We have been following the election closely. One of the key aspects noted by everyone is the intensity of turnout on one specific side, the Republican side. Some pundits and analysts call this “the Trump Effect” – because Donald Trump has brought more people into the election process than ever before.
We have also been predicting a tsunami election not only in ‘primary’ support of candidate Donald Trump, but also with overwhelming participation in the general election which would seemingly crush any Democrat opponent.
However, in an effort to find independent analysis for our own research we reached out with an offer for anyone to compile their own modeling. A data analyst by trade, Nick H took up our offer and here’s his findings:
Analysis by “Nick H” – Preliminary Notes / Assumptions:
- 2008 was a peak year in the Democrat party for voter turnout. Previous years did not come close to the 38 million voters in the Democrat primary process and the 68 million in the general was a record. President Obama was a key factor. (2008 turnout exceeded 2012 by roughly 6%)
- Current energy on the Republican side in 2016 is statistically higher than ever before. The four primary elections/caucuses that have taken place so far prove it because the current results smashed 2012’s results. (2012 were previous records).
- The primary predictions are hard to do because in previous years, once a nominee was presumed, turn out (regardless of party) drops. People see their vote will not matter, so they don’t vote. The current predictions assume (maybe incorrectly) that both parties will fight this out later rather than sooner.
- The terms “nomination process” and “primary process” are used interchangeably. Some of the data (like Iowa) is for caucuses, technically not a primary. To make absorption easier refer to the whole process as “nomination”, but typing “primary” is so much easier than “nomination process”. (Both used, they mean the same thing)
The analysis begins by looking at the current rate of Republican turnout and current rate of Democrat turnout, and finding the growth or decline.
Since four states from four regions voted, I bucketed each state into a region: south, north, middle, and west. For the Republican side, I looked at 2008, 2012, and 2016. What was the increase in turnout from 2008 to 2016 and 2012 to 2016? Once found, I took the average of those two and got the following results:
At first glance, the % increase is shocking, and seem inflated, however numbers are numbers and they do not lie. These are the raw voter turnout numbers in 2016 compared to 2008 and 2012.
♦ Currently the trend is 29% more voters in the republican primary process in 2016 as compared to 2012.
Next we look at the Democrat primary voters and see what is the trend for 2016 as compared to 2008. *Note there wasn’t a primary in 2012, so the comparison is Primary Turnout ’08 -vs Primary Turnout ’16.
The media and punditry have mentioned a down-turn in Democrat turnout, but the numbers are far more revealing than they discuss:
♦ Currently the trend is a 25.12% DECREASE in voters in the Democrat primary process in 2016 as compared to 2008.
Using the hard data of the first four primary results we switch to a predictive model.
Predicting the primary process turnout is based on the region of each state. Basing the region in proximity to the states that have already voted. Some maybe questionable, but most are reasonable assumptions. An example of a questionable one is Oklahoma. This model puts OK in the “middle” region, however, some could argue it should be included in the “South” region.
Based on the region of each state, I then applied the following to the 2016 estimated primary turnout:
- Republican – the average growth between “2012 to 2016” and “2008 to 2016”.
- Democrat – the average decline between 2008 to 2016.
Using these numbers, I multiplied the 2012 (2008 for dems) primary results by the regional change (percent mention above). An example: in 2012 Georgia had 901,470 votes in the Republican primary. The % increase in the South is +45%, we add another 405,661 votes to give us a predicted total of 1.3 million votes in Georgia for the 2016 Republican primary vote. (*note we will be able to verify this prediction, GA, tomorrow)
Based on the model described above, here is a look at previous year actual turnout vs 2016 estimates for the nomination process (click to enlarge):
In 2008 slightly less than 22 million voted in the Republican primary. In 2012 just over 19 million voted in the primary. The lower 2012 participation was due to everyone except Mitt Romney basically gone after the Florida Primary vote (destroying Newt Gingrich). Romney coasted to the convention.
[Side Note: The 2012 numbers almost mirror the 2000 numbers when Bush got the nomination against McCain.]
The 29.5 million estimated (projected based on current trend) for 2016 in the Republican process is massive by comparison to ’08 and ’12. However, the projection is also revealing given the fact that previous nominees where able to achieve the nomination with far fewer votes. The Republican nominee in 2016 will have a much larger group of voters having supported him to the nomination than Mitt Romney by a factor of almost 50%.
On the Democrat side things look much better on the chart than it actually might be for a few reasons: #1) Hillary had far less support this time around in New Hampshire than in 2008, despite only two people in the race this year. #2) In 2008, the primaries for the Democrats went on till the very end.
The 28.5 million estimated “projection” this year (28,450,721) is based on that long primary race happening again. However, unfortunately for Sanders, given the size of Clinton’s victories he might be forced to drop out soon, or the Sanders supporters become disenfranchised and stop voting en masse, in effect losing interest.
The 2008 primary voter turnout for the Democrats was huge. More than half of the Democrat general election voters participated in the primary race in 2008 and also 2012:
You can see that Democrats have high turnout operations in primary races. In 2008 38.1 million Democrats voted in the primary and 66.5 million Democrats voted in the general. That is a very high participation rate for primary contests in comparison to Republicans.
You can see below Republican turnout in the primaries in the past two presidential elections has been much lower compared to the general election:
In 2008 21.9 million Republicans voted in the primary and 58.1 million voted in the general election (265% increase). In 2012 19.2 million primary -vs- 59.2 million in the general (308% increase).
If those increases are even remotely maintained given the scope of the current increase in Republican Primary participation, the general election vote would be through the roof.
And here is where you begin to understand the potential scope of the 2016 primary voter for the Republican nominee, Donald J Trump. This is the bottom line the media and establishment republicans do not want to see discussed. This is the potential for the “Trump Effect”:
73 million is a lot, but the numbers do not lie. That’s 7.2 million more potential votes than Barack Obama carried in 2008, and almost 13 million more than Mitt Romney carried in 2012. That’s YUGE !
The huge boost in Republican primary votes (double in Nevada over 2012), the massive crowds at Donald Trump rallies, and the record number of viewers for the GOP debates (300% – 500%+) confirm a huge interest and support for the frontrunner Donald Trump.
The Democrat turnout is harder to put a number on since there was no primary in 2012 to draw a current trend to. However, as stated in the assumptions, 2008 was an absolute voter peak for Democrats. Anyone arguing Hillary will do as well as Obama in 2008 (66 million votes) would be hard pressed to explain that validity.
Even if Republican projection turnout was off by 5 million votes, Trump still wins in a landslide. Heck, even if the projection turnout was off by a staggering 10 million votes, the republican nominee (Trump) would still gets more votes than President Obama did in 2012 and it is highly doubtful Hillary could turn out that level of support.
SUMMARY: If the continual primary estimates hold (we’ll know tomorrow), and if these numbers translate to the general election, Donald Trump wins hands down. The question is, how many Republicans will stay home (anyone but Trump) and how many Democrats will stay home (anyone but Clinton)?
While some Republicans may refuse to vote for Trump it’s doubtful they will vote for Hillary. However, inversely, a fair amount of Democrats who refuse to vote for Clinton will likely cross over and vote for Trump.
- Reference Material – I pulled 2008, 2012, and 2016 (to date) primary or nomination process data from http://www.electproject.org/home/voter-turnout/voter-turnout-data. Needless to say there was no nomination process data for the Democrats 2012.
- The general election data for 2008 and 2012 I obtained from:
- I also used https://en.wikipedia.org and http://www.nytimes.com/ to spot check the data and to look up other years (like total votes cast in 2000 and 2004).