The States: Prediction map

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This map will track the predictions I make, to give you a visual flavour of Democrat/Republican areas of the country.

If you prefer your information delivered to you in a visual format, then you’ve come to the right place. With every new article in the series, I will update this map with which party I think will win the state, as well as a running Electoral College total for each party.

A couple of things to remember:

  1. A party/candidate needs 270 Electoral College votes to win the presidency.
  2. Red states are predicted Republican wins, blue states are predicted Democrat wins. White states are still waiting to be written about!
  3. Electoral College votes are winner-takes-all.
  4. Maine (ME) and Nebraska (NE) split how their Electoral College votes are won. They both award two Electoral College votes to the winner of the statewide popular vote, then apportion others (three more for Nebraska, two more for Maine) for the winners of the popular vote in each of the state’s House of Representative districts.
  5. In the event of a 269-269 Electoral College tie (which could be possible in 2020!), the House of Representatives decides which candidate will become president and the Senate decides who will become vice-president. In that circumstance, it’s perfectly possible to imagine a Democratic president and Republican vice-president (or vice-versa).

Running Electoral College prediction

Most up to date prediction map

Republicans (incumbents): 74

Democrats: 81

Winner needs 270

The States #12: Hawaii

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Hawaii (HI).

Location

Hawaii is a state with many unique attributes! It is the newest state in the entire Union, having been granted statehood in 1959. It is also the state furthest away from the mainland United States, sitting some 2,500 miles off the California coast. Finally, it is the only state to have a foreign nation’s flag as part of its state flag – the Union flag is present thanks to a strong historic relationship between the United Kingdom and Hawaii (before it was absorbed into the United States)

In the fifteen elections Hawaii has participated in, it has voted Democratic in thirteen, only voting Republican for Richard Nixon in 1972 and Ronald Reagan in 1984.

There are four Electoral College votes up for grabs in Hawaii.

State voting history

State history

Hawaii has been comfortably Democratic since the turn of the millennium, with the smallest margin of victory recorded at just over eight per cent in 2004. Barack Obama, perhaps Hawaii’s most famous son, scored particularly well in the state, as is to probably be expected given he is one of their own. Even ten per cent of third-party voters couldn’t prevent Hillary Clinton from amassing a victory in Hawaii by a margin of over thirty per cent.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

For the first time, I have included Native Hawaiians as a racial category from the census. In every state so far, apart from Hawaii, Native Hawaiians have registered at less than 1% of a state’s population. However, given that we are now in the state of Hawaii, it seems appropriate to introduce the new racial category. Hawaii is a fantastically diverse state, and is also a majority-minority state, with a very small proportion of white people. Asian-Americans are the majority in the state, with a sizeable proportion of Native Hawaiians and Hispanics/Latinos too.

Gender

State gender

The gender breakdown of Hawaii is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 5-17 category is worthy of discussion. Anyone aged eight or over at the time of the 2010 census is now eligible to vote. Although we can’t break down the 5-17 age range any further, there’s a strong likelihood that the majority of this 15.9% of Hawaiians will now be eligible to vote in 2020. Young people tend to be Democrats at the polls.

However, they are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. Can Joe Biden turn out the young vote?

Analysis of the 2016 presidential race in Hawaii

2016 graphic

Hawaii is one of just two states where Hillary Clinton won every county across the entire state, the other being Massachusetts. Hawaii only has five counties, given how small the population of the state is and how widely spread they are. Since Donald Trump didn’t win any counties, this part of the article will be shorter than usual!

Kalawao County was the county that delivered the widest margin of victory for the Democrats in 2016. It’s number one on the map (if you can’t see it, it’s the little green speck!).

Kalawao race

The first thing to note about Kalawao County is how small it is. Ninety people took the census there in 2010. Yes, ninety. As a result, Kalawao County is not representative of Hawaii or most states in the Union for that matter. However, it is an incredibly diverse county. Kalawao County is, in fact, a majority-minority county. Almost half of the state is Native Hawaiian, a group that powers Democratic victories across the state. Hillary Clinton took fourteen of the twenty votes cast in Kalawao County in 2016, compared to just one for Donald Trump.

Kauai County, number two on the map, is more representative of the state of Hawaii. Only Honolulu County delivered a worse result for the Democrats than in Kauai County.

Kauai race

Interestingly, Kauai County is home to a tiny percentage of Native Hawaiians, unlike Kalawao County. Kauai County also has a far higher population than Kalawao County – almost 70,000 people. The racial and ethnic makeup of Kauai County is still diverse, although in comparison to Kalawao County is more white, and almost a third of the population is Asian-American. Such diversity is one of the hallmarks of a strongly Democratic county, and although this was one of Hillary Clinton’s worse counties, she still romped to victory with 62.5% of the vote.

Honolulu County was both the county where Democrats picked up the most votes in the state, and also the county in which Donald Trump performed best. This is to be expected given that Honolulu County is home to almost a million people – half of Hawaii’s entire population. It is number three on the map.

Honolulu race

Honolulu County, like so many major population centres within the United States, is home to a wonderfully diverse mix of people – it is also another majority-minority county, where Asian-Americans outnumber white Americans more than two to one. There is also a significant amount of Native Hawaiians (as is to be expected in the county representing the state’s capital) and Hispanics/Latinos to be found in Honolulu County. As with Kauai County, diversity is a strength for the Democrats when it comes to election days. Will Hawaii be as good to the Democrats in 2020 as it has been in previous elections?

The demographics of the state certainly play into the hands of Democrats. Republicans will find it tough to win in a state where so few of the demographics are favourable to their general winning trends – small, heavily white counties tend to favour the Republicans, and these simply don’t exist in Hawaii. The diverse nature of the Hawaiian population means that 2020 should be another good election cycle for the Democrats in the Aloha State.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Georgia’s case, the winner will receive sixteen Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Hawaii polling data as of 19th May 2020

Sadly we have to go back to a world with no polling data. FiveThirtyEight has no record of any polls from Hawaii. Check back soon!

Prediction

This is the part of the article I’ve most been looking forward to the most! Hawaii is probably as easy to predict as California; I’m very glad to be back to easy predictions!

The demographics and election history of Hawaii mean there is simply no other prediction you can make – Hawaii will be a Democratic-voting state in November.

Prediction: Democratic win

The Obama administration DID leave instructions on how to deal with a global pandemic – and here’s what they say

On May 12th 2020, the leader of the Senate, Mitch McConnell, joined “Team Trump” on a Twitter live-stream and attempted to smear the Obama administration for their predecessors any instructions on how to deal with a potential global pandemic.

For starters, it shouldn’t be for the previous administration to explain in detail every potential global disaster – there could be hundreds. It’s also been nearly four years since President Obama left the White House. If like Donald Trump, you’re obsessed with being reminded of exactly how long it has been, you can click this link.

But the most important thing of all is that Mitch McConnell was not telling the truth. The Obama administration did leave instructions (sixty-nine pages of them to be precise) on how to deal with the outbreak of a global pandemic, should one occur during Trump’s presidency. You can read it for yourself at the link in Joe Biden’s Tweet below.

If you don’t want to read all sixty-nine pages of it, fear not! I did it for you! I’ll take you through the most contentious points of the Playbook, as the Obama administration called it.

It begins with a definition of what the Playbook is for: “The goal of the Playbook is to assist U.S. government experts and leaders in coordinating a complex U.S. government response to a high-consequence emerging disease threat anywhere in the world with the potential to cause an epidemic, pandemic, or other significant public health event.”

That’s on Page Four.

Page Eight accurately predicted that COVID-19 could become a global pandemic: “Pathogens that would cause heightened concern include, but are not limited to, novel (non-seasonal) influenza viruses, SARS, and other novel coronaviruses…”

Page Eleven is where things get interesting: “Collaboration with the host government to reduce issues of duplication and determine the best use of resources for the response is also necessary.”

In this case of COVID-19, the host government was initially the Chinese government. What the Playbook suggests is to collaborate and work together with the Chinese (and eventually others) to provide a global response to a global disease.

Instead, Trump did things like this. The furthest thing from collaboration.

The same page of the Playbook also suggests that “an evolving public health emergency will require continual reassessment of the pathogen [and its] location (urban vs. rural)…”

Instead of continually reassessing the pathogen and comparing rural and urban response efforts, the Trump administration has sat back and watched as some states have reopened their economies against national and international advice.

On Page Twelve, we find some of the most obvious advice, yet the most disregarded by the Trump administration: “standardised clinical care must be informed by evidence.”

Dr. Anthony Fauci, the public face of the U.S. COVID-19 efforts, has already said many times that the U.S. government has acted very differently to how the evidence suggests they should have acted.

Tapper-Fauci
Jake Tapper (left) interviews Dr. Anthony Fauci (right) on his CNN show, State of the Union (image from CNN)

Page Twelve also tells us that “in an international incident, U.S. departments and agencies will need to work with the host government and its Ministry of Health, the World Health Organisation (WHO)…and United Nations partners.”

Yet instead of working with the World Health Organisation to find a multilateral solution to a crisis which has affected almost every country on the planet, Trump did this. The literal opposite. The Obama administration was known for focusing heavily on diplomacy and working with partners – we’ll see that throughout the Playbook. Trump? Not so much.

On Page Seventeen, the Playbook warns that the “U.S. government relationship with the host government will impact the U.S. ability to provide assistance, obtain important public health data and/or cooperate on outbreak issues.”

Whether or not China would ever have provided accurate public health data is doubtful – the country is known for its secrecy, especially when something potentially embarrassing to the country’s international reputation occurs.

But there was more of a chance that China would have been receptive to American assistance if President Trump had not created a trade war with the Chinese. Relations between the two nations have naturally soured as first the U.S., and then China in response placed tariffs on each other’s goods.

Page Eighteen is the first mention of border crossings during a pandemic; the border has been the Trump administration’s key issue since Donald Trump announced his bid to become president in 2015.

This may be the one instruction that the Playbook gets wrong, at least when it comes to this particular outbreak: “The issue of border screening is complexit is rarely appropriate to put border screening measures in place at an elevated threat level.”

In this example, border screening probably is necessary, to make sure that people who are testing positive for COVID-19 are not bringing the disease into the United States.

Page Twenty-Two suggests that “developing response options for worsening siutations as soon as possible given the unpredictability and speed of evolving epidemics” is the best course of action, as well as “an early decision to request supplemental funding from Congress, if needed.”

China reported its first cases of COVID-19 in Wuhan on December 31st, 2019 (though at this time it was a new, unknown respiratory illness). At the start of February 2020, the U.S. announced its first cases of COVID-19, and not until March 25th did the U.S. government agree on a stimulus package to aid workers. The Trump administration simply failed to take COVID-19 seriously enough when it mattered.

Page Thirty-Six advises that “early coordination of risk communications through a single federal spokesperson is critical to collect and disseminate data elements from across state, local, tribal and territorial (SLTT), and federal agencies.”

Yet press conferences announcing U.S. government policy regarding COVID-19 have been held by several different people, including the President himself, his new Press Secretary Kayleigh McEnany, or sometimes not at all.

Finally, as if all this wasn’t enough, across pages Sixty-Two through to Sixty-Six, the Obama administration left examples of how it dealt with the pandemics the world experienced over its eight years in power, including

Obama
Image from Getty Images

MERS and Ebola.

So, Mitch McConnell, not only did the Obama administration leave advice for dealing with a global pandemic (contrary to the Senate Majority Leader’s claims), the advice left behind was some of the best advice the new government could have received. Having dealt with both MERS and Ebola during its eight years, the Obama government was experienced in managing pandemics. Unfortunately for the United States, and potentially the world, the Trump administration chose to ignore this advice and deny its existence. Only time will tell how damaging this decision was.

The States #11: Georgia

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Georgia (GA).

Location

Much like its southern neighbours, Georgia was a reliably Democratic state until the 1950s/early 1960s, when Republicans began targeting southern white voters who were alienated by the Democrats’ political position on civil rights. Since then, the state has been a largely Republican state, voting for Democrats only three times since 1972: twice for Jimmy Carter (a Georgia native) in 1976 and 1980, and Bill Clinton in 1992.

There are sixteen Electoral College votes up for grabs in Georgia.

State voting history

Voting history

As previously mentioned, Georgia has not voted for a Democratic candidate since 1992, and for a time was one of the most reliably Republican states in the Deep South – George Bush Jr. won the state comfortably in both 2000 and 2004. However, since the elections of Barack Obama, Georgia has become less and less comfortable for the Republicans. In 2016, Donald Trump’s margin of victory was a little over five per cent, compared to George Bush Jr.’s 2004 margin of just over seventeen per cent.

Part of this will be because of Barack Obama’s ethnic background; there is a large portion of African-Americans in Georgia as we will discuss later. A growing population has also contributed to Georgia sliding slowly towards the Democrats.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

The main thing to notice about Georgia is the sizeable African-American population in the state. Almost one-third of the state is African-American, a group who tend to be a reliable Democratic voter base. Indeed, Hillary Clinton can thank African-American voters for getting her to within five per cent of Donald Trump in 2016. However, a majority of the population is white, which would help explain why the Democrats have not been able to win the state at a general election since 1992.

Gender

State gender

The gender breakdown of Georgia is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign – especially if he chooses Stacey Abrams, who almost became the first African-American, female governor in history when she ran for the governorship of Georgia in 2018.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 5-17 category is worthy of discussion. Anyone aged eight or over at the time of the 2010 census is now eligible to vote. Although we can’t break down the 5-17 age range any further, there’s a strong likelihood that the majority of this 18.6% of Georgians will now be eligible to vote in 2020. Young people tend to be Democrats at the polls.

However, they are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could young people hold the key to Biden’s hopes of flipping Georgia?

Analysis of the 2016 presidential race in Georgia

2016 graphics

Of Georgia’s 159 counties, Democrats won just thirty, yet they were still able to limit the Republican margin of victory to just over five points. How did they manage this? Well, the clues lie in where Democrats picked up their victories, and as you can tell by the counties I have chosen to analyse, Democrats did particularly well in the north of the state. Clayton County, number one on the map, is where Democrats won by the biggest majority in Georgia.

Clayton race

Clayton County is a majority-minority county, which means that an ethnic minority is the majority of the population in Clayton County. In this case, African-Americans make up just over two-thirds of the entire county population. There is also a significant Hispanic/Latino population in Clayton County, further adding to the county’s rich diversity. African-Americans and Hispanics/Latinos tend to be reliable Democratic voters, and this helps to explain why Hillary Clinton was able to rack up an impressive victory in Clayton County, winning just under 85% of the vote there.

Fulton County, number two on the map, was where the Democrats picked up their most votes in one individual county.

Fulton race

Fulton County is also a majority-minority county, albeit only slightly. Fulton County is also home to Atlanta, the state capital of Georgia, and is home to just over 900,000 residents, making it Georgia’s most populous county. As we have seen in previous posts, counties with large population centres tend to be Democratic-leaning counties. Fulton County (and the city of Atlanta) was no exception, giving 69% of its vote to Hillary Clinton in 2016.

Yet not every area around Atlanta was quite so generous to the Democrats in 2016. Cobb County, number three on the map, is one such example of a county in the Atlanta area that was a very close contest in 2016.

Cobb race

Cobb County does not incorporate the city of Atlanta (ninety per cent of which is incorporated in Fulton County as discussed above), and so it is more of a suburban county. This is reflected in its demographics – it is far more white than either Clayton County or Fulton County, yet a quarter of the population is African-American, and 12% of the population in Cobb County is Hispanic/Latino. Suburban, white areas tend to be more Republican-leaning, and this explains how Hillary Clinton recorded such a small margin of victory in Cobb County. Democrats will have to do much better in counties like Cobb County to stand a chance of making the state competitive.

Cobb age

Yet there are reasons to think that Cobb County could move in favour of the Democrats in 2020. Nearly 19% of Cobb County’s population was aged between five and seventeen at the time of the 2010 census. Anyone who was aged eight or over at the time of the census is now eligible to vote. It is likely that a sizeable portion of this population bloc is now eligible to vote.

Given that almost a fifth of the Cobb County population fits into this age range, this results in a lot of new young voters, probably voting in their first election in 2020. Younger people tend to be Democratic voters, but also are notoriously bad at voting in the first place. Can Joe Biden harness the youth vote in 2020 and strengthen the Democrats’ hold on Cobb County, and maybe even flip Georgia?

As the map above shows, Republicans took a vast majority of counties across the state of Georgia. Glascock County, number four on the map, was where Republicans recorded their biggest margin of victory in 2016.

Glascock race

Any guesses why Glascock County voted overwhelmingly Republican in 2016? One clue lies in its demographics – Glascock County is significantly more white, and significantly less diverse, than any of the counties we have looked at in this article so far.

Glascock County is also far smaller than any of the counties above – all of the counties in which Democrats did well in had populations that reached the hundreds of thousands; the population of Glascock County is, according to the 2010 census, just over 3,000. Small, white counties are the dream Republican county, and that’s exactly the definition of Glascock County.

Yet Republicans also did very well in counties of large populations – Cherokee County, number five on the map, has a population of just over 200,000. It was also the county in which Donald Trump received the most votes in the 2016 presidential election.

Cherokee race

Here, as with the Democrats’ large victories, location is key. Cherokee County is part of the Atlanta suburbs, like Cobb County. In many ways, Cobb County and Cherokee County are relatively similar – they are suburban counties with high populations. The one crucial difference? Demographics. Cherokee County is much less diverse than Cobb County – the proportion of the population in Cherokee County that considers itself African-American is significantly lower, as is the proportion that considers itself Hispanic/Latino. Republicans have traditionally performed very well in suburban counties, and Cherokee County is the perfect example of that in Georgia. There is reason to believe that Republican dominance in suburbs might slowly be coming to an end in the not-too-distant future, however.

Finally, Twiggs County is, at least on our map, the outlier county! This article has been dominated by talk of Atlanta and its suburbs, but Twiggs County (number six on the map) is by my very quick Google Maps-ing at least 100 miles south of Atlanta. So why include it? Well, Twiggs County was the county in Georgia that delivered a win for Donald Trump by the smallest majority at the 2016 presidential election.

Twiggs race

Twiggs County is such a fascinating county because it proves that elections really are decided by who comes out to vote (that may seem obvious but it will make more sense as I explain) – this county is almost 50-50 split between African-Americans and whites. Given that it is a small (population: just over 9,000), rural county, you would expect Twiggs County to be a Republican-leaning county, and it was in 2016. However, the significant proportion of African-Americans who call Twiggs County home (and who generally, but not always, tend to vote Democrat), made the 2016 contest very close indeed for the Republicans. It’s not the biggest county, but if Democrats want to flip the state of Georgia, they would do well to win places like Twiggs County – a feat that is not entirely impossible.

Georgia’s demographics make for exciting analysis because it is a state in transition – as we saw at the start of the article, The Peach State used to be a solidly southern, pro-Republican state. Now, it is very much purple, realistically on the table for either party (although you might say that purple blend is slightly more red than blue). Can Democrats win Georgia? Sure! Stranger things have happened in politics, and with a more popular candidate than in 2016, particularly one who is held in such esteem by the African-American community, who knows?

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Georgia’s case, the winner will receive sixteen Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Georgia polling data as of 11th May 2020

Sadly we have to go back to a world with no polling data – the last time registered voters were polled in Georgia was in January. Check back soon!

Prediction

This is the part of the article I’ve most been looking forward to the most! Georgia is not yet as tough to predict as Florida, but in one or two elections time might just be.

The history is certainly in the Republicans’ favour, although the margin by which Republicans have won election-by-election has decreased, to the point where it was just over five per cent in 2016. With a more popular candidate on the ballot this time around, there is certainly reason to believe that a surprise could come and Joe Biden could flip the state of Georgia.

Unfortunately, I don’t think the demographics of Georgia have quite tilted enough in favour of the Democrats just yet, which is why I think either in 2024 or 2028 Georgia may go blue, but for now, I think the Republicans pick up the win here, by a hair.

Prediction: A very small Republican win, possibly requiring a recount.

The States #10: Florida

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Florida (FL).

Location

One of the world’s favourite tourist destinations (thanks Disneyworld!), Florida has become increasingly important in presidential election history. A traditionally Democratic state in the early period of American history, it followed the other southern states who seceded from the Union by flipping Republican in the late-1950s/early-1960s. Since then, it has given its vote to both Democrats and Republicans. Florida is the ultimate swing state.

Who can forget the drama of 2000, where Florida gave its Electoral College votes to George W. Bush after weeks of intense legal wrangling? Even the Supreme Court had to get involved. That year, Florida was decided by just roughly 600 votes, and ultimately decided one of the closest presidential elections in history.

Interestingly, Florida has voted with the winner of the general election ever since 1964 (excluding 1992). Could 2020 be another year where Florida gets it right again, and will the result in Florida accurately predict the overall outcome of the 2020 election?

There are twenty-nine Electoral College votes up for grabs in Florida.

State voting history

Voting history

The election of 2000 in Florida, as previously discussed, was genuinely a true tie. Ever since the state has fluctuated between Democrats and Republicans and got it right every step of the way. No more than six per cent has separated presidential candidates since 2000, and it is unlikely that the margin will be any greater this time around. Hillary Clinton pushed Donald Trump close in 2016, and if enough third-party votes had gone her way, she would have picked up Florida’s twenty-nine Electoral College votes. Whatever happens, expect another close race in 2020.

Don’t forget, Florida and Donald Trump have a very special relationship – his prized resort, Mar-a-Lago, is found here.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

Florida really is the ultimate battleground state, and the demographics back that up! It is white enough to consistently be in contention for the Republicans, yet also diverse enough to be winnable for the Democrats. The African-American and Hispanic/Latino populations of Florida are sizeable enough for popular Democratic candidates to make a real push to win Florida.

Gender

State gender

The gender breakdown of Florida is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With 17% of the population aged over-65 in Florida, Trump has a relatively significant block already generally backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter the 65+ vote?

Analysis of the 2016 presidential race in Florida

2016 graphic

Of the sixty-seven counties in Florida, Democrats picked up wins in just nine, with Republicans winning the other fifty-eight. It’s astonishing, therefore, that just over 100,000 votes (or 1.2%) out of over nine million decided the state in favour of Donald Trump. Let’s begin with the county where Democrats picked up their biggest margin of victory – Gadsden County, number one on the map.

Gadsden race

Gadsden County is the only majority-minority county in the entire state – a significant majority of the population is African-American, and with a healthy per cent of  Hispanics/Latinos in the county too, this county was a sure-thing for the Clinton campaign, which did well with minority voters across the entire nation. However, there are enough white people in the state to keep Clinton’s margin of victory in the state to just under eighteen per cent.

A similar result for Hillary Clinton was seen in Broward County, which is number two on the map.

Broward race

Where the African-American population is lower in Broward, the Hispanic/Latino population is much higher, as is the per cent of white people. Despite a majority white population, which generally votes Republican, Hillary Clinton was still able to rack up a decent win here, winning 67% of the vote.

Yet some of the counties Hillary Clinton won were very close contests and could be up important target counties for the Republicans if they want to build on their 2012 and 2016 victories in Florida. Hillsborough County, number three on the map, is one such example.

Hillsborough race

Hillsborough County and Broward County look very similar in demographics, yet one county (Broward) gave 67% of its vote to Hillary Clinton and the other (Hillsborough) gave just 51.5% of its vote to Hillary Clinton. Why is that? Well, Hillsborough County is slightly more white than Broward County, which is generally a good indicator of a more Republican-leaning county. The per cent of African-Americans in Hillsborough County is also significantly less than in Broward County, by roughly 10%. That 10% could have been hugely important for the Clinton campaign, which did extremely well with the African-American community nationwide.

As you can see from the map, Donald Trump took convincing victories in several Floridian counties. His widest margin of victory came in Holmes County, number four on the map.

Holmes race

Holmes County is by far the whitest county we have come across in Florida so far. It’s also the smallest, with a population of just under 20,000. For context, Broward County has a population of just under two million. As we have seen in almost every state we have covered so far, whiter and smaller counties tend to vote for Republican candidates. That was very much the case in Holmes County, where Donald Trump picked up almost 88% of the vote.

Lafayette County, number five, gave a very similar proportion of its vote to Donald Trump. Any guesses why that might be the case?

Lafayette race

Lafayette County is a very white county, with over three-quarters of its population considering itself white in the 2010 census. It looks pretty similar to Hillsborough County, with one key difference. Based on what we have seen in this series so far, the one determining factor that means LaFaayette County voted Republican and Hillsborough County is a lower level of diversity in Lafayette County. Hillsborough County has a higher percentage of African-Americans (albeit slightly), and a much higher level of Hispanics/Latinos.

But that’s not all that is special about Lafayette County. Bonus points to anyone who thought that its gender demographic could be a part of Lafayette County voting Republican.

Lafayette gender

Lafayette County is far more male than many of the counties we have looked at so far in this series, and indeed the United States as a whole. It’s not an argument that is considered very often, but people did generally vote for the candidate who matches their gender – women voted for Hillary Clinton, men for Donald Trump (unfortunately this is the only election in history where a female candidate has been a party nominee). A male-dominated was therefore always likely to lean in favour of Donald Trump.

Yet if you look at the graphic above, several of the counties that Donald Trump won in 2016 are the palest shade of red, meaning the Republican margin of victory was small. None more so than in Pinellas County, which is number six on the map and where Donald Trump recorded his smallest margin of victory in Florida back in 2016.

Pinellas race

Pinellas County is fascinating, at least from a demographic point of view, because on paper it should be more reliably Republican than Lafayette County, based on the demographic trends that we have considered in this series so far. It is much less diverse than Lafayette County and more white, so why was it only a marginal county for the Republicans in 2016?

Well, one clue may lie in its population size. Pinellas County is home to just over 900,000 people. As we have seen, larger counties (particularly those such as Denver County in Colorado) which contain big cities tend to be more Democratic-leaning. This is certainly true with Pinellas County, which is home to St. Petersburg, which is Florida’s fifth-largest city and tends to be very, very Democratic in its vote.

Put simply, the outcome of an election in Pinellas County is a contest between who comes out to vote on the day – St. Petersburg’s liberal inner-city voters or the rest of the county’s more suburban Republican voters.

Florida is one of the most exciting states in the country when it comes to presidential elections (reminder: Election 2000!). Its twenty-nine Electoral College votes make it a hugely important state, only New York (29), Texas (38) and California (55) have the same, or more, Electoral College votes going into the 2020 presidential election. Since New York and California are historically Democratic states, and Texas is a historically Republican state, how Florida votes is almost always crucial to the overall election result. If Florida votes Democratic, they capture at least three of the four most valuable states (in terms of Electoral College votes) in the country.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Florida’s case, the winner will receive twenty-nine Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Florida polling data as of 17th May 2020

Polling data

Hooray! Finally, we have a decent amount of polling data to look at. As is the case in most swing-state polls right now, Joe Biden has a small lead over President Trump, although in Florida, at least, Trump appeared to be reining that lead a little. Newer polling data gives Biden a six per cent lead over Trump. If such a lead can hold until November, Joe Biden could be looking at swinging the state of Florida back into the Democrats’ column. Florida will be polled regularly as it is an important swing state, so we’ll keep an eye on this one!

Prediction

This is the part of the article I’ve most been looking forward to the most! This is by far the hardest prediction I’ve had to make so far, and only a handful of later states will be as difficult to predict as Florida.

Tiny margins of victory have meant that whoever wins Florida will probably do it with a margin of less than five per cent. Florida is diverse enough for the Democrats, particularly a relatively popular Democrat such as Joe Biden, to push the President very close in a general election race. And don’t forget, the last time Joe Biden was part of an election campaign (as Obama’s Vice-President in 2008 and 2012), Florida voted for the Democrats.

However, something is just telling me right now that Republicans, particularly Donald Trump, have enough in their campaign to grind out another tiny victory. Florida is the President’s to lose; if the election was held today I would expect Florida to vote Republican, and unless Joe Biden pulls something special out of the locker, or Trump monumentally messes up somehow, Florida will just about vote Republican in November.

Prediction: Republican win, by another very small margin.

 

The States #9: District of Columbia

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to the capital of the United States, the District of Columbia (DC).

Location

The District of Columbia is the only place we will visit in this entire series that is NOT a state. D.C., as it is more commonly known, is the only area in the entire United States that is not a state yet is still enfranchised to vote. The capital of the United States was granted the right to vote in 1961 thanks to the Twenty-Third Amendment – that’s right, before 1961, citizens in D.C. had to watch as everyone else in the country voted!

There are three Electoral College votes up for grabs in the District of Columbia.

State voting history

Voting history

If I told you that D.C. has voted in favour of the Democrats in every single election since it was admitted into the Electoral College in 1964, would you be surprised? No other state will be so totally lopsided as D.C., that much I can guarantee you. Even the most unpopular Democrat in polling history (which extends back, ironically, as far as D.C. has been allowed to vote), Hillary Clinton, took over 90% of the vote in D.C.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

The racial and ethnic breakdown of D.C. is the single biggest clue as to why the city is so Democratic-leaning. The state is what’s called a majority-minority state, where a minority (in this case, African-Americans), make up a majority of the state’s population. African-Americans were almost certainly Hillary Clinton’s biggest backers in 2016. The state is almost one-tenth Hispanic/Latino as well, further adding to D.C.’s ultra-diverse nature.

Gender

State gender

The gender breakdown of D.C. is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With 11% of the population aged over-65 in D.C., Trump has a small block already largely backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter the 65+ vote?

Analysis of the 2016 presidential race in D.C.

This section is going to be very, very short, and pictureless. Why? Well, the District of Columbia is officially only one county! The state does break down into wards, but those are so small that the census does not release this data.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In D.C.’s case, the winner will receive three Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

D.C. polling data as of 1st May 2020

Unfortunately, there has been no polling of any form in D.C. at all! Check back soon!

Prediction

This is the part of the article I’ve most been looking forward to the most! Thankfully, predictions don’t come much easier than this. D.C. has never voted Republican in its entire voting history, and won’t do so in 2020 either. D.C. is, by far, the easiest prediction I will have to make through this entire series.

Prediction: Democratic win

The States #8: Delaware

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Delaware (DE).

Location

We’re hanging around on the east coast once again with this article on Delaware. Like Connecticut, Delaware can proudly boast that it is one of the original thirteen colonies, and one of the few states to participate in all fifty-right presidential elections. However, Delaware can go one better – it was the very first state to be established after the Americans won the War of Independence; Delaware has been recognised as a state in its own right since 1787!

State voting history

Voting history

There are similarities to Connecticut here as well – Delaware also hasn’t voted for the Republican candidate at a general election since 1988, but the margin of victory for the Democratic candidate has, in recent times, slipped somewhat. In 2008, Barack Obama won Delaware by twenty-five per cent, yet Hillary Clinton won the state by just eleven per cent. Will the downward trend continue, or will Democrats find themselves revitalised in Delaware in 2020?

ell, normally I would leave that question open, but there’s a crucial piece of information that must be considered, specific to the 2020 election. For over thirty-five years, Joe Biden, the Democratic nominee for president in 2020, represented Delaware in the Senate. Bearing in mind that senators are re-elected every six years, Joe Biden was a hugely popular representative. With that in mind, it’s certainly hard to see the margin of victory for the Democrats decreasing any further.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

Delaware is also a fairly diverse state, like Connecticut. A fifth of the state’s almost 900,000 population is African-American, and for a state this size, the per cent of Hispanic/Latino people is also sizeable. Even though the state is majority white, Democrats have still been able to win impressively here, but how?

Gender

State gender

The gender breakdown of Delaware is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With a healthy 14% of over 65s in Connecticut, Trump has a solid block already largely backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter a solid 65+ vote?

Analysis of the 2016 presidential election in Delaware

2016 graphic
Image from Wikipedia

Before we get into the fun stuff, this section will be much shorter than the others so far, since there are only three counties in Delaware!

In a visual sense, you might be wondering how it makes sense that Donald Trump could win a majority of the counties in a state and yet not win the state. Indeed, the Republicans weren’t even competitive in 2016.

The clue is the demographics of the populations in the counties. Let’s take a look at New Castle County, the only county Hillary Clinton won. It’s number one on the map.

New Castle race

There are several clues in New Castle County that explain how winning just this county meant that Hillary Clinton was able to win the state. 60% of the entire of Delaware’s population lives in New Castle County, that’s roughly 540,000 people. The county is therefore very representative of the state as a whole – the demographics of the state and New Castle County are very similar. States and counties with large African-American and/or Hispanic/Latino communities tend to be reliably Democratic voter bases, and New Castle County is no exception. Hillary Clinton won the county with 62% of the vote.

Donald Trump also did very well in just one county, that’s Sussex County which is number two on the map.

Sussex race

Sussex County is much more white and much less diverse than New Castle County. The Hispanic/Latino population is roughly the same as a percentage, you might say. True. But the population of Sussex County is far smaller than that of New Castle County – 540,000 in New Castle County compared to 200,000 in Sussex County. Roughly 8.5% of the population being Hispanic/Latino in New Castle County equates to 46,000 people; the same percentage in Sussex County equals 17,000 people. The whiter, less diverse county in Delaware naturally, on paper, voted for the Republicans.

However, one county was very much up for grabs in 2016 and could determine the margin of victory in 2020. Kent County is number three on the map.

Kent race

Kent County looks, purely on a comparison of the graphs, almost identical to New Castle County, yet voted Republican by a slim margin; so slim in fact that it did not produce an absolute majority result. Kent County is the smallest county of the three, with a population of roughly 170,000. If the white community in the county was particularly inspired to vote in November 2016, this could be one explanation for how Republicans were able to eke out a victory in the county. Whatever the reason, Kent County is very much in play in 2020 – could Joe Biden win two counties this time around?

Delaware looked, on the graphic from 2016, like a Republican state right? Fortunately for the Democrats, they performed so well in New Castle County that losses in the other two counties were not enough to hand the state to the Republicans. With at least one county in play for either side in 2020, could Republicans flip Delaware for the first time since 1988? Well, with Joe Biden, long-time Delaware Senator, as the candidate for the Democrats this time around, it’s very unlikely that the Republicans will be able to win Delaware in 2020.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Delaware’s case, the winner will receive three Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Delaware polling data as of 30th April 2020

Unfortunately, there has been no polling that fits our very rigorous criteria! Check back soon.

Prediction

This is the part of the article I’ve most been looking forward to the most! A majority of media outlets will tell you that Delaware, like Connecticut, will be a very safe Democratic state in 2020, but I don’t think Democrats should be complacent when it comes to Delaware either. After all, Wisconsin was believed to be safe in 2016.

That being said, with Joe Biden as the Democratic candidate in 2020, a candidate who is much more popular across the entire country (but especially in Delaware, thanks to his three-and-a-half decades in the U.S. Senate on behalf of the state), I can’t see the Democrats doing worse in Delaware than they did in 2016. If anything, they will probably rival their 2008 margin of victory. Even Kent County could turn blue!

Prediction: Democratic win

The States #7: Connecticut

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Connecticut (CT).

Location

Connecticut was one of the original thirteen British colonies, which stretch from New Hampshire in the north (excluding Vermont) down to Georgia in the south, along the east coast. It is one of the few states that can proudly boast the fact that it has participated in all fifty-eight of the United States’ presidential elections. During that time it has swung from right to left, as is to be expected over 230 years of election history! However, since the 1988 election, Connecticut has been a solidly Democratic state.

There are seven Electoral College votes up for grabs in Connecticut.

State voting history

Voting history

Since the turn of the millennium, Connecticut has been as safe a state for the Democrats as they come. However, the margin of victory has slipped in recent times, from 22 per cent in 2008 to 14 per cent in 2016. Will the Democrats be able to secure Connecticut more safely in 2020, or will their per cent of the vote continue to slip?

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

Connecticut is a fairly diverse state, with a healthy proportion of African-Americans and Hispanics/Latinos, both of which tend to be reliable Democratic voter bases. Nevertheless, the state is still hugely white. The clues as to how Democrats have consistently been able to win in Connecticut lies in where Democrats won, which we will consider later on.

Gender

State genderThe gender breakdown of Connecticut is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With a healthy 14% of over 65s in Connecticut, Trump has a solid block already largely backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter a solid 65+ vote?

Analysis of the 2016 presidential race in Connecticut

2016 graphic
Image from Wikipedia

Before we get into the fun stuff, this section will be a little shorter than the others so far, since Donald Trump only won two of Connecticut’s eight counties in 2016.

You might be thinking, how does a Democrat in 2016 win 54% of the vote and yet do so well in the state as a whole. Or, you might be asking yourself, how do the Republicans take 41% of the vote and only win two out of eight counties? Let’s take a look, shall we?

Hartford County is the county in which the Democrats did best in 2016, it’s number one on the map.

Hartford race

Hartford County is more diverse than the state as a whole, which would explain why this was the county in which the Democrats did best. The proportions of both African-Americans and Hispanics/Latinos is higher in Hartford County than statewide. The African-American and Hispanic/Latino vote is crucial to the Democrats winning Connecticut, and will continue to be so in 2020.

Fairfield County also proved to be a happy hunting ground for the Democrats in 2016, where they won another solid majority. It’s number two on the map.

Fairfield race

The demographic statistics for Fairfield County are very similar to Hartford County, so it’s no surprise that the election results from the two counties are pretty similar as well. The diverse nature of both counties is evidence that non-white communities turned out for the Democrats in significant numbers. Will the same be true in 2020?

Not every county that the Democrats won was as generous as Hartford and Fairfield. Tolland County, number three on the map, was a very close contest, one that was settled by just over 3,000 votes, out of a population of .

Tolland race

Tolland County is far less diverse than both Hartford County and Fairfield County. The huge white population in Tolland County is the reason that this vote margin in this county was far smaller than the other counties we have looked at so far. Tolland County could very much be up for grabs for the Republicans if they can persuade white voters to come out in even larger numbers than in 2016.

Republicans did very well in just one county in Connecticut. That count was Litchfield County, number four on the map. Republicans took 54% of the vote there.

Litchfield race

Litchfield County is remarkably similar to Tolland County, yet produced a six per cent difference in the vote margin, which was enough to tip the county in favour of the Republicans. A smaller African-American and a larger white population could be the one and only factor that resulted in Republicans winning Litchfield County.

Windham County was as much of a narrow victory for the Republicans as Tolland County was for the Democrats.

Windham County

Tolland County and Windham County look very similar don’t they? The only major difference between the two counties is a tiny reduction in the proportion of African-Americans in Windham County compared to Tolland County. What’s clear here is that both Tolland County and Windham County are both very much in play in 2020.

On paper, Connecticut seems like it might be less safe for the Democrats in 2020 than in reality. Yes, the Democrats won six out of eight counties in 2020, including in major population centres like Hartford County and Fairfield County. But there two battleground counties in Connecticut based on 2016 results – Tolland County and Windham County. If these counties manage to align their vote one way or another, they could tip Connecticut to either a safe Democrat result, or put the state on course for a marginal Republican victory. However, these two counties alone will probably not be enough for the Republicans to flip the entire state. Even if Republicans narrow the margin statewide once again, the Democrats should have enough to win.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Connecticut’s case, the winner will receive seven Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Connecticut polling data as of 7th May 2020

Polling data

Some polling for Connecticut has finally been taken! Admittedly there’s only one poll but it’s something. Connecticut is, as is to be expected, comfortably polling in favour of the Democrats.

Prediction

This is the part of the article I’ve most been looking forward to the most! A majority of media outlets will tell you that Connecticut will be a very safe Democratic state in 2020, but I don’t think Democrats should be complacent when it comes to Connecticut. After all, Wisconsin was believed to be safe in 2016.

That being said, Connecticut will probably be a blue state come November. A reliably blue state since 1992, there’s nothing that immediately suggests to me that Republicans pose a threat to that record in 2020. Connecticut is diverse enough, with some significantly sized population centres, to get the Democrats another seven Electoral College votes.

Prediction: Democratic win

The States #6: Colorado

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to Colorado (CO).

Location

Colorado gained statehood in 1876 and has had a mixed voting history ever since. It has regularly swung between periods of voting Democrat and periods of voting Republican. Between 1920 and 2004, Colorado was primarily, but not exclusively Republican. However, since 2008, the state has shifted significantly towards the Democrats. Including 2016, the Democrats have won Colorado for the past three elections. Can Joe Biden make it four?

There are nine Electoral College votes up for grabs in Colorado.

State voting history

Voting history

Interestingly, 2016 was the first election since the turn of the millennium where one of the candidates didn’t win an absolute majority of the vote. Indeed, third-party candidates could have swung the vote for Donald Trump if they had voted Republican. The margin of victory in Colorado has not been more than nine per cent since 2000, which suggests that whoever wins in 2020 will do it with another slim margin of victory.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

Much like Arizona, Colorado is majority white, but with a healthy Hispanic/Latino population too. Hispanics/Latinos tend to be a reliably Democratic voter base, which is probably what has pushed the Democrats to small victories in the state recently. However, large concentrations of white voters tend to vote Republican. 80% of  Colorado’s population is white, meaning that the Republicans will always be competitive in the state, even if they ultimately fall short at the polls.

Gender

State gender

The gender breakdown of Colorado is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With a healthy 11% of over 65s in Colorado, Trump has a solid block already largely backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter a solid 65+ vote?

Analysis of the 2016 presidential race in Colorado

2016 graphic
Image from Wikipedia

On the face of it, Colorado should have been a slam dunk for the Republicans. They won a huge majority of the counties in Colorado – all bar twenty-one of Colorado’s sixty-nine counties voted Republican. So how did the Democrats pull off a victory? The answer lies in where Democrats won. Let’s look at Denver County, which is number one on the map, and yes, it does include that weird sticky-out bit!

Denver race

Denver County is the most populous county in Colorado, with roughly 600,000 inhabitants according to the 2010 census. The county includes the state capital, surprisingly enough, it’s also called Denver. Although the county is majority-white, there is a significant Hispanic/Latino population there, voters who did reliably turn out for the Democrats back in 2016. A healthy 10% of the population is also African-American, a demographic that was a key voter bloc for the Clinton campaign. The demographics of Denver County certainly go a long way to explaining how Clinton was able to dominate, taking 73% of the vote.

Boulder County is another example of a massive Clinton victory. It’s number two on the map and is home to roughly 290,000 people.

Boulder race

Boulder is considerably more white, with much smaller African-American and Hispanic/Latino populations. However, the Hispanic/Latino population is still significant; roughly one-in-seven people in Boulder County is Hispanic/Latino. Again, the demographics help explain why Hillary Clinton was able to win big there, but with a reduced margin of victory in comparison to Denver County.

So far, we’ve seen that the Democrats won big in areas with large populations. When you look at the smaller counties in Colorado, the Democratic vote shrinks significantly. Look at Gilpin County, number three on the map. Gilpin County is where Clinton’s margin of victory was smallest. Indeed, she didn’t even win a majority of the vote in the state.

Gilpin race

Gilpin County is significantly more white than any county we have looked at in Colorado so far – and far less Hispanic/Latino. Gilpin County also only has a population of 5,000 people, in comparison to Denver County’s 600,000 people or Boulder County’s 200,000 people. Smaller, more white counties tend to be areas where Republicans win. Indeed, don’t be surprised to see if counties like Gilpin were to vote Republican in 2020, such was the fine margin Clinton won by in 2016.

Major Republican victories in 2016 came in areas like Gilpin County. Kiowa County saw the biggest Republican margin of victory in 2016. It’s number four on the map.

Kiowa race

The eagle-eyed among you may notice that these figures total over 100%. That is because the United States Census considers Hispanic/Latino to be an ethnicity and not a race. It asks a separate question for whether people consider themselves to be of Hispanic/Latino origin. As a result, a white person could also answer that they are of Hispanic/Latino origin if their family line can trace back to Mexico, for example.

Kiowa County is about as safe for Republicans as counties come. It’s incredibly white. Where the graph says 0% Asian-American, that’s because the 2010 census listed zero Asian-Americans living there. It’s also far from the main population centres like Denver. Compared to Denver County, Kiowa County is tiny as well – it’s only home to 2,000 people according to the 2010 data. Small, white and rural are the three key factors that generally guarantee a sizeable Republican victory. Kiowa County gave 75% of its vote to Donald Trump in 2016 – expect similar margins in 2020.

Washington County also returned a handsome victory for Donald Trump in 2016. It’s number five on the map.

Washington race

With a small population, just under 5,000, Washington County is very similar to Kiowa County. It is also very small, incredibly white and not very diverse. Expect figures of around 84% for Donald Trump again this time around.

Yet Donald Trump was able to scrape victories in areas with a significant population base too. Number six on the map is Pueblo County, it has a population of just under 200,000, far bigger than either Kiowa County or Washington County, but not as large as Denver County or Boulder County, both of which were Democrat wins.

Pueblo race

Yes, you could argue that Pueblo County is incredibly white, and so Donald Trump should have won. But as with Gilpin County for Hillary Clinton, this was a county that did not return an absolute margin of victory. The best explanation for this? A larger proportion of white people voted for Donald Trump, but a significant majority of Hispanic/Latino people voted for Hillary Clinton. In the end, the white vote was enough to propel Donald Trump to a victory, but it was a close-run thing. Could Joe Biden flip what is a sizeable county, and potentially strengthen the Democrats’ grip on Colorado as a state?

Like other southwest states, Colorado’s Hispanic/Latino population will be key to determining how the state votes in 2020. If enough of them turn out and follow historic voting patterns, expect to see Colorado remain Democratic and potentially even more Democratic than past elections. If Hispanic/Latino voters are not particularly enthusiastic about voting in this election, Republicans could steal an unlikely, but not entirely unpredictable victory.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In Colorado’s case, the winner will receive nine Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

Colorado polling data as of 25th April 2020

Unfortunately the last polls to be taken in Colorado that fit the criteria for this series were back in December 2019. As this data is so old, it is more than likely to be out of date, so for the time being there is no polling data. Check back soon!

Prediction

This is the part of the article I’ve most been looking forward to the most! In the California article, I mentioned that most states are “safe”, or predictable. Colorado is just about safe, but not entirely. It’s not a battleground state like Arizona, but neither is it wholly safe like Alaska for the Republicans or California for the Democrats.

A recent history of marginal Democratic victories and a relatively large Hispanic/Latino population puts the state in the Democrats’ hands.

Yet Trump and the Republicans could steal a victory here if the Democrats do not give Hispanic/Latino voters the reasons they need to go out and vote. For now, Colorado is the Democrats’ to lose.

Prediction: Democratic win, just

The States #5: California

Welcome to The States, where I profile every state in the U.S. and use the information to predict how each state will vote in the 2020 presidential election. This series will look at, but not rely on polling data. It will also look at demographics, previous voting history, and polling data as a collective, to build up a picture of what the state looks like and how it thinks. The states will be profiled in alphabetical order so as not to appear biased in any way. This article is dedicated to California (CA).

Location

Famous for the Gold Rush of 1849, when a nugget of gold was found in Sacramento, California has since grown to become the most populous state in the entire country. A former Republican stronghold in the late 1860s and early 1900s, and then again through the 1950s to 1980s, California has since become one of the most reliably Democratic states in the country. Not since 1988 has California voted Republican, and the margin of victory for the Democrats has tended to increase election on election. Let’s take a look at California’s election returns since 2000.

There are fifty-five Electoral College votes up for grabs in California, the most in the country.

State voting historyState elections

Some pretty impressive Democrat returns there. As previously mentioned, the Republicans haven’t won in California since 1984 – nine elections ago. Hillary Clinton took the best result for the Democrats since the turn of the millennium; Donald Trump took the worst result for the Republicans during the same period. 2016 was also the third election in a row where the Democrats took 60%+ of the vote. Even the 7% of voters who picked a third-party candidate couldn’t stop the Democrats steamrolling to victory in 2016.

Statewide demographics

Before we get into the statistics, bear in mind that the last time the United States took a census was in 2010, so, unfortunately, the most reliable data we have to hand is almost a decade old. However, it will still paint a mostly accurate picture of state demographics.

Why demographics?

Demographics are far more reliable than polling data when it comes to predicting a presidential race because of how wildly inaccurate polling data could be in a country the size of the United States. The voting population in the U.S. could be as high as 160 million people in 2020. As carefully and accurately as polls are done nowadays, they can never replicate what 160 million people plan to do on one day in November. People can lie or change their minds when answering who they plan to vote for in an election.

U.S. polls are often also taken of either likely voters (LVs) or registered voters (RVs). The difference is crucial. A likely voter is not necessarily a registered voter, but a registered voter is almost certain to be a likely voter. Many polls in 2016 that predicted a Hillary Clinton sweep were taken of likely voters. The polls that predicted a much closer race (or in some rare cases, a Trump victory), were taken of registered voters.

Of course, demographics don’t mean that everyone in a particular race, gender or age thinks the same. Not every African-American voted for Hillary Clinton in 2016, despite her good relationship with that community. Not every white person voted for Donald Trump, despite that being the demographic his campaign targeted. They do, generally provide a better trend for analysis than polls, however.

Race/Ethnicity

State race

This just screams “liberal dream”. The state is majority-white, true. But the main thing to notice is just how diverse the state is. Almost 38% of the state considers itself Hispanic/Latino, and just over 13% of the state is made up of Asian-Americans. Given California’s border with Mexico, it is perhaps no surprise that such a large proportion of California considers itself Hispanic/Latino. The racial and ethnic demographics suit the Democrats down to a tee. Can Joe Biden better Hillary Clinton’s  result in 2016?

Gender

State gender

The gender breakdown of California is not particularly different from that of the United States as a whole. However, both Joe Biden and President Trump have chequered histories when it comes to their behaviour towards women. That didn’t seem to matter in 2016 when Hillary Clinton took the nationwide female vote by a far smaller margin than she would have liked. Will women turn out to vote for two candidates who women will probably find difficult to back? Biden’s Vice-Presidential pick could hold the key to turning out the female vote for his campaign.

Age

State age

Before I analyse this graph, I have to admit that the way the U.S. census group ages is horrendous. 18-64 is a huge age range encapsulating within it a massive range of political beliefs and attitudes. However, the 65+ category is worthy of discussion. People aged 65+ are very good at turning out to vote, and generally (but not always) back a Republican candidate. With a healthy 11% of over 65s in California, Trump has a solid block already largely backing him. Young people, on the other hand, are notoriously bad at turning out to vote, even for Bernie Sanders who was overwhelmingly backed by young people in the 2016 and 2020 Democratic primaries. How will Joe Biden turn out the young vote, and could a revitalised youth vote be enough to counter a solid 65+ vote?

Analysis of the 2016 presidential race in California

2016 graphic
Image taken from Wikipedia

What’s most interesting about California is the East-West divide between the Democrats and Republicans. Democrats dominate the west coast, where the major population centres of California are found. Eastern counties tend to be smaller in terms of population, as well as more rural. San Francisco County (number one on the map) and Los Angeles County (number two) are two of the three most populated counties in the state. Let’s take a look at San Francisco first.

San Fran race

San Francisco, much like the state as a whole, is incredibly diverse. So diverse that no one race or ethnicity represents a majority of San Francisco’s 800,000 residents. With a healthy proportion of Hispanic/Latino voters and a third of the population considering themselves Asian-American, the diversity of San Francisco gave Hillary Clinton 85% of the vote in 2016.

Los Angeles County is the largest county in all of California, so it’s perhaps no surprise that the Democrats did well here.

Los Angeles race

Los Angeles County is home to nine million residents, making it the most populous county in the state. It is a little bit more white than San Francisco County, but it is also far more Hispanic/Latino. Given Hispanic/Latino voters are generally reliable Democratic voters, it’s no surprise that, with almost half of the population considering itself Hispanic/Latino, Clinton took almost 72% of the vote in Los Angeles.

Democrats did so well in California, they were able to flip a couple of counties that had previously voted Republican. One of those was Orange County, which is number three on the map. Just over three million people live in Orange County.

Orange County race

The slightly less diverse nature of Orange County, plus its majority-white population go a way to explaining how Republicans had previously been winning Orange County. However, the large Hispanic/Latino population could explain how Hillary Clinton was able to flip the county in favour of the Democrats, particularly when her opposition, Donald Trump, took an aggressive stance against Mexican (Hispanic/Latino) immigration in the 2016 campaigns.

But Donald Trump did have successes in some parts of California. His biggest margin of victory came in Lassen County, which is number four on the map.

Lassen race

Lassen County is much more white and much less Hispanic/Latino than the counties in which Hillary Clinton won big in. It’s also far smaller. Bearing in mind that San Francisco has a population of almost nine million, Lassen County is home to just 35,000 people. As we saw in Arkansas, small, white and rural counties tend to be where Republicans dominate. Lassen County is another example of this trend.

The gender breakdown of Lassen County may also explain how Donald Trump was able to win big.

Lassen gender

Lassen County is far more male than the United States as a whole, or indeed any other county we’ve come across so far. People tend to vote for a candidate of their gender, particularly women. A male-dominated county, generally, would return a male-dominated vote.

But Trump was also able to win some larger counties too. Kern County, number five on the map, is home to almost 900,000 people.

Kern race

Kern County is both majority white and also diverse. The majority white population probably turned out in a larger force on Election Day in 2016, as Trump was able to win Kern County with roughly 53% of the vote. But Trump could face an uphill task to retain Kern County in 2016.

Kern age

Kern County is a relatively young county. Only 9% of its population fell into the 65+ category during the 2010 census. Also bear in mind that almost 22% of the population was aged 5-17 ten years ago. Anyone who was aged eight or above during the 2010 census is now eligible to vote – which is probably a large proportion of that 22% of Kern County’s 5-17 year olds. Could they tip Kern County in favour of the Democrats, if they turn out to vote?

Butte County is the county that Donald Trump won by the slimmest margin. Indeed, if we’re being picked, he didn’t even win it since he didn’t manage an absolute majority in the county. It’s number six on the map.

Butte race

Butte County is also a fascinating example of how every trend we explore in this series rings true. With a population of 220,000, Butte County is one of California’s smaller counties, although in contrast to other states, 220,000 is huge for one county! It’s also incredibly white, far more so than any other county we have considered in California. On paper, this should have been an easy win for the Republicans. Yet for whatever reason, Trump won the county by a slim margin. Could Joe Biden find the magic formula to flip marginal counties like Butte, nationwide?

California is one of the most diverse states in the country, and that lends itself particularly to well to the Democrats. The Hispanic/Latino vote in California has recently powered the Democrats to big victories in the state, rarely giving Republicans a chance to make the state competitive, let alone steal a win there. Will that change in 2020? Well, Joe Biden isn’t particularly popular with the Hispanic/Latino population, so the margin of victory may not be as impressive as elections past. That being said, it would take nothing short of a miracle for Donald Trump to win California in 2020.

Polling data

A couple of notes about the polling data before we dive into the numbers. I will be taking the data from FiveThirtyEight, which is probably the most reliable polling model on the entire planet. It collates every major poll and lists them. It also ranks the polls on an A+/F scale according to how reliable the data is as well as taking biases into account. For this series, I will be using as many polls as possible with a B rank or above, to filter out unfair and inaccurate polls. As discussed earlier, registered voter polls tend to be much more reliable, so I will only be using data from these polls.

We must also bear in mind that there has been zero polling done in some states. In this case, the “Polling data” section will, unfortunately, be skipped. They will be updated if/when polls are finally taken. Check back often!

Also, remember that a candidate doesn’t need an absolute majority to win a state in a presidential election (put simply, they don’t need over 50% of the vote to win). The winner is simply the candidate who takes the largest percentage of the vote. The winner receives all of a state’s Electoral College votes. In California’s case, the winner will receive fifty-five Electoral College votes.

Finally, I will only be using data concerning Joe Biden and Donald Trump. Polls were taken of other Democrats against Donald Trump, but as they are no longer in the race and Biden is the presumptive Democratic nominee, these polls are sadly now redundant.

California polling data as of 17th May 2020

Polling dtat

Admittedly, there isn’t much to look at, but there’s something! Polls currently show Joe Biden with a predictably healthy lead over Donald Trump in California. Biden has expanded his lead over Donald Trump since I first wrote this post in April, now holding a commanding thirty per cent advantage over the president.

Prediction

This is the part of the article I’ve most been looking forward to the most! You’re probably beginning to gather by now that only a handful of states are, usually, competitive at general elections. California is not one of them.

A recent history of huge Democratic wins, an ultra-diverse population and several major population centres all combine to form one of the most reliably Democratic states in the entire country.

Trump might tighten the margin of victory Biden has in California in 2020. Will Trump win the state? Not a chance.

Prediction: Democratic win