Nate Silver says GOP takes Senate on bad data - Hell no says Princeton Election Consortium

[egbertowillies]

FiveThirtyEight Nate Silver released his new probabilities on who will control the Senate after the 2014 elections. He has it 60/40 in favor of Republicans.

How can you release probabilities on bad data?

But......Then..... He goes on national TV and says there is very little polling data so he is inferring.

Princeton Election Consortium has the probability of Democrats retaining the Senate at 70%.

Is Nate Silver damaging his reputation to stay relevant in the short term? Disregard the Mainstream Media and GOTV folks.
http://egbertowillies.com/2014/09/07/fivethirtyeight-nate-silver-gop-senate/

[Kablooie]

1. If he admits he is inferring and not using hard data it shouldn't be a problem.

If he tries to fake it, then it becomes a problem but if he's honest and explains that his conclusions aren't as solid as last time I don't see the issue.

[egbertowillies]

2. My issue is that he has strong credibility.

As such if he is giving probabilities based on light or bad data it creates a self fulfilling prophecy and suppresses the vote.

[Jackpine Radical]

4. He needs to be consulting with Marge.

You know her--Marge Inoverra.

[still_one]

3. unfortunately, he is not wrong. It is an uphill battle for us. Here is what the trend is

http://www.realclearpolitics.com/epolls/2014/senate/2014_elections_senate_map.html

That by no means does not mean we cannot win, but there are a lot of factors that need to line up so we do. First of all we MUST GOTV.

Next, women and minorities MUST vote in large numbers.

Our biggest problem, from historical standards is that midterms to not generate a lot of enthusiasm, and unless people get off there butts and vote, we don't stand a chance.

What really bothers me is that after the 2000 election there should be no doubt about the importance of elections, and the importance of people voting. However, election after election, compared to other countries, our turnouts are pathetic

and we ask why we are where we are today?

[RadicalGeek]

5. Thing is,

The issues to generate enthuaism among the Democratic "base" are there;
Wisconsin, the War On Women, Trayvon, Jordan, Renisha, the shutdowns.

It may be that the DNC/DLC can't upset their corproate donors, perhaps then the drive needs to come from PCCC, MoveOn, etc?

[still_one]

6. It should be very obvious to the country the republican war on women. In fact Atkin in Missouri

lost because of that. It will be quite sad if people, and especially women are apathetic about that.

Every issue you named should generate the enthusiasm to get out and vote.

I am in California so I am not privy to a lot of the ads where the races are critical for us in the country, but I have to believe that the DNC is working to support those candidates on those issues.

Of course MoveOn has always been part of the drive.

In the end, it is people who need to get out and vote in order to stop this path of disintegration that the republicans have been inflicting upon the country

[AnalystInParadise]

7. Other than the war on women meme

those other topics are not going to win us statewide elections. Please explain to me how you square the circle that we will win 50.1% of the vote campaigning on three African American's, the state of Wisconsin and the shutdowns. No disrespect to anyone previously mentioned, but seriously how do you win elections on that?

We drive GOTV and win by attacking the war on women, by refusing to get involved in another war ANYWHERE, by constantly touting the ACA, and campaigning on new ideas for the economy.

[RadicalGeek]

9. You "expand them out"

Those three African-Americans can tie into the massive amount of guns in America and the campaign of fear of "the other" that the GOP has run (Bill Moyers has two pieces on this).

Wisconsin, the GOP efforts to weaken labor in order to build the Corporate State that the Kochs and ALEC want.

The shutdowns, the politics of obstruction and, as you said, a lack of new ideas.

Women alone cannot win the Senate!

[AnalystInParadise]

15. Guns is a losing argument man

[RadicalGeek]

16. Much like how

Civil Rights was?

Point is, if the GOP continues to let the "Redneck Taliban" set it 2nd Ammendment policy, they look more and more out of step with an America that is becoming less and less Rural--IMO.

[AnalystInParadise]

17. Are you really trying

to connect the valiant struggle for Civil Rights with the desire to strip away 2nd Amendment rights? I am not even a big gun supporter, but are you serious?

[LovingA2andMI]

8. Statistical Data is all based on....

Probabilities. That if this happens, this would occur. The probabilities can change due to a number of human ( in this case) factors, like humans voting in large numbers.

However, counting on humans is not an exact science either.

Also, the narrative has been for months the GOP is going to take the Senate. Democrats thought for some reason the Fear Factor might inspire folks to the polls. Thus, they had their cable TV talking heads play into this game.

Now reality is creeping in....oh no, we might really lose the Senate...Oh No. Now, Nate Silver and every other statistician using the probabilities theories to qualify what Dems talking heads have been saying for months on end...are now wrong link.

If the Dems lose the Senate, they should blame it on the self fulfilling prophecy theory and nothing else.

[Bluenorthwest]

10. Nate Silver has predicted just one midterm cycle in his career, 2010.

He was off by about 9 seats in the House and called at least 3 Senate Seats incorrectly including Harry Reid's.
So does what he says really matter much?

[1StrongBlackMan]

12. Further ...

I remain convinced that EVERY "guesser" is using out-moded predictive models, based on "historically lower minority/youth turn-out/participation" models. While this old man can't speak to the youth turn-out/participation for 2014, I strongly believe, based on places like VA, NC, GA, and MS, the 2014 models will miss the mark, even more than the 2010/2012 models.

[santamargarita]

11. In any case, get out and vote!

[Orangepeel]

13. Silver's method has always used more inference the farther it is from the election

The closer it gets to the election, the more weight the he gives the polls. There will be more polls and they will be less likely to change due to circumstances.

That doesn't mean he's right, of course. His opinion is worth a lot when it comes to interpreting statistical data. But there's no reason to believe he's better than the average joe at interpreting voter psychology two months out.

[johnnyrocket]

14. I cannot believe that STATEWIDE, Republican votes will be there.

Maybe in the crazy districts, sure, Republicans win...but for statewide Senate races, I don't see people going out of their way to go to the polls and vote for a dammed Republican. The voters aren't that stupid.

I could still be a squeaker, but the Dems hold on, worst case in Senate is 50-50, with Biden breaking the tie.