Archives for November 2020

It has been awhile since I’ve done one of these. Leave your thoughts on the election in the comment thread!

5:07: Four years ago on election night my house lost power for a few hours. Today, a bit after 2pm, I lost power for a few hours. I hope that isn’t a bad omen!

5:24: Ahhhh…finally got to pour myself a beer!

5:31: MSNBC: “Breaking news…Polls close in Arkansas…here’s what we can tell you…It is too close to call” Not really “breaking” news as much as prematurely ejected news.

5:41: Texas is interesting, but way too early. Florida isn’t looking good, but way too early to tell.

I am having flashbacks of 2018 when my twitter feed was full of people hysterically clutching their pearls that the blue wave hadn't materialized around this time. But when all the votes were counted …

Everyone said be patient. We knew it could come down to Pa. #ComeOnMan

— digby (@digby56) November 4, 2020

6:30: The first Senate seat to flip…Hickenlooper in Colorado.

He spent 4 years golfing. And people are still voting for him. I will never understand it

— Sarah Cooper (@sarahcpr) November 4, 2020

Senate Democrats*	Senate Republicans
greater than 99.9% probability	less than 0.1% probability
Mean of 53 seats	Mean of 47 seats

Since yesterday’s analysis, about 20 new polls have come out. The results have not changed much. This is my final analysis of the for the Senate races.

After 100,000 simulated elections, Democrats have a Senate majority 99804 times, there were 196 ties, and Republicans control the Senate 0 times. This analysis suggests that Democrats will almost certainly control the Senate in 2021.

Here is the distribution of Senate seats from the simulations:^*

This graphs shows the probability of at least each number of seats controlled by the Democrats:^*

[Read more…]

Biden	Trump
> 99.9% probability of winning	< 0.1% probability of winning
Mean of 360 electoral votes	Mean of 178 electoral votes

It is election day! I’ve added a couple dozen more polls since The yesterday analysis, but the results have not changed much. Yesterday, Vice President Joe Biden was leading President Donald Trump with an average of 362 to 176 electoral votes.

Today, I tightened up the “current poll” window to 10 days. This has the effect of adding uncertainty in under-polled states like Washington and Oregon. But doing this should better capture late trends in hotly contested states.

Today, after 100,000 simulated elections, Biden wins 100,000 times. Biden received (on average) 360 to Trump’s 178 electoral votes. The election is today and this model predicts Biden will win with near certainty. The most likely outcome (14.6% probability) is 351 electoral votes. Biden’s worst outcome in the simulated elections was 308 electoral votes, suggesting that Biden has almost no chance of an outcome below 308 votes. The highest electoral vote total was 418.

Here is the distribution of electoral votes [FAQ] from the simulations:

The trajectory of this race over the past year can be seen from a series of elections simulated every week including polls from the preceding ten days (FAQ).

Ten most probable electoral vote outcomes for Biden (full distribution here):

[Read more…]

Biden	Trump
> 99.9% probability of winning	< 0.1% probability of winning
Mean of 362 electoral votes	Mean of 176 electoral votes

There have been 131 new polls released since the previous analysis just 3 days ago in the race between Former Vice President Joe Biden and President Donald Trump. The sheer volume of polls has let me tighten the “current poll” window to polls in the past two weeks. This has resulted in the race tightening a bit. Three days ago (with a 3-week “current poll” window, Biden averaged 373 electoral votes.

Now, 100,000 simulated elections later, Biden wins all 100,000 elections and receives (on average) 362 to Trump’s 176 electoral votes. In other words, Biden would almost certainly win an election held today. We’ll have to see what happens tomorrow….

What are the big shifts over the last three days? In Georgia, there are 13 polls total that are split six with Trump ahead, six with Biden in the lead, and one tie. Biden’s chances have dropped slightly from a 94% probability of winning three days ago to a 77 percent chance of winning the state today. In Iowa, five new polls were released and five aged-out for nine polls total. Biden leads in four and Trump in four with one tie. The net result is that Biden’s slight lead (56% probability of winning) has moved to Trump leading (53% probability of winning today). Basically, Iowa is a toss-up.

We lost one of the two polls in Nebraska’s second congressional district, so Biden’s chances have slipped from a 98% chance to a 70% chance of winning. This is, essentially, a situation of too little polling.

North Carolina is the state everyone is talking about. And there is lots of polling. I have 27 current polls, with six new polls added and a loss of 14 older polls. Biden leads in 18 of the polls and Trump leads in seven polls; there are two ties. These shifts in polls have caused Biden to slip slightly from a 99% chance of winning to an 87% chance of winning. Finally, in Texas we have 11 current polls, with 5 new ones and six old ones dropping out. Trump leads in seven polls and Biden in three, with one tie. The net result is Trump’s chances have gone up again from 59% probability of winning to a 76% probability of winning.

There are some states that went for Trump in 2016 but the analysis gives Biden a solid lead, including Arizona (96% probability of winning), Florida (>99.9% probability of winning), Michigan (>99.9% probability), Pennsylvaina (>99.9% probability) and Wisconsin (>99.9% probability).

The long term trends in this race can be seen from a series of elections simulated weekly for the past year, always including polls from the preceding two weeks (FAQ).

For an election held today, here is the distribution of electoral votes [FAQ] from the simulations:

Ten most probable electoral vote outcomes for Biden (full distribution here):

[Read more…]

Senate Democrats*	Senate Republicans
greater than 99.9% probability	less than 0.1% probability
Mean of 53 seats	Mean of 47 seats

The previous analysis of the Senate races four days ago showed control of the Senate likely to go to the Democrats, with an expected 54 seats. Since then there have been many new polls released. As a result, the race has tightened slightly.

Now, the Monte Carlo analysis using 100,000 simulated elections gives Democrats have a Senate majority 99,773 times and produces 227 ties. Republicans never control the Senate. That is an almost imperceptible “slide” for the Democrats from the previous analysis. But the mean number of seats has dropped by one from 54.2 to 53.2. Part of the reason is that I’ve tightened the “current poll” window from 1 month to 3 weeks, as there is enough polling to justify the narrower window. And doing so increases the chances of catching late trends in the race.

There were a few big changes. In Alaska, where Democrat Al Gross is attempting to unseat Republican Sen. Dan Sullivan, the newest polls take the race from a tie to Sullivan in the lead with a 78% chance of winning an election today. For Georgia’s seat 1 race where Democratic challenger Jon Ossoff is trying to unseat Republican Sen. David Perdue, the last 6 of 14 polls have put Ossoff in the lead. Perdue had a 65% chance of winning four days ago, but now Ossoff has the edge with an 87% chance of winning. In Iowa, where Democrat Theresa Greenfield is challenging Republican Sen. Joni Ernst, Greenfield’s once strong lead has slipped. Four days ago she had a 76 percent probability of winning. Now she has a 61% chance. The open seat in Kansas has Democrat Barbara Bollier head-to-head with Republican Roger Marshall. Boiller had a slight lead that translates to a 63% probability of winning. But the new polling has Marshell ahead with a 91% chance of winning.

The Race in Maine has tightened as well. Democrat Sara Gideon has been leading Republican Sen. Susan Collins. In fact, Gideon still leads in all three current polls, but one poll is nearly a tie. Gideon’s lead has slipped from 97% a few days ago to 87% probability today. In another challenge to an incumbent, Democrat Mike Espy is challenging Republican Sen. Cindy Hyde-Smith for her Mississippi seat. The polling tightened up recently, and Hyde-Smith had a 56% chance of winning a few days ago. But a new Civiqs poll has Hyde-Smith up by +8%, boosting the Senator’s chances to a 90% probability of winning. Finally, in the South Carolina race, Democrat Jamie Harrison’s challenge of Republican Sen. Lindsey Graham has seen a shift in Graham’s chances of prevailing from 56% to an 88% chance.

Here is the distribution of Senate seats from the simulations:^*

[Read more…]

Most of Washington voters have already voted. But if you haven’t yet, there’s still time. Drop off your ballot at one of these locations by 8:00. Or get it postmarked by tomorrow. If you are putting it in the mail, I would drop it off today if you can.

And in any case, wash your hands right now.