538 Senate Forecast Megathread: Currently Projecting 6 Democratic Pick Ups

[Gass3268]

Democrats currently sit at a 62.5% of taking back the Senate. Democratic chances were only at 51.4% on October 9th but there have been generic congressional poll has really started to break for the Democrats recently (Ipsos +10, NBC +6, Fox +6, Morning Consult +5).

Odds:

Colorado 95.7% (D)
Illinois 88.4 (D)
Wisconsin 87.3% (D)
Indiana 62.9% (D)
Pennsylvania 57.2% (D)
New Hampshire 55.4% (D)
-------------------------------
Nevada 50.8% (R)
Missouri 58.2% (R)
North Carolina 58.3% (R)
Florida 73.1% (R)
Louisiana 86.2% (R)
Kentucky 88.7% (R)
Ohio 95.4% (R)  

[Gass3268]

There have already been some changes since this morning

[Gass3268]

Going with the polls-plus model (that's the one I used above, and the one that you are directed to right away): 65.3%

Odds:

Colorado 95.9% (D)
Illinois 89.0 (D)
Wisconsin 88.6% (D)
Indiana 66.7% (D)
Pennsylvania 59.7% (D)
New Hampshire 55.6% (D)
Nevada 53.7% (D)
-------------------------------
North Carolina 55.2% (R)
Missouri 56.0% (R)
Florida 74.5% (R)
Louisiana 84.6% (R)
Kentucky 87.6% (R)
Ohio 95.1% (R)

[Blackacre]

Alright Nevada!

Also, why does 538 use Polls-Plus as the default for the Senate Forecast anyway?

[Sorenroy]

Alright Nevada!

Also, why does 538 use Polls-Plus as the default for the Senate Forecast anyway?

I think the real question is why they don't for the presidential race. I forget where, but Silver said that the polls-plus model is the one they have the most confidence in.

...Actually, now that I think about it, I think that the low number of polls would make anything but the polls-plus model swing wildly with any change. With fewer polls, the polls-only could end up looking very similar to the nowcast (although for some reason it doesn't, it just looks like the polls-plus adjusted in favor of one of the candidates ).

[Figueira]

Looks like the coattails have arrived!

[Gass3268]

Looks like the coattails have arrived!

Yeah, the generic congressional poll is sitting around 5 points and it appears to be growing. This is definitely not enough to take the House, but probably enough to take the Senate.

[Blackacre]

Looks like the coattails have arrived!

Hillary ordered the coattails on Amazon.com and didnt click free 2-day shipping, so they took roughly 14 days to get to her.

However, I always buy my tortured metaphors on Amazon Prime

[Blackacre]

Senate Dems now clear 70% in polls-plus, Nevada is blue in all 3 forecasts and Cortez Mastro cleared 60% in polls-plus, and NC and MO have gone blue in the now-cast. The races are breaking for the Dems, though after NC and MO I don't expect any other seats to go blue unless something crazy happens in Florida.

In the polls-plus, Dems have a 30.1% chance of getting a majority between 53 and 59 seats, and 53/47 is just .3% more likely than 50/50.

[OneJ]

In MO, Kander has 48.2% chance of winning compared to Blunt's 51.8%. http://projects.fivethirtyeight.com/2016-election-forecast/senate/missouri/

[Blackacre]

Thats not the weird one.

North Carolina, Polls-plus. Ross a 50.2% favourite, but the expected margin shows an exact tie, and on the "who's ahead in each Senate race" chart, NC is shown as R+0.1. What even?

[Figueira]

NC is now tied in margin and chances.

[Antonio the Sixth]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

[Vosem]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

He takes generic congressional ballot polls into account.

[The Other Castro]

Missouri has now flipped to Kander for the first time in the Nowcast.

[heatcharger]

Missouri has now flipped to Kander for the first time in the Nowcast.

And in the polls-only and polls-plus forecasts as well.

Democrats chances to take back the Senate:
Polls-plus: 74.0%
Polls-only: 71.0%
Now-cast: 81.0%

[Blackacre]

Holy crap, Missouri in the polls-plus.

I fail to check the forecast for 45 minutes or so to research for an essay due 5 days before the election, and suddenly polls-plus MO goes from 49.5% chance Kander win to a 60.1% chance Kander win. Completely got me by surprise, esp because I was fixated on the weird fluctuations in the NC race on the model

[The Other Castro]

Holy crap, Missouri in the polls-plus.

I fail to check the forecast for 45 minutes or so to research for an essay due 5 days before the election, and suddenly polls-plus MO goes from 49.5% chance Kander win to a 60.1% chance Kander win. Completely got me by surprise, esp because I was fixated on the weird fluctuations in the NC race on the model

It's all because of the Google Consumer Survey poll that was just added.

[Antonio the Sixth]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

He takes generic congressional ballot polls into account.

Meh, I wouldn't put much trust into generic ballot polls. People don't have an option to vote for "generic Democrat" or "generic Republican", they can only vote for the candidates that are actually running where they live.

[Figueira]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

He takes generic congressional ballot polls into account.

Meh, I wouldn't put much trust into generic ballot polls. People don't have an option to vote for "generic Democrat" or "generic Republican", they can only vote for the candidates that are actually running where they live.

Sure, but it means that if Ayotte, Burr, and Toomey win, it won't be because voters want a "check" on Clinton or whatever.

[Antonio the Sixth]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

He takes generic congressional ballot polls into account.

Meh, I wouldn't put much trust into generic ballot polls. People don't have an option to vote for "generic Democrat" or "generic Republican", they can only vote for the candidates that are actually running where they live.

Sure, but it means that if Ayotte, Burr, and Toomey win, it won't be because voters want a "check" on Clinton or whatever.

Yeah, that would at least be the silver lining.

More important, however, is making sure they don't win.

[Figueira]

NC is now tied in margin and chances.

I don't get why. There hasn't been any poll showing Ross ahead recently, and the most recent one has Burr+1.

He takes generic congressional ballot polls into account.

Meh, I wouldn't put much trust into generic ballot polls. People don't have an option to vote for "generic Democrat" or "generic Republican", they can only vote for the candidates that are actually running where they live.

Sure, but it means that if Ayotte, Burr, and Toomey win, it won't be because voters want a "check" on Clinton or whatever.

Yeah, that would at least be the silver lining.

More important, however, is making sure they don't win.

Obviously.

But I think it's clear that if they win it's because of factors in those particular races, not because voters think it's somehow noble to want a Republican Congress and a Democratic President.

[HAnnA MArin County]

Yay Missouri! Maybe there still is hope for this state.

[Gass3268]

Polls-Plus - 77.4% D
Polls-Only - 75.3% D
Nowcast - 84.2%


Odds (Using Polls-Plus):

Colorado 97.7% (D)
Wisconsin 95.1% (D)
Illinois 90.5 (D)
Indiana 69.8% (D)
New Hampshire 63.67% (D)
Pennsylvania 62.4% (D)
Missouri 61.2% (D)
Nevada 60.6% (D)
North Carolina 54.4% (D)
-------------------------------
Florida 70.4% (R)
Kentucky 72.4% (R)
Louisiana 83.1% (R)
Arizona 89.2% (R)
Ohio 95.1% (R)

They are currently projecting 7 Democratic pick ups.

[Blackacre]

Is the DSCC going to take the money they beat out of Murphy and give it to the Kentucky Dems then, or are they going to come to their senses and give Murphy the money he needs to win?