. The establishment survivor simply needs to stay in the race. What the establishment would be aiming for is something like this on the first convention ballot:

Trump 1,050
Cruz 960
Establishment Survivor 430
Abstain/Scattering 26

The number needed to be nominated this year, per the green papers, is 1,233. No one is there, so no one can be nominated via the first ballot. Under party rules, delegates are freed from pledges to support a candidate and re-vote on the nomination. The hope would be that a sizable number of Cruz/Trump delegates don't actually support them and only voted that way on the first ballot because they had to under party rules, and would therefore gravitate to Rubio (etc.) and get him nominated. Of course, Trump would be furious and blame the party for not giving him reliable delegates, but he would have no legal recourse under the party rules and it would be too late to mount a meaningful independent run.

[Mr. Morden]

The hope would be that a sizable number of Cruz/Trump delegates don't actually support them and only voted that way on the first ballot because they had to under party rules, and would therefore gravitate to Rubio (etc.) and get him nominated.

This is a point which'll get more attention if we still have three or more viable candidates in a few months: How the delegates are selected.  As far as I can tell, the rules vary quite a bit from one state to the next.  In some states, the presidential candidate can hand pick his delegate slate, and other states he can't.  The degree to which delegates are "bound" to vote for the candidates that they've pledged to vote for (even on the first ballot) also varies from state to state.  This is why whenever you have the running delegate counts during the primary season, every media organization ends up giving a slightly different number, because how you count it is messy.  This mess always gets swept under the carpet, because one candidate ends up as the de facto nominee in advance of the convention anyway, but if there was a contested convention scenario, you'd have a giant mess.

E.g., the Atlas actually lists a whopping 29% of 2012 GOP delegates "unallocated":

https://uselectionatlas.org/RESULTS/national.php?f=1&year=2012&elect=2

I think that exaggerates reality, and the RNC has tightened the rules on caucuses now, but in addition to the delegates who are definitively allocated to the individual candidates, there's also a block of delegates who will either be only weakly assigned to a candidate or not assigned at all.

[Slander and/or Libel]

I think it would inspire massive unrest among party voters if they gave most of their votes to Trump/Cruz yet at the convention the establishment orchestrated the nomination of Rubio/Christie/Kasich.

[Landslide Lyndon]

But if we go by who predicted TRUMP's "imminent demise" and who didn't, then the only pundits left will be Paul Krugman, Michael Tomasky and kos.

I don't think that's true at all.  The "party decides" fundamentalists like Jonathan Bernstein haven't predicted Trump's imminent demise.  They've just not aimed to predict the short term polling trends at all, because they consider them irrelevant.  Their perspective is that the party elite will get a candidate they like in the end, and who cares who's leading in the polls right now, or tomorrow?  (I think Silver has basically been in this category as well.  Did he ever predict that Trump would have an "imminent demise" in the polls?  I don't think he has, but if someone has a quote from him to that effect, please post it.  I think, like Bernstein, he doesn't get too worked up about who's going to be leading in such-and-such a state next week.)

Now, maybe the "party decides" people are going to be proven wrong in an epic way.  But let's not get ahead of ourselves, and declare them wrong yet.  No one's voted yet.  The GOP doesn't yet have a nominee.  Let's just wait and see whether they're right or wrong.

I'm talking more about the people who stubbornly refused, despite all evidence to the contrary, to take TRUMP seriously as a candidate and confidently predicted his collapse after he disparaged McCain's service, after he insulted Megyn Kelly, after he feuded with FOX News, after the Paris attacks, etc.
To say even now that TRUMP won't be the nominee is a not unreasonable assumption. But to say that he has no chance, there is some serious delusion and stubbornness there.   

[SillyAmerican]

Again Silver is using data, he believes in the 'party decides' theory that the best early indicators are endorsements (from currently elected politicians) and Trump has zero. This cycle will test the 'party decides' theory. Silver is also right that early polling, especially early national polling has not been a good indicator. However we are now transitioning into the time when it isn't early polling and Trump still has a lead in NH and is in second in IA, so history is slowly but surely moving in Trump's direction to at least be a serious contender.

Agreed. And I think that any belief in the notion of 'party decides' will be soundly squashed this time around. Just a gut feeling...

[Slander and/or Libel]

Can someone explain how proponents of the "party decides" theory think that there will not be a revolt by GOP voters if the party nominates someone who came in third or fourth place in the delegate count by virtue of votes cast in primaries?

[Torie]

Technically, the establishment survivor doesn't need to get many delegates in May/June - all that needs to happen is Trump and Cruz split them fairly evenly. The establishment survivor simply needs to stay in the race. What the establishment would be aiming for is something like this on the first convention ballot:

Trump 1,050
Cruz 960
Establishment Survivor 430
Abstain/Scattering 26

The number needed to be nominated this year, per the green papers, is 1,233. No one is there, so no one can be nominated via the first ballot. Under party rules, delegates are freed from pledges to support a candidate and re-vote on the nomination. The hope would be that a sizable number of Cruz/Trump delegates don't actually support them and only voted that way on the first ballot because they had to under party rules, and would therefore gravitate to Rubio (etc.) and get him nominated. Of course, Trump would be furious and blame the party for not giving him reliable delegates, but he would have no legal recourse under the party rules and it would be too late to mount a meaningful independent run.

Even assuming that the Cruz and Trump delegations have a lot of Quislings in their ranks, if it appeared that the Pub establishment was going to poach the Quislings, and successfully give the nomination to the Establishment choice, one would think that Cruz and Trump would cut a deal, and one of the two would release his delegates before the first ballot, giving the nomination to the other. And then presumably the other would get the VP slot. Morden etc. is right that in your scenario, if the Establishment did what it did, the nomination would probably become worthless. It would look like the will of the people had been ignored. It won't be happening. Alternatively, the establishment might pick Cruz.

[Mr. Morden]

Sure if one candidate gets a clear majority of votes and delegates from the primaries, then the nomination isn't going to be stolen from them.  But I'm talking about more ambiguous cases.  Say Trump has won a 45% plurality of all votes in the primaries, but whether he has a majority of delegates is a matter of dispute, because of the fuzzy math I talked about in terms of unbound vs. unpledged etc. that seems to create all these disagreements in the tracking of the delegate count every four years.  There'll then be various attempts to create mischief with the convention in a way that there wouldn't have been if all of the candidates were broadly acceptable to the party leadership.

That's what I mean about having to pay attention to how the delegates are actually chosen.  In a true contested convention scenario, the delegates are free agents.  They're not necessarily going to do the bidding of the candidate who they're "supposed" to back.

[mencken]

Are not all these predictions dependent on fair play by the Quisling forces? They prevented Paul-friendly delegations from Maine and Louisiana from being seated last time, why wouldn't they perform similar shenanigans against Trump or Cruz this time? The Quislings' only incentive to play fair is bad PR resulting in a general election loss, but a good many of them might prefer a certain Clinton presidency to a small chance of either Trump or Cruz winning. The Trump/Cruz forces better be prepared to take over a good chuck of the state party apparatuses to make sure that the rightful nominee gets on the ballot in November, rather than someone from a smoke-filled room.