Why Trump could lose the popular vote but win the electoral college vote

[RR1997]

There's a very high possibility that Trump loses the popular vote, but wins the electoral college.

First of all, Trump has a clear disadvantage in terms of popular vote. Hispanics and Asians are going to turnout at record levels, which will hurt Donald Trump in terms of PV.

Trump is going to severely underperform in western states like Utah and Arizona. The fact that he's probably going to underperform in western states makes it very hard for him to win the popular vote.

http://www.realclearpolitics.com/widgets/2016/turnout_two_party_vote.html

If we assume that Trump gets 20% of the Hispanic vote with a 54% turnout, 23% of the Asian vote with 52% turnout, and 9% of the black vote with a 63% turnout, he will need 65% of the white vote with a 67% turnout in order to barely win he popular vote. Romney got 59% of the white vote and Reagan got 66% of the white vote during 1984 landslide.

I might be underestimating Donald Trump with Hispanic voters though. Most polls have shown him getting the same percentage of the Hispanic vote as Romney got. Some polls have shown Trump doing better than Romney with Hispanic voters. Romney got 27%. You have to remember that Romney hit a floor with Hispanic voters in 2012.

McCain and Bush were relatively liberal on immigration, which is why they got 35% and 44% of the Hispanic vote respectively. Romney was very right-wing on immigration , which is why he only 27% of the Hispanic vote. There are around 20% of Hispanic who'll vote Republican no matter what. These are Hispanics who actually agree with the Republicans on immigration. These are Hispanics who hate illegal immigrants that take the easy route. These are completely assimilated Hispanics who live in heavily Republican areas and vote similarly to their peers. There is no way Trump does worse than 20% of the Hispanic vote. 25% is his ceiling though. Trump will obviously do worse than Romney with Hispanics, but I wouldn't be surprised at all if he only did slightly worse. I think Donald Trump will only do slightly worse than Romney in terms of percentage of Hispanics, but turnout will soar. I think that Hispanic turnout will soar to 54% (compared to 48% in 2012). This largely hurts him in terms of popular vote.

Even if we assume that Trump gets 25% of Hispanics/Asians (with a 54% and 52% turnout respectively) and 10% of blacks (with a 63% turnout), he will still need 64% of the white vote in order to barely win the popular vote (with a 67% turnout).

However, he doesn't need 65% of the white vote to win the electoral college vote.

The problem with the RCP calculator is that the universal swing system is not accurate.

I'm going to quote Wulfric and TNVolunteer here (since I'm too lazy to explain it with my own words):

The idea is that TRUMP will turnout a highly unusual selection of the white population, just like he has in the primaries, that loves him to death. It will be disproportionately in rustbelt states (PA, MI, OH, WI), so a caculator like Nate's that uses a uniform swing system is not accurate. Romney plus those four states equals 270 exactly, and Trump says he'll figure out a way to win FL for good measure.

I'm not saying I have any faith in it all or hope it's correct. But Trump outlived literally everyone's expectations in the primaries, so who really knows for sure about the general. Hillary can't just sleep until the election once Sanders is under lock and key.

lol

You realize that Trump has been in leading Clinton in most Florida polls, right? He can definitely win FL. And then OH and PA will do it. He could win the EC but lose the PV.

I also have a hard time believing that Trump won't try everything to get that Hispanic number to at least Romney 2012 levels (27%). He knows he can't win with those abysmal numbers among non-Whites. One other thing: This calculator assumes an uniform swing among White voters, which isn't likely to happen. Trump will almost certainly do better with Whites in OH than in CO or maybe VA. 62%/63% may sound like much when you think about it, but Republicans have much room to improve with Whites in the North. That said, I have no doubt that Trump would lose the election if it were held today.

Trump has an unfair advantage in terms of electoral college votes.

Trump is the perfect fit for whites in Florida. Florida is probably going to have more Obama 2012-Donald Trump 2016 voters than any other state in the country. The whites in Florida are typically staunchly anti-immigration and will love Trump's populist message. Romney hit a floor with Hispanics in 2012, so I can't see Trump doing that much worse with Hispanics in FL.

Many polls have shown Trump in the lead (or close to winning) in Florida.

Same thing with Ohio and PA. Donald Trump is a perfect fit for working-class Rust Belt whites. A lot of anti-free trade and immigration white voters in these two states. Trump is also going to do slightly better with blacks than Romney did.

Many polls have shown Donald Trump in the lead (or close to winning) OH/PA.

For example, PPP (a very reliable pollster) recently conducted polls in PA and FL. According to their polls, Donald Trump is tied with Hillary in PA and is beating her in FL, despite the fact that he's trailing her nationwide.

Despite the fact that Trump is losing to Hillary by a very large margin in national polls, he's still beating (or close to beating Hillary) in FL/OH/PA.

All he needs is FL/OH/PA+Romney 2012 states to win (and winning these three states doesn't require winning the popular vote).

Don't be surprised at all if Trump wins the electoral college but loses the popular vote.

[Meclazine for Israel]

Ohio, Pennsylvania and Florida are must win's for Trump.

Even before the devastating events in Florida, your analysis looks sound.

There is plenty of time remaining for him to get other states from the Democrats as well.

I think we will see some interesting changes in the polls between now and November.

[Panhandle Progressive]

Starting with the 2012 election results, Trump could win back Florida, Ohio, Iowa, and New Hampshire and still lose. Think about that. Virginia really is the key this cycle. If Hillary can hold it she is almost guaranteed to become president. This is why I believe she will select a Virginian for her running mate.

[Xing]

It's plausible, but I don't think there's a high chance that he magically does well in OH/PA/FL while doing poorly nearly everywhere else. Flipping PA will be a much taller order than a lot of people here seem to realize. If he relies mostly on blue collar white men, he has to win them by about 15 points more tha Romney did. And Romney honestly didn't do that badly among them. Not to mention, Trump could easily do worse than Romney among women and educated voters in the state. Trump is probably only flipping PA if he's enjoying quite a bit of success nationally. I would not be quick to assume that PA will be more Republican than the nation as a whole.

[RaphaelDLG]

It's more likely Hillary wins the EC but loses the PV.  Both possibilities are non -negligable, but both are very unlikely

[RR1997]

It's more likely Hillary wins the EC but loses the PV.  Both possibilities are non -negligable, but both are very unlikely

No.

Hillary winning the EC but losing the PV is nearly impossible. Donald Trump winning the EC but losing the PV is somewhat likely.

It's more likely Hillary wins the EC but loses the PV. 

How? No way Trump loses FL, OH, PA if he wins the popular vote.

^^

[ProgCon]

It's funny you bring this up considering I was playing with the demographic numbers at fiverhityeight and one of the reasonable conclusions I came to is that Trump could likely lose the popular vote, but win the electoral college. I generated this result by adjusting the numbers to the following:

College Educated White: 54% Republican, 83% Turnout
Non-College Educated White: 69% Republican, 59% Turnout
Black: 89% Democrat, 59% Turnout
Hispanic/Latino: 76% Democrat, 55% Turnout
Asian/Other: 71% Democrat, 54% Turnout

Popular Vote Result: 49.3% Clinton (D), 48.9% Trump (R), 1.8% Others
Electoral Vote Result: 273 Trump (R), 265 Clinton (D)

The map would look like this:

[RaphaelDLG]

If the swing is uniform from 2012, Trump could win Romney plus OH VA FL and still lose.

Your scenario assumes Trump tanks severely with Hispanics and stays that way but white voters swing massively.  I don't think Trump wins without improving a little bit with Hispanics.

Let's say Trump loses five points with Hispanics and gains with everyone else and there are slight fluctuations in turnout.  He wins 50-50 barely 270-268 with Romney plus VA/OH/NH/FL.

Trump winning the rust belt is an overrated meme.

Maybe you're right and at this point for this candidate the Republicans have a slight edge in the EC.  But either way what I said about either result being pretty unlikely is true I think.

[Xing]

It's more likely Hillary wins the EC but loses the PV.  

How? No way Trump loses FL, OH, PA if he wins the popular vote.

He obviously wouldn't lose all three, but he could certainly lose one of them while winning the popular vote.

[RaphaelDLG]

Also let's put it this way, I may very well be wrong, but I don't think the trends have moved WI, VA, and NH to the point that, in 2016, Trump is more likely to win WI than the other two.  Not yet.  I apologize because I know VA and NH are considered safe blue around here

[RaphaelDLG]

I do respect your analysis, though, it reflects intelligent thought.  I think Trump is going to have a money disadvantage and the Dems will be POURING money into PA if the race is close.

[IceSpear]

Why would Trump be guaranteed to do better in Florida? Cubans hate him and Romney absolutely demolished Obama among whites, so I highly doubt there's some sleeping giant of Obama/Trump white voters that you describe.

[MT Treasurer]

It's more likely Hillary wins the EC but loses the PV.  Both possibilities are non -negligable, but both are very unlikely

No.

Hillary winning the EC but losing the PV is nearly impossible. Donald Trump winning the EC but losing the PV is somewhat likely.

It's more likely Hillary wins the EC but loses the PV.  

How? No way Trump loses FL, OH, PA if he wins the popular vote.

^^

Well, congrats. People here really underestimated the influence CA and TX have on the popular vote.

[elcorazon]

down the stretch 538 had chances of this result at around 10-11%. I was quite worried about it in advance. Sad that it happened.

[Beet]

Wow, this thread was very prescient.

[Fuzzy Bear]

Why would Trump be guaranteed to do better in Florida? Cubans hate him and Romney absolutely demolished Obama among whites, so I highly doubt there's some sleeping giant of Obama/Trump white voters that you describe.

I was happy to be part of this sleeping giant.

[Mr. Smith]

It's funny you bring this up considering I was playing with the demographic numbers at fiverhityeight and one of the reasonable conclusions I came to is that Trump could likely lose the popular vote, but win the electoral college. I generated this result by adjusting the numbers to the following:

College Educated White: 54% Republican, 83% Turnout
Non-College Educated White: 69% Republican, 59% Turnout
Black: 89% Democrat, 59% Turnout
Hispanic/Latino: 76% Democrat, 55% Turnout
Asian/Other: 71% Democrat, 54% Turnout

Popular Vote Result: 49.3% Clinton (D), 48.9% Trump (R), 1.8% Others
Electoral Vote Result: 273 Trump (R), 265 Clinton (D)

The map would look like this:

That was my map in the final week, only after Comeyghazi.

And, that only differences here are MI/PA  over NH and the UT result.

...Scary.

[Meclazine for Israel]

This was spot on the money.

I was glad i was part of this thread.

I love the celebrating Hillary signatures of members who left after election night.