The Sanders Hypothesis: This Weekend's Primaries

This weekend there are Democratic Party primaries or caucuses in Kansas, Louisiana, Nebraska, Maine and Puerto Rico. The model I developed for predicting primary and caucus outcomes indicates the following results as most likely:

Screen Shot 2016-03-05 at 8.43.16 AM

Sanders is losing the primaries, so far, and Clinton is on the path to victory. However, Sanders has a fair amount of time to catch up. Perhaps he just needs his strategy to take hold. The idea was to have a revolution, which in this case, means a lot of people show up. A lot of people did show up, but not enough. Sanders needs to get the rest of them to show up in these states! (My model only addresses states, sorry Puerto Rico, and yes, you should be a state).

So, in a way, this weekend's events are individual tests for the hypothesis that Sanders can pull his nuts out of the fire and catch up to Clinton. (Unlike Republicans, we speak not of genitalia here, but rather, chestnuts. Roasting. In the open fire of politics. You've heard the expression, right?)

This could be a good weekend for Sanders in terms of victories. I'm predicting he takes Nebraska, possibly Kansas, and certainly Maine. Three wins all at once will invigorate his campaign.

However, note that these numbers are from a model that predicts a Clinton lock on the nomination by mid April. A sign that Sanders is doing better than that projection would be doing less badly in Louisiana, and significantly better in Nebraska and Kansas. I don't think doing better than projected in Maine will mean much, because there is a probably favorite son effect there.

More like this

Yesterday, the Democrats held three contests, in Louisiana, Nebraska and Kansas. I had predicted a Sanders win in Nebraska and Kansas, and a Clinton win in Louisiana, using my ever-evolving ethnicity-based projection model. Those predictions came to fruition. Like this: Predicted on top, Actual…
I developed a predictive model for the Democratic primaries that was designed to have the following features: 1) It does not rely on polling; 2) It does use exit polling and other information to set certain parameters; 3) It mainly uses prior primary or caucus results to predict the future, and…
Today and tomorrow we have the Alaska, Hawaii, and Washington Democratic primaries. According to the model I developed prior to the last primary, which predicts future primaries using information about primaries to date (which I've not updated from last Tuesday), Sanders will win all three…
Almost exactly 50% of the votes have been cast in the Democratic Party primary and caucus process. I've been updating a model to predict primary and caucus results all along, and the model has done fairly well. The most recent update, however, was a bit off. That update involved separating states…

Although only 83.8% of Nebraska has reported, the "model" is currently sporting a mean absolute percentage error of 26.8%.

Sanders won Nebraska, Kansas, and Maine by much larger margins than the model predicts; in Kansas and Maine, he won more than twice as many delegates as Clinton. Hillary Clinton would like us to believe that because she won lopsided victories in a few southeastern states, the race should be considered decided and Bernie should just quit, no need to hear from people living in other regions. When Republicans display the identical attitude in a general election campaign, it is called the "Southern strategy," and it is increasingly not a winner for them.