Thanks! My comment about probabilities not summing to 100% (or nearly so) did not apply to the binary (or nearly binary) races like Senate seats, but to nominee markets where there are several choices.
I think a lot depends on Trump, if he wins NC easily then Tillis wins. If Trump manages to lose here or just barely wins that could mean Tillis loses.
f(x) = Φ(c * Φ[sup]-1/sup) has the property that |0.5 - x| > |0.5 - f(x)| when c > 1 (i.e. it pushes things away from 50%). So if we had a + b + c = 1 with a, b, c < 0.5, then f(a) + f(b) + f(c) < 1 is guaranteed since we have f(a) < a, f(b) < b, and f(c) < c.
For this reason and others this debiasing formula is not sufficient for races with three or more possible outcomes.
We should hope the voting is correlated, since the White House does us little good without the Senate. But aren’t there a large group of baffling voters who will say “I voted D for President, so will vote the R Senator for balance” ? “Gridlock! Yayyy!!!”
In 2016 Trump won NC by 3.5% and Roy Cooper won the governor race by 0.2%.
Currently both Trump and Cooper are each favored slightly again. this certainly indicates that the senate seat seat is in play, and the most recent poll result is Cunningham (D) +7 over Tillis (R).
My adjusted PredictIt reflects this as well. Last night’s posted results were:
NC Electoral College: P(D) = 46.8%
NC Senate: P(D) = 64.5%
Also see (from January): NEW POLL: Tillis’ Approval Ratings Still Underwater Even After Spending More Than $700k on Primary Ad Campaign
I still don’t know how Hagan won for NC Senate in 08 but I guess Obama helped her and Dole tried to call her an atheist. Also Bev Perdue won for governor that year and that may have helped Hagan too.
You know, I use to hear that, I haven’t for years, in fact for decades now. I feel like that reduced a lot after the 90s.
You’re right that voting tends to correlate, both inside a state and between states.
But this statement is still demonstrably untrue, septimus. In any emergency care situation the first step is stopping the bleeding. Taking the White House is an important first step. I think we have a chance at the Senate - more data as more polls closer to E-day will make things clearer - but taking the White House and changing the direction of the federal government is overwhelmingly important.
Or, to rephrase, which is preferable? Gridlock or Trump and McConnell continuing to work together for another four years?
I don’t actually find that all that baffling. There are people who either a) don’t want the government to be able to do anything much or b) want the government to only be able to do things on which there is wide agreement.
The second of those would IMO make considerable sense if it weren’t that for some years now we’ve had people in the US government who disagree with anything that the other party’s in favor of, purely on the grounds that the other party’s in favor of it. So it’s gotten unreasonably difficult to get wide agreement.
Some encouraging senate polls this week, but keep in mind that there’s a long way to go.
Iowa: Ernst (R) 43 +1 over Greenfield (D) 42, PPP May 4
Montana: Bullock (D) 46 +7 over Daines (R) 39, MSU Bozeman May 5
North Carolina: Cunningham (D) 50 +9 over Tillis (R) 41, Civiqs May 5
Colorado: Hickenlooper (D) 48 +17(!) over Gardner (R) 31, MSU Bozeman May 5
PPP is a Dem polling organization, so sadly can’t be too happy about the Iowa poll. The rest look very hopeful and a Montana win would be unexpected. So cool. Daines ® is the incumbent.
I wouldn’t get too happy about any of the polls this far out and when polling data in each race is pretty sparse, but while PPP is a partisan organization their polling is pretty much on the level. 538 has them as a B rated pollster with a slight 0.3% Dem lean.
Of the four polls in my post, the PPP is the most reliable. Neither Civiqs nor Montana State Bozeman have the track record necessary to even be properly rated, and the Civiqs poll was paid for by Daily Kos.
Man, Montana would be a gift I never would have dreamed of receiving a few months ago.
I think she will lose. But there’s a huge amount of pressure there, and the GOP will have to choose where to spend it’s resources. IMHO, they will spend it on keeping control of the senate, not shoring up trump.
So even in races like kentucky, where the dem candidate will likely lose due to GOP voter fraud and suppression, keeping the pressure on is critical.
That’s why I see texas as critical- Biden and trump are neck and neck in polls. The GOP will have to pour money there to keep it red- or pour $ to keep Moscow Mitch?
Incidentally, I think Amy Mcgrath would be a wonderful Senator. I even sent her a small donation.
I follow this blog that has been doing polling analysis/forecasts as hobbies on elections since 2004. He’s a self-declared conservative but a numbers man and has a solid record at predicting elections.
His current projections are:
President: Biden 278 - Trump 260
Senate: Republicans 51 (-2) - 47 Democrats (+2)
House: Democrats 240 (+6) - Republicans 195 (-5)
This blogger has also been going since 2004 and also has a solid record.
Although his electoral college forecast is way too optimistic for me as he assesses through 2020 polls which in some states there haven’t been many. And in some states like Texas where the polls show a statistical tie, he marks it as a tie at this point. The other guy makes his calls as if the election were to be held today.
This guy doesn’t make much sense. Even though his map has CO as Likely GOP, he says this about the race:
He’s only just starting to cover senate races. That write-up he’s done was all done in the last 24 hours (check the bottom of the page) so I expect the map to reflect swiftly. Up to now he has only concentrated on the presidential race.
Just got a fundraising letter from his likely Dem opponent, Barbara Bollier, in which a picture of Trump pointing approvingly at Kobach is featured quite prominently.
As did I! There’s also a super PAC helping her: https://ditchmitchfund.com/
Just sent to me by a politically-savvy Ohio lawyer I know:
*Friends,
I hope that each of you remain safe and healthy—far more so than our democracy currently is. From the bungled response to the pandemic to the corrupting of the DOJ as evidenced by the Flynn debacle to the trumpeting of a made-up scandal as a diversionary tactic, it’s clear that the Conman-in-Chief is corrupting our government and our institutions as badly as we’d each feared. Perhaps even worse.
So what’s the antidote? We all know the #1 mission: Dump Trump this November. And we’ll all do our part here in Ohio…
We each have to make decisions about how to allocate our donation dollars to get the best bang for our bucks. Along those lines, I just gave to Sen. Gary Peters (D-Mich.). Here’s why:
· Michigan is more in play than Ohio is, and it’ll be nearly impossible for the Orange Overlord to win a second term without winning Michigan.
· Sen. Peters’ poll numbers correlate strongly with Biden’s. Engaging Democrats benefits both candidates in lockstep.
· Mission #2 after dumping Trump is to improve the Senate. To do that, we need to defend the seats we already have. Two Democratic Senators stand for reelection in states that Trump won in 2016—Sen. Peters, and Sen. Doug Jones. Of the two, Sen. Peters’ race will likely be tighter, and it is the only one taking place in a state that’s in play in the Electoral College.
· Sen. Peters has deep relationships in Michigan’s urban centers and can engage those voters to increase turnout that will benefit Democrats up and down the ballot.
For all these reasons, in my estimation, Sen. Peters’ campaign is an efficient destination for donations—one that can help us accomplish two goals with the same money. As you all consider how to allocate your donation dollars this cycle, I hope that you’ll consider Sen. Peters’ campaign.
I’m happy to talk more about Sen. Peters with anyone who’s interested. And for those of you who would like to support his campaign, here’s a link you can use to do so: ActBlue
Many thanks for your time and consideration. Stay safe. Wash your hands.*