 |
 |
It has been mid-May since I published one of these analyses. back then, Biden led Trump in the electoral college with a mean of 348 to 190 votes.
Now, with over 40 new polls, Biden still leads. In the 10,000 simulated elections, Biden won all 10,000 times. Biden received (on average) 354 to Trump’s 184 electoral votes. If the election was today, we would expect Biden to win with a greater than 99.9% probability.
Here is the distribution of electoral votes [FAQ] from the simulations:
Ten most probable electoral vote outcomes for Biden (full distribution here):
- 353 electoral votes with a 3.33% probability
- 335 electoral votes with a 3.00% probability
- 351 electoral votes with a 2.94% probability
- 356 electoral votes with a 2.79% probability
- 369 electoral votes with a 2.56% probability
- 354 electoral votes with a 2.17% probability
- 350 electoral votes with a 2.16% probability
- 363 electoral votes with a 2.12% probability
- 357 electoral votes with a 2.10% probability
- 341 electoral votes with a 2.07% probability
After 10,000 simulations:
- Biden wins greater than 99.9%, Trump wins less than 0.01%.
- Average (SE) EC votes for Biden: 354.0 (22.5)
- Average (SE) EC votes for Trump: 184.0 (22.5)
- Median (95% CI) EC votes for Biden: 353 (310, 404)
- Median (95% CI) EC votes for Trump: 185 (134, 228)
Each column of this table shows the electoral vote total aggregated by different criteria for the probability of winning a state (Safe=100%, Strong=90%+, Leans=60%+, Weak=50%+):
| Threshold | Safe | + Strong | + Leans | + Weak |
|---|---|---|---|---|
| Safe Biden | 205 | |||
| Strong Biden | 69 | 274 | ||
| Leans Biden | 61 | 61 | 335 | |
| Weak Biden | 0 | 0 | 0 | 335 |
| Weak Trump | 34 | 34 | 34 | 203 |
| Leans Trump | 60 | 60 | 169 | |
| Strong Trump | 58 | 109 | ||
| Safe Trump | 51 |
This table summarizes results by state. Click on the poll count to see the individual polls included for the state.
| 1 | 1 | EC | # | Total | % | % | Biden | Trump | |
|---|---|---|---|---|---|---|---|---|---|
| 6 | 2 | Votes | polls | Votes | Biden | Trump | % wins | % wins | |
| AL | 9 | 2* | 1085 | 39.4 | 60.6 | 0.0 | 100.0 | ||
| AK | 3 | 1* | 272 | 47.1 | 52.9 | 25.9 | 74.1 | ||
| AZ | 11 | 6 | 3493 | 52.5 | 47.5 | 98.5 | 1.5 | ||
| AR | 6 | 1* | 604 | 38.4 | 61.6 | 0.0 | 100.0 | ||
| CA | 55 | 2 | 1175 | 66.5 | 33.5 | 100.0 | 0.0 | ||
| CO | 9 | 3 | 1625 | 58.8 | 41.2 | 100.0 | 0.0 | ||
| CT | 7 | 1 | 814 | 62.9 | 37.1 | 100.0 | 0.0 | ||
| DE | 3 | 1* | 394 | 58.1 | 41.9 | 99.1 | 0.9 | ||
| DC | 3 | 0* | (100) | (0) | |||||
| FL | 29 | 8 | 16740 | 50.6 | 49.4 | 87.5 | 12.5 | ||
| GA | 16 | 4 | 2923 | 50.0 | 50.0 | 49.4 | 50.6 | ||
| HI | 4 | 0 | (100) | (0) | |||||
| ID | 4 | 0 | (0) | (100) | |||||
| IL | 20 | 0 | (100) | (0) | |||||
| IN | 11 | 1 | 929 | 42.8 | 57.2 | 0.1 | 99.9 | ||
| IA | 6 | 1 | 1149 | 48.9 | 51.1 | 29.7 | 70.3 | ||
| KS | 6 | 1 | 643 | 43.5 | 56.5 | 1.1 | 98.9 | ||
| KY | 8 | 1 | 1038 | 41.5 | 58.5 | 0.0 | 100.0 | ||
| LA | 8 | 0 | (0) | (100) | |||||
| ME | 2 | 1* | 819 | 55.3 | 44.7 | 98.4 | 1.6 | ||
| ME1 | 1 | 1* | 459 | 60.3 | 39.7 | 100.0 | 0.0 | ||
| ME2 | 1 | 1* | 438 | 51.6 | 48.4 | 68.3 | 31.7 | ||
| MD | 10 | 1 | 765 | 62.6 | 37.4 | 100.0 | 0.0 | ||
| MA | 11 | 2 | 1546 | 66.2 | 33.8 | 100.0 | 0.0 | ||
| MI | 16 | 11 | 10539 | 53.2 | 46.8 | 100.0 | 0.0 | ||
| MN | 10 | 1 | 744 | 52.7 | 47.3 | 84.8 | 15.2 | ||
| MS | 6 | 1 | 442 | 43.7 | 56.3 | 2.9 | 97.2 | ||
| MO | 10 | 1 | 460 | 47.8 | 52.2 | 25.9 | 74.1 | ||
| MT | 3 | 1 | 390 | 46.7 | 53.3 | 17.5 | 82.5 | ||
| NE | 2 | 0 | (0) | (100) | |||||
| NE1 | 1 | 0 | (0) | (100) | |||||
| NE2 | 1 | 1 | 417 | 55.9 | 44.1 | 95.5 | 4.5 | ||
| NE3 | 1 | 0 | (0) | (100) | |||||
| NV | 6 | 1 | 717 | 52.2 | 47.8 | 79.3 | 20.7 | ||
| NH | 4 | 1 | 750 | 54.5 | 45.5 | 96.4 | 3.6 | ||
| NJ | 14 | 3 | 2042 | 60.8 | 39.2 | 100.0 | 0.0 | ||
| NM | 5 | 1 | 929 | 56.5 | 43.5 | 99.7 | 0.3 | ||
| NY | 29 | 3 | 2286 | 65.4 | 34.6 | 100.0 | 0.0 | ||
| NC | 15 | 13 | 9753 | 50.7 | 49.3 | 84.5 | 15.5 | ||
| ND | 3 | 1* | 372 | 40.9 | 59.1 | 0.7 | 99.3 | ||
| OH | 18 | 3 | 2128 | 50.0 | 50.0 | 48.9 | 51.1 | ||
| OK | 7 | 1* | 450 | 36.7 | 63.3 | 0.0 | 100.0 | ||
| OR | 7 | 0* | (100) | (0) | |||||
| PA | 20 | 9 | 6235 | 52.5 | 47.5 | 99.5 | 0.5 | ||
| RI | 4 | 0 | (100) | (0) | |||||
| SC | 9 | 1 | 555 | 44.7 | 55.3 | 4.2 | 95.8 | ||
| SD | 3 | 0 | (0) | (100) | |||||
| TN | 11 | 1 | 477 | 40.5 | 59.5 | 0.2 | 99.8 | ||
| TX | 38 | 5 | 4734 | 48.9 | 51.1 | 15.0 | 85.0 | ||
| UT | 6 | 2 | 2481 | 45.1 | 54.9 | 0.0 | 100.0 | ||
| VT | 3 | 0 | (100) | (0) | |||||
| VA | 13 | 1 | 507 | 56.6 | 43.4 | 98.3 | 1.7 | ||
| WA | 12 | 3 | 2023 | 61.0 | 39.0 | 100.0 | 0.0 | ||
| WV | 5 | 1* | 495 | 33.3 | 66.7 | 0.0 | 100.0 | ||
| WI | 10 | 9 | 5826 | 53.1 | 46.9 | 100.0 | 0.0 | ||
| WY | 3 | 0 | (0) | (100) |
* An older poll was used (i.e. no recent polls exist).
Details of the methods are given in the FAQ.
The most recent analysis in this match-up can be found from this page.
Darryl, questions about the methology in the Y2 Analytics poll used in Utah. The pool of respondents appears to be a sample of a sample, and if I’m reading it correctly that same panel of 2,000 voters is called on repeatedly to be polled. So any bias in that sample is perpetuated. I guess to me it seems less “random”. It is a randomized poll? Is the MOE any different because of the methodology? Any other comments? Thank you.
godwinha @1,
The pool of respondents appears to be a sample of a sample,
Yes…this is a pool-based sampling scheme.
if I’m reading it correctly that same panel of 2,000 voters is called on repeatedly to be polled.
This is my understanding.
So any bias in that sample is perpetuated.
Biases will definitely be present in the pool, which is fixed. Of course, all samples, whether randomly selected, or a sample within a sample, likely have bias, so that isn’t new. The concern is that the same panel is being resampled repeatedly and non-randomly for serial polls.
The Y2 Analytics folks try to correct for that by use a double weighting scheme to create something of a synthetic group of respondents. The first weighting is based on the probability of voting based on voting record in 2016 and characteristics like age and party registration. (Their methods are not described in enough detail to understand exactly how this weighting is done; I can imagine a repeated re-sampling method that would provide a distribution of outcomes that could be part of the sampling error, although there is nothing that suggests they are doing anything like this.) The purpose of this first weighting is to more heavily weight individual responses for individuals who are more likely to vote.
The second weighting is traditional poll weighting to match characteristics of Utah’s voting population.
I guess to me it seems less “random”. It is a randomized poll?
The sample of respondents do not approximate a random sample. In the past, I wouldn’t consider including such a poll in these analyses. Sadly, the death of the landline, combined with restrictions of automated polling on cell phones, have forced pollsters to broaden their polling methods. A lot of traditional pollsters are now incorporating some internet polling, recruited panels, etc. Y2 Analytics current polling method is not an internet poll, even if the poll is completed over the internet. The original panel was recruited from the state’s registered voter file. The don’t specify if that original recruitment was via phone or mail.
Is the MOE any different because of the methodology?
The margin of error for the sample size should be larger because of their double weighting scheme. Unfortunately, they don’t specify how they estimated the margin of error. But using traditional methods (that apply to randomly sampled respondents with a true D/R split of 50%) for a sample of 1,099 respondents, the MOE is 3%, which is what they give. So, they have likely underestimated sampling error in their poll, even if they have attempted to correct for any bias in their sampling scheme.
Any other comments?
As you probably know, 538 uses polls close to an election to grade pollsters. Y2 Analytics get a B/C from 538, based on 4 previous races. They got the election results correct in all 4 races, but the final margins were 0.4% biased toward Democratic candidates. That said, I believe their methods were substantially different for the polls included in 538s grading exercise. If so, we really have no idea how good Y2 Analytics polling is.
One way to assess them now is to see how they are doing compared to other pollsters. So far Rasmussen is the only other polling firm doing Utah:
Rasmussen has Utah as leaning way more toward Trump than Y2 Analytics. The last poll from each firm was included in the above analysis, giving Utah 100% probability of going for Trump in an election held now.
It doesn’t matter to me. I will not be voting for Trump or Biden in November.
3 You want 4 more years of “I take no responsibility” trump. Got it.
Can someone remind me of the polls, of the over/under, of the betting odds at this time in 2016. . .
@5 On the superficial scale you’re a 10. Godwinha at least has a capacity to ask intelligent questions and carry on above-3rd-grade level conversation when he wants to.
I don’t care what their fucking grievances are. The looting of businesses and destruction private and public property are fucking crimes just as anything they’re protesting is.
Concern troll @7: We don’t care what your fucking grievances are. If you won’t listen to us, we won’t listen to you. What’s good for the goose, is good for the gander.
If you really care about saving property, tell your friends to stay home.
https://www.youtube.com/watch?v=axth3x32T2s
[Deleted — Off Topic]
9+ MINUTES OF VILE YOU WILL NEVER GET BACK – YOU’VE BEEN WARNED
This Woman Thinks the Chicago Police Are Shirking Their Duty to Disarm Store Owners, So That Black People Can Loot Safely
* * *
LOL – CLOWNS TO THE LEFT OF ME. . .
Big Bird and Elmo Will ‘Address Racism’ in Virtual CNN Town Hall
CNN is teaming up with the perennial children’s program “Sesame Street” in a virtual town hall that will “address racism” in America in the wake of the police involved death of George Floyd.
* * *
GO FOR IT GRETCH!
Michigan Gov. Gretchen Whitmer told radio show Mojo In The Morning on Thursday that she understands the “Defund Police” sentiment being pushed on social media by young people in the wake of George Floyd’s death.
GARCETTI HAS YOU BEAT GRETCH – NOW IF HE ONLY HAD THE BALLS TO DEFUND AND DISBAND THEM COMEPLETELY.
LA Mayor announces significant CUTS to police budget to reinvest in “communities of color”
* * *
“IT IS STUNNING.”
[ Deleted — Off topic. Take it to an open thread]
CONSEQUENCES:
ALERT: DITCH THAT iPHONE – APPLE AIN’T ‘WOKE’ ENOUGH
* * *
Oh yes, and then there’s these. . . HAPPY THURSDAY!!
Missouri: That Lake of the Ozarks party didn’t move the COVID-19 needle after all
* * *
Lancet Formally Retracts Fake Hydroxychloroquine Study Used By Media To Attack Trump
A study of 96,000 patients indicting the benefits of hydroxychloroquine on coronavirus-stricken patients used to attack Trump has been retracted.
[Off Topic]
(Darryl, please feel free to delete this for O/T )
[As you wish :-) ]
[Deleted. Sorry…the content really had nothing to do with the election analysis, polls, or even the 2020 election…]
A couple more weeks of Trumpalo Toy-Soldiers beating the shit out of peacefully protesting children and grandfathers while Stupid Hitler “inspects” his Panic Room may cost him two or three of his must-win states forever.
I’ll be closely watching poll trends in WI and MI for declines in Stupid Hitler numbers, and in PA for a rise in Biden numbers.
There is probably going to be a time lag of some duration between the markets’ recovery and some measure of Trump’s acceptability, whether that be approval, percent chance of victory according to the analysis that is the subject of this thread, or whatever. Graphing some measure of Trump acceptability vs. the DJIA 50-day and 200-day moving averages might be interesting over the next several months. Granted, those would be national approval and international financial figures while the EC is an aggregate of individual state results. However, apparently the jobs news reported today is the largest one-month increase in jobs ever. It’s the economy, stupid.
It’s certainly a relief to me that HA has QoS McHillbilly to closely watch @ 15 poll trends in states like WI and PA. Because what would Darryl know about those states:
@16 – Anthropology? Can you actually get a job with that credential?
I can only imagine.
But I’ll try not to. When most of us picture The Q Clearance Pussy in need of relief the image includes leather zipper masks, ball gags, and plenty of sobbing masturbation.
Instead of waiting, why not try the same analysis for three or four past presidents? In your dotage with nothing but time on your hands that should be something you can handle.
The results might even be surprising.
@ 19
Because I can afford to wait – I have the luxury of time. I’m not the one who has hitched my wagon to the physical health of RBG and the mental health of Biden.
I see the looters are being persnickety about their targets, going after Rolex watches and other high value loot.
Hey looter – Stealing Rolexes does not turn you into decent people. You’re all still pieces of shit.
You’re quite right. You’re the one who hitched your wagon to Vladimir Putin and his Stupid Hitler Bunker Boi.
Doctor Dumbfuck and his orange moron’s Covid-19 fuck up just added trillions to the national debt, cost tens of millions their jobs, and you have no idea where all that money went, only that it’s gone and tens of millions of Americans are still unemployed. But do keep your eye on those Republican incel looters stealing shiny Rolex watches. Doctor Dumbfuck and his orange moron are counting on you doing that cuz maybe it’ll help improve their dismal poll numbers. Know what I mean?
@4 Where he’s really going is he wants 4 more years of “I take no responsibility” for Trump. I have yet to see any Trumpalo own up to being responsible for putting Trump in office.
Dear Jay, Jenny and Dow,
During the entirety of your “pandemic” lockdown, I have been to the post office, to Costco, to Fred Meyers, to Safeway, to Winco, to Trader Joe’s, to a health food store, to Home Depot, to Walmart, to Les Schwab, to a doctor, to my mechanic and to a fabric store.
During this “pandemic” lockdown I have NOT been to my parish for any reason, let alone a Mass, nor to a restaurant, nor to get a hair cut, nor to simply browse and shop.
At none of the places I visited have I been asked for personal information for “contact tracing”.
In order to go to an OUTSIDE Mass limited to 100 people, I am now required to devulge my information for “contact tracing”.
Jay, Jenny and Dow, until you can prove you have “contact tracing” for the “peaceful protestors”, those hoarding thug savages destroying our cities, you can go stick you heads where the sun don’t shine, and evidently where you kleep your brains (limited as they are) IF you can get your noses out of each others.
You won’t. Your onion tears appeasement to the savages prove this entire “pandemic” lockdown and destruction was phony nonsense by cowardly pretenders in search of self-aggrandizing power.
And, Archbishop Ettiene, grow a spine. Your allegience is to God and to your faithful, not to temporary “leaders.” Be a shepherd not a sheep. Your job is not to bow down to petty earthy dictators but to stand and defend your faith.
@24 oh give me a f*ing break. I voted fro Trump because he wasn’t that oft drunk Marxist liar Hillary and you voted for her because she wasn’t him. That can be said for most people on both sides
The same thing will happen this year. We have become a nation of the least common denominator, glorifying ignorant celebrities and more ignorant politicians, getting angrier at each other as years pass and it continues.
Your allegience is to God and to your faithful, not to temporary “leaders.” Be a shepherd not a sheep. Your job is not to bow down to petty earthy dictators but to stand and defend your faith.
Lemme guess, Mel Gibson’s dad was a personal friend.
And you can’t celebrate the Mass in Latin ‘cuz of … reasons.
I CAN celebrate the Mass in Latin. Our parish is increasingly using Latin and more traditional Gregorian music, but yes, there are churches that celebrate with the complete Latin Mass I celebrated every single school day of my elementary education. I still have (and use) my 1959 St Joseph’s Daily Missal
There are 8 parishes in the Seattle Archdiocese with 2 in Seattle and 2 in Tacoma; another 4 in the Spokane and Yakima diocese combined. There are 7 in the Portland Diocese with 5 in Portland.
And thank you for capitalizing ‘Mass’.
ftfy
This dawning realization about yourself that you’ve stumbled upon since your last episode of masturbating to the videos of George Floyd’s murder was of course inevitable.
But equally inevitable is the simple inescapable truth that this sorry state you are in is irreversible.
Modern enlightened society has stood beckoning you away from the edge for the past fifty years, patiently coaxing you with compromise and excuses. But times up. You’ve slipped into the abyss from which there can be no return. Modernity threw you its last best lifeline four years ago. Instead you voted for the nominee you told you right to your face that he sexually assaulted lots of women. He didn’t care. You didn’t care. And now we’re done with all of you.
“Don’t take anything personally. Even when a situation seems so personal, even if others insult you directly, it has nothing to do with you. Their point of view and opinion come from all the programming they received growing up. When you take things personally, you feel offended and your reaction is to defend your beliefs and create conflict. You make something big out of something so little because you have the need to be right and make everybody else wrong.”