Over the past month, I’ve used FiveThirtyEight‘s Democratic Primary Forecast and the always-fun 270ToWin map generator tools to produce a timelapse of the national Democratic primary. I started this project the day of the Iowa Caucuses, and have collected data up to the results of Super Tuesday (although, technically votes in CA are still being counted, the state is clearly a W for Bernie Sanders).
The Rise and Fall of Bernie Sander’s “Revolution” in 44 Images
For those of you unfamiliar with FiveThirtyEight‘s forecast tool, one of it’s many useful features is to provide a range of percentages of each state’s votes that each candidate it likely to get, and then to provide an average vote percentage for each state.
For example: as of the morning of March 3, Sanders wins between 14% and 35% of the vote in 80% of simulations in Missouri. His average vote total in these simulations is 24%. In the same state, Biden wins between 28% and 54% of the vote in 80% of simulations. His average vote total in these simulations is 41%.
I then took these average percentages for each state and found the difference between them.
From the previous example with Missouri, on average Biden leads Sanders by 17%.
After finding that difference, I used the following scale to determine how likely a candidate was to win the state.
- Safe – Leading by 21% or more
- Likely – Leading by 20-16%
- Lean – Leading by 15-11%
- Tilt – Leading by 10-1%
After that, it was simply a matter of applying color-coding to that scale and filling out the map through 270ToWin.
Originally, I was generating one map per day. However, as events began to unfold more quickly (Bloomberg’s brief surge, the Nevada Caucuses, two debates, etc.), I began making two or four maps per day to reflect the quickly-shifting landscape.
1. The South was always Bernie’s weakness and Biden’s strength.
Even when things looked at their worst for Joe Biden’s campaign (right after the Iowa caucus), most of the South was only tilting towards Sanders. Bernie never had a secure hold on the South as a whole, and only ever had a significant lead in Louisiana, Florida, and Georgia.
Biden’s total domination in the Southern states on Super Tuesday should not have been terribly surprising to anyone.
2. Things can change very, very fast.
Between February 3rd and 6th, Bernie went from doing decent in the West and some of New England to literally leading in every single state.
Between the South Carolina primary on February 29 to March 4, Joe Biden went from doing decent in the South to owning it, as well as making tremendous gains in the Midwest, Great Plains, mid-Atlantic, and New England.
Things are moving fast in this primary now. Multiple states are voting almost every single week for the next couple of months. Debates are going to keep happening. Endorsements will pile on.
3. Debates performances matter.
This can probably be seen most clearly with Bloomberg’s polling. In mid-February he was gaining significant traction in many Southern states, New York, and some Great Plains states.
Then he was in a couple debates.
After that, his polling plummeted and that was reflected in the vote totals of Super Tuesday and his almost-immediate departure from the race.
4. Joe doesn’t have this in the bag…yet.
After Biden’s indisputable thrashing of Sanders, Warren, and Bloomberg this past Tuesday, it would be easy to say that Biden’s path to the nomination is wide-open. But as we can see from Bernie’s massive wave that overwhelmed almost the entire country after Iowa that became a mere shadow of it’s former self within a few weeks, nothing is ever set in stone until someone gets a majority of delegates of the convention in Milwaukee decides the nominee.
Now, I am willing to say that the race is now Biden’s to lose. But I also said in early February that the race was Bernie’s to lose. And clearly, he figured out a way to do that.