Nate Silver: G.O.P. Stays on Upswing in Senate Forecast

[depakid]

Republican chances of taking over the Senate have improved again in this week’s forecast. They are now 22 percent — up from 18 percent last week and 15 percent two weeks ago. Republican chances are now approaching the point where they stood prior to the Delaware primary, when they had peaked at 26 percent before Christine O’Donnell’s victory.

There are six states currently held by Democrats where Republicans have at least a 75 percent chance of winning, according to the model. These are North Dakota, Arkansas, Indiana, Pennsylvania, Wisconsin and Colorado. Although some of these races — particularly Wisconsin and Colorado — could tighten, the G.O.P. has a fairly strong chance to win all six if they are having a good night overall. (They will probably win North Dakota, Arkansas and Indiana even if they are having a disappointing night.)

Although earlier in the cycle, it had looked like some Republican-held seats were in play, that is less of a concern for them now. Instead, some races that had once seemed competitive — like Ohio, North Carolina, New Hampshire and Florida — have seen the Republican candidate gain ground in recent weeks.

One partial exception is Kentucky, where some polls show the Democrat, Jack Conway, moving into a somewhat stronger position. It is not that Republicans couldn’t lose Kentucky — Mr. Conway has been running some strong ads lately — but it is unlikely to be the state that prevents them from gaining a majority.

Instead, it is a set of seven states that are likely to determine Republican chances to control the Senate. These are Nevada, Illinois, West Virginia, Washington, California, Connecticut, and the New York special election. Republicans would need to win four of these seven races to take claim of the Senate.

More: http://fivethirtyeight.blogs.nytimes.com/2010/09/30/g-o-p-stays-on-upswing-in-senate-forecast/#more-1637

[RBInMaine]

1. Enough of the punditry. Dems, just WORK. Call, canvass, donate, rally. Screw the pundits.

[jefferson_dem]

2. GOP need at least four of these: NV, IL, WV, WA, CA, CT, NY.

As it stands now, I don't see that happening. In order of Dem. vulnerability (from most to least), in my opinion: IL, WA, NV, WV, CT, CA, NY.

[Cosmocat]

3. I'm sorry ...

first off, we are talking about a 25 percent, give or take chance ... Again, if the Ds were in that spot, it would not be cast as a "chance" to gain the chamber, it would be cast as extremely unlikey ...

Also, I totally respect how strong Silver has been in the past with his calls, but seriously ... Reid has not been down for months now in Nevada, Blumenthal has never been down as far as I know ... I mean, seriously ... Sestak has his margins down in Pa to what a lot of the margins have been all along for Rs in other races, and they are seen as competitive and Pa is seen as a dead race for Sestak ... Seriously, McMahaon has never been any closer than Sestak has been ...

California is done, has been for a while now, but it still is seen at risk for dems ...

I also think this carries over to the house end, too ...

The trends are not great, but they have tended D for a while now ...

[karynnj]

9. Good comparision of CT and PA

In CT there is no doubt the numbers are declining for Blumenthal, but the closest is one poll where he is up 3. In PA, the pattern is a little bit less smooth, but Toomey routinely had about a 7 to 8 margin and the last two polls were 3 and 5.

I think the difference is that Silver is answering a question from the Republican POV, Can they take over the Senate? The 22% statistic is the real answer. He then has to explain the set of states they need to take. the fact that many of these states are more likely to go to the Democrats - and the Republicans need all but 2 is why there is a 78% probability they won't take the Senate. (I suspect that by next week, CA will essentially be out of reach for them - meaning they need 4 of the 5 - and most currently favor the Democrat.

What it means is that there is a real threat - though it is 3 times more likely to NOT happen than happen. That means we need to work.

[karynnj]

5. CT, WA, NY and CA seem unlikely - especially NY and CA

Not to mention, election booth "conversions" are likely to benefit Blumenthal - not the WWE Queen. In addition, I think many of those saying "green" in Chicago will likely - as happens in nearly every race - return to Giannoulias. (Yes, I know he has baggage - but I know why I didn't vote for John Anderson, who I preferred to Carter.)

I suspect that we have a better chance to win all of them than they do. I think the reason - complicated math aside that the results are near where they were when O'Donnell was chosen is that West Virginia surprised everyone. However, I think CA is really unlikely to be in play any more.

The article itself has the probability of winning each race - and the Gillibrand seat has a 7% chance of going Republican. I really don't know why it was included - because there are states he awarded to the Republicans with greater chance of going to the Democrats. (KY is 11% and NH is 9 - MO is a bit lower with 6%)

Edited to add NY, which I didn't see before - and which has never been in play.

[Mass]

6. The 3 races we could lose in the 7 are IL, WV, and NV

I cant even start to figure out how NY came in this list.

For the rest, I guess we will figure out who has good likely voter models and who does not.

[boomerbust]

4. Depa fail

Wisconsin will not go Republican

[depakid]

18. Talk to Nate, not me

He's the one running the numbers and cranking out the odds.

[FBaggins]

19. Or just list the last time a senate incumbent that far behind came back to win

with a month left in the race.

[depakid]

20. It happens both in sports (where Nate's background is) and in politics

Typically though in order to reverse a trend or tendency, there has to be a "halftime adjustment" or a game changing play or two.

We've seen a momentum shift in California for example (although losing either of the big races there would have been nothing short of astonishing from the outset).

Personally, I've been watching this bellwether game pretty closely:

http://elections.nytimes.com/2010/forecasts/senate/nevada

[FBaggins]

23. It CAN happen... but only rarely.

And as you said... it requires something that changes the game. Something beyond the bald guy with the painted head yelling "we are NOT going to lose this game!"

The reason I asked for an example is that I've been following politics for a long time and can't think of an example in the last 20 years.

[golfguru]

24. Wisconsin is toast from what I am seeing n/m

[alcibiades_mystery]

7. DVT

Useless.

[FrenchieCat]

14. Made his morning!

Definitely DVT!

[bigdarryl]

8. he's delusional we will make SURPRISES in this election I'm listening to

various African American talk shows on radio and the host are really telling the audience WE CANNOT sit out this election because of the things the rethugs are planning I think the African American vote and Hispanic vote will be much higher than the polling firms are telling people

[MadMaddie]

21. I agree with you!

[Dawgs]

10. Probably because of Feingold seat.

I'm assuming it went from toss-up to leaning Republican.

[FBaggins]

11. Actually, it's WI and WV

The chances of holding Wisconsin have declined by about 10% in the last week (according to Silver), the chances of losing WV have climbed from 10% to 37% in the last two weeks.

[bigdarryl]

13. Not according to this poll WV is DEADLOCKED!!

http://www.huffingtonpost.com/2010/09/29/west-virginia-senate-poll_n_743146.html

[FBaggins]

15. One poll does not a deadlock make.

There's no question that the race has tightened in WV, but Silver's estimation looks pretty accurate. It's not a 50/50 proposition at this point, but it's getting close.

If there were other (non rasmussen) polls that showed an R lead, the chance of loss would go up significantly.

[Aramchek]

28. and Nate Silver can't really predict the future. He just pretends to.

Then takes credit for the successes and blames the failures on polling models.

[FBaggins]

37. Can you back that up or are you just blowing smoke?

For instance... Of the races that he's claimed are 50/50... How many went each way?

My bet is that you've seen ten races that he rated as 90% one way or the other and nine of the ten went the way he predicted... And you're latching on to the one he got "wrong" without realizing that you just proved him correct.

For Cook (as an example), almost exactly 50% of the races that he's rated as "tossup" have gone each way.

[Aramchek]

39. He's made lots of predictions, and lots of them were wrong.

You can treat his numbers like facts if you want. And that will affect what you think is possible and thereby what you do.

I will view them as what they are, guesses by a guy who makes a lot of guesses.
They don't change what is possible.

[FBaggins]

40. You could have saved typing time by just saying "blowing smoke"

Or can you provide some examples? Or someone with a BETTER track record?

Announcing that you've decided to stick you head in the sand isn't particulalry compelling.

[Aramchek]

41. somebody likes blowing Silver, that's for certain

do I really have to provide evidence that the dude's not infallible?
you are definitely well-versed in his polling. you know his failings even though you ignore them.

Quit acting like anyone should base their actions on what Nate Silver says.
He's much closer to a bookie than a Nostradamus.

[FBaggins]

45. Infallible? Boy... you really are laying the strawman arguments on thick aren't you?

So sorry that someone asking you to back up claims of fact is so disturbing to you.

He's much closer to a bookie than a Nostradamus.

That's almost exactly right. And you're one of those true-believer football fans that:

1) Bets straight-up on his team when the bookies all say they're 14 point dogs.
2) Accuses anyone who won't make that ridiculous bet of really rooting for the other team.
3) Ends up dead broke nine times out of ten.
4) Hollers that the one time the bet worked out in his favor is proof that the odds-makers don't have a clue what they're talking about.

[muffin1]

42. +1,000,000

:rofl:

[Mefistofeles]

43. What's the ratio, more or less, of his wrong predictions vs. his correct ones?

To the best of your knowledge.

[karynnj]

12. Feingold has been in trouble a long time - West Virginia is the major change

The Governor was said to be a shoo in - the most popular politician in the state. Some polls now show him losing.

[Bluenorthwest]

16. Nate Silver is full of numbers, lacking in all other factors

Plus, he makes his money by 'predicting the future' like Jean Dixon, so he has to keep predicting interesting things, that is his product. His way of looking at things is so compartmentalized that it is worthless.
How did Nate's Numbers predict Meg Whitman's troubles? Could Nate's Numbers have done so? Nope. And so it goes with Nate's Numbers.

[FBaggins]

17. Sorry. That's untrue.

He is a wiz with the numbers... no question, but his model does take a great deal of other factors into account. In fact it's highly unlikely that you can name one that isn't included. You might try reading his explanation of the model before jumping to conclusions.

he makes his money by 'predicting the future' like Jean Dixon, so he has to keep predicting interesting things

Nope. He makes his money predicting the future, and therefore he has all the incentive in the world to get it right.

How did Nate's Numbers predict Meg Whitman's troubles? Could Nate's Numbers have done so? Nope. And so it goes with Nate's Numbers.

That's pretty ridiculous. All political prognostication is necessarily based on "what it currently known". If Schumer dropped dead tomorrow (heaven forbid), Nate's prediction that he has essentially no chance of losing would have to change - but that doesn't mean it's currently incorrect. Can you question the weatherman's prediction of an 80% chance of rain tomorrow by saying his model didn't predict the meteor that fell last night?

[Aramchek]

26. Nate Silver is Jesus and Charlie Cook is God, right?!?

[Mefistofeles]

27. That's a strawman

Nobody claimed or suggested they're God. Please give us real arguments, with premises and stuff.

[Aramchek]

29. you don't know Frodo, do you? Don't you dare blaspheme Charlie Cook!

[FBaggins]

30. Not divine of course... But if you could think of someone with a better track record

I presume you would have named one by now.

Certainly superior to the childish alternative of "I don't WANT them to be right... So they must be wrong"

[KingFlorez]

31. Don't even bother

Anything that shows Democrats losing badly is the best news ever to some people around here.

[Mefistofeles]

33. What is your prediction for November?

How optimistic are you?

[KingFlorez]

34. I don't think it's going to be complete disaster

From what I'm seeing, there has been movement toward Democrats in enough places to maintain control.

[Mefistofeles]

36. Thanks. Here's mine

We keep the Senate and lose the House. Just like Silver and Sabato predict.

[Aramchek]

35. my prediction is that Mr. Mefistofeles has a Pizza Party by November

some folks just can't hide how they really feel.

[MadMaddie]

22. Good for him that he can come up with statisics.

However; one thing he can not predict is actual turnout. They can do as many surveys and other predictions but they cannot accurately determine actual turnout.

This election cycle outcome will be entirely dependent on "VOTER TURNOUT". This is why I am not paying attention to Silver or any polls.

Whichever side shows up wins it's as simple as that. So if there is anyone is out there that is thinking about sitting this out they should really think about what that means. If you don't vote a RepublicanTeapartier will. That's one more vote for them and one less for us.

[Aramchek]

25. is bitterness its own reward?

[BeyondGeography]

32. Alternative headline: GOP Senate takeover chances now only slightly less remote

Big deal, Nate.

[ClarkUSA]

38. So what? Historically, the party in the WH nearly always loses seats during the midterm elections.

Democrats have had the largest share of Senate majority in decades, thanks to Obama08 turnout. It was inevitable that Democrats would lose seats, especially given the state of the improving but still struggling economy affecting amnesiac independent voters.

Complacency is the bane of Democrats now, not dissatisfaction. Listen to President Obama's Gen44 Summit speech, during which he addressed the issue.
He asked Gen44 members (of which I am one) to get engaged and start GOTV.

The important point to take away from Nate Silver is this: a Senate takeover is still unlikely. And that's in part because liberal Obama supporters like me and others in OFA groups across the nation are working hard to GOTV.

In fact, local OFA GOTV efforts on behalf of Sestak are having a noticeable effect, with Sestak trending upwards and pulling within 3 points/the margin of error in the latest poll after trailing by double digits for months. The PA Democratic Chairman gave us heartfelt thanks on a conference call a couple of days ago. Sestak is actually slightly edging out Toomey with independent voters now.

:)

[Phx_Dem]

44. Quit raining on our parade, Nate!

Just kidding. He's a good guy who calls it as he sees it.

:P