Welcome to my first blog post!
I’m no expert on Electoral College math, so I’m asking a simple question more focused on the popular vote:
What factors will impact the popular vote difference between the Republican POTUS candidate and the Democrat POTUS candidate in November 2020?
In my view, the math for what November’s vote count looks like is something like this (where T is the number of votes for Trump and D is the number of votes for the Democrat candidate[1]; T2020 represents the votes for Trump in 2020):
T2020 = T2016 - (People that voted for Trump in 2016, but don’t in 2020) + (People that didn’t vote for Trump in 2016, but do in 2020)
and similarly:
D2020 = D2016 - (People that voted Democrat in 2016, but don’t in 2020) + (People that didn’t vote Democrat in 2016, but do in 2020)
This part is mostly obvious, but let’s break down some of the items (with their new variable names in parentheses):
“People that voted for Trump in 2016, but don’t in 2020” (Tminus) is comprised of two subgroups:
1. People that choose not to vote at all in 2020 (Tabstain), and
2. People that choose to change their vote to Democrat (Tbadflip)
It’s notable that the first subgroup is a -1 popular vote wise, while the second is a -2 since it takes a vote away from Trump and adds it to the Dem side.
“People that didn’t vote for Trump in 2016, but do in 2020” (Tplus) is similarly comprised of two subgroups:
1. People that didn’t vote in 2016, but vote Trump in 2020 (Tnew), and
2. People that voted for Hillary in 2016, but choose Trump in 2020 (Tgoodflip)
And, similar to the previous section, the first subgroup gives Trump a +1, the second a +2.
Skipping a bit of written work (take off marks if you need to), we’re left with:
T2020 = T2016 - Tminus + Tplus
T2020 = T2016 - (Tabstain + Tbadflip) + (Tnew + Tgoodflip)
Re-organizing this a little gives:
T2020 = T2016 + Tnew + Tgoodflip - Tabstain - Tbadflip, and for the Dems:
D2020 = D2016 + Dnew + Dgoodflip - Dabstain - Dbadflip
Recognizing that Tgoodflip (people that voted for Hillary and now are voting for Trump) is the same as Dbadflip(people that voted for Hillary and are now voting for Trump), we can simplify a little:
T2020 = T2016 + Tnew + Dbadflip - Tabstain - Dgoodflip
D2020 = D2016 + Dnew + Dgoodflip - Dabstain - Dbadflip
And re-arrange:
T2020 = T2016 + Tnew + Dbadflip - Tabstain - Dgoodflip
D2020 = D2016 + Dnew - Dbadflip - Dabstain + Dgoodflip
Now, computing the popular vote difference (where a positive number is a D victory, popular vote wise):
Difference2020 = D2020 - T2020
Difference2020 = D₂₀₁₆ + Dnew - Dbadflip - Dabstain + Dgoodflip - (T₂₀₁₆ + Tnew + Dbadflip - Tabstain - Dgoodflip)
Difference2020 = D2016 + Dnew - 2Dbadflip - Dabstain + 2Dgoodflip - T2016 - Tnew + Tabstain, or
Difference2020 = (D2016 - T2016) + Dnew - 2Dbadflip - Dabstain + 2Dgoodflip - Tnew + Tabstain
Knowing that the difference in 2016 was 2,868,686, we’re left with:
Difference2020 = 2,868,686 + Dnew - 2Dbadflip - Dabstain + 2Dgoodflip - Tnew + Tabstain
It looks pretty busy, but let’s look at each component part:
· Dnew – how many net new voters will vote Democrat in 2020 (“net new” in the sense that they didn’t vote in 2016). I think that we can assume that it will be significant for several reasons:
1. Some people stayed home because they just didn’t like Hillary. Presuming that Biden is the nominee and, even acknowledging his perceived shortcoming, it’s hard to imagine him being hated in the way that HRC was in many circles.
2. Some people stayed home because they thought a Hillary win was “in the bag” – the abject shock of a good portion of the nation after election night suggests that this was a prevailing attitude. This year, those people know what’s at stake.
3. Some voters that would vote Democrat stayed away because they were pro-Bernie and couldn’t stomach voting for his rival – this somewhat overlaps with #1, but I’m expecting that this will be slightly reduced (I did say slightly) in 2020.
4. People have watched Trump operate for over three years and those not enamored of the MAGA/KAG movement might show up and vote since they perceive something is at stake for them personally (in a way that they did not in 2016).
· Dbadflip – these are people that voted for Hillary in 2016 and now cast their vote for Trump. I travel in the US, talk to my US friends & family and follow US politics fairly carefully (although, with full disclosure, I follow many more D-friendly sites, accounts, etc., so I might not have the best assessment of this), but I can’t picture too many people in this camp. And remember, this is the killer group because they represent a 2 vote flip – kind of like losing a game late in the season to the team you’re fighting for a playoff spot.
· Dabstain – this is the group that voted D in 2016 and now chooses not to vote. Given potential misogyny against Hillary, the dislike of her by many, I think if you voted for her in 2016, you were pretty committed to the party (and her). I don’t see this group as particularly huge, but maybe I’m missing something (not an intentional “abstain” joke).
· Dgoodflip – these people were Trump voters in 2016 and change to Dem voters for 2020. Again, it might be biased by what/whom I follow, but I “feel” like there are a lot of these. People who voted for a well-known business person or the reality show star or for a change or to “drain the swamp” (or they just hated Hillary that much) will have 4 years of Trump’s track record to judge. I have a potentially-biased view of that, but I think this will be a big number and it’s another powerful 2 point switch.
· Tnew – These are people who didn’t vote in 2016, but wake up in 2020 and decide to vote for Trump. I think it’s well-understood that his current base is strongly behind him, but I don’t see much evidence (and I could be wrong) of new people who will show up to vote for him based on what has happened so far in his term (and I believe that was true even before COVID-19).
· Tabstain – these people voted for Trump in 2020, but stay home in 2020. For some of the same reasons as the “Dgoodflippers”, some will stay home, but in their case, they can’t stomach a vote for the Dems. There could be a large number of these (solely based on my opinion).
So, based on this (and here’s where it gets a lot less mathematical until someone does some unbiased surveys), let’s assume (from the above analysis) that the following are “bigger” numbers:
Dnew, Dgoodflip, Tabstain
And the following are “smaller” numbers:
Dbadflip, Dabstain, Tnew
This leaves our equations as (and please ignore anything you learned in Algebra at any level):
Difference2020 = 2,868,686 + big - 2small - small + 2big - small + big
Dangerously grouping like terms:
Difference2020 = 2,868,686 + 4big - 4small
Depending on how big “big” is and how small “small” is, this certainly looks favorable to the Democrats. Do with this what you will – I’m sure that I can be accused of bias in this, but it was simply a thought exercise for me to understand what factors will impact the vote in November.
And I know that popular vote does not equate to Electoral College success, but I might argue that this math applies (maybe to a greater or lesser extent) in each Electoral College district. I look forward to your comments and thank you for reading all the way to the end!
[1] I’m intentionally not using H, since there will be a new Dem candidate unless something very crazy happens!
