The truth about Democratic chances in the House

[ProSense]

The truth about Democratic chances in the House

by Chris Bowers

Nate Silver currently gives Democrats a 21% chance of keeping control of the House. Pollster.com also gives Democrats a 21% chance. Seems like a consensus to me.

With that in mind, here are three seemingly common things that are less likely than Democrats winning the House in 2010:

1. First, it’s less likely that your birthday is in either August or September (18%) than it is for Democrats to keep control of the House. However, we all know lots of people born in either August or September.
   
   
2. Second, your odds of rolling doubles in Monopoly, thus winning a second roll or getting you out of jail, are lower (17%) than the odds of Democrats winning the House. But everyone who has ever played Monopoly has rolled doubles many times.
   
   
3. Third, it’s less likely for an NFL team trailing by seven points at the start of the 4th quarter to win the game (about 15%) than it is for Democrats to keep the House. Still, every football fan, player and announcer knows comebacks like that happen all the time. Two Sundays ago, it even happened twice in one day (here and here). A third, even larger comeback happened on that same day.
   Yes, Democrats are behind. However, a 21% chance of victory is far from being defeated.
   
   more

Maybe Tim Kaine and Howard Dean are right?

[phantom power]

1. That is good to keep in mind.

The question that it all begs is, why are we staring at an 80% chance at losing the House to the party who is currently represented by the fucking Tea Cracker Freak Show?

:shrug:

[depakid]

10. +1

Inquiring minds want to know.

[Liberal_Stalwart71]

2. Anyone who does statistics and does them right knows and understands that there is what

we call "random noise." Statistics is never a perfect science because it is impossible to know all the NOISE that exists in the world: (1) voter turnout; (2) election fraud; (3) weather impacts; (4) people not counted in a sample due to the cell phone as the only reliable means of contact; or, (5) a catastrophic event that could alter the election just days before it occurs (e.g., the bin Laden tape that surfaced the weekend before the election that may have affected Kerry's changes in 2004).

The problem with polls is that they simply provide a snapshot in time that is used to project future outcomes. While they are great at estimating likely outcomes based on available information that one may have at any given time, they are not great at projecting how NOISE will affect the true outcome.

With this in mind, I'm going to commit to not driving myself crazy with these polls. Instead, let's all commit getting out the vote. I voted early in Maryland precisely so that it frees up time to drive people to polls, to make calls, to canvass, etc.

I don't think we should give up.

[FrenchieCat]

3. Polls ARE noise!

[Bluenorthwest]

7. Yes indeed the polls are the noise, they are noise about noise

What the stats people call noise is really called the world, life, reality. Right now there is a woman playing the violin on Countdown, and she's not noise either, she's the world. Damn.

[jberryhill]

4. Although noise itself is quantifiable

My favorite undergraduate engineering course was EE 301 "Random Signals and Noise"

It looks great on a college transcript, to those unfamiliar with tech stuff.

"You sat through a whole semester of random signals and noise?"

Sums up a lot of classes really...

[Doctor_J]

5. It's actually pretty damn close to being defeated

This is a really bizarre fantasy piece. I am not sure what the purpose is. "I've roled doubles in Monopoly, so Speaker Boner is not a sure thing". Well, that's comforting

[polichick]

6. Most polls are meaningless so I'm with Kucinich on this (POTUS radio today):

"I'm not conceding anything."

[Beetwasher]

8. Remember The Following: GARBAGE IN GARBAGE OUT

Nate and other analysts use the data given to them from all the polls. Rasmussen and Gallup inundate the field with their polls and therefore have a HUGE net effect on the data. It is becoming clearer, that there may be some serious problems with both of those outfits methodology that produces significant pro-Republican results (intentional or not).

Garbage in garbage out.

[Valienteman]

12. Your analysis is flawed. Nate doesn't include Rasmussen

Please edit.

[Beetwasher]

15. Source It-He Uses Ras ALL Over The Place, Are You Fucking Kidding?

:rofl:

[BklnDem75]

16. What are you talking about?

It's the only poll he uses in North Dakota.

http://elections.nytimes.com/2010/forecasts/senate/north-dakota

[Valienteman]

18. True. I saw that. He adjusts for Rasmussen's +2R house effect though

http://fivethirtyeight.blogs.nytimes.com/methodology/

[Clio the Leo]

9. AND there is currently a greater chance that Dems will keep the House....

.... than there is that the GOP will take the Senate.

(but that's according to Nate Silver, you all can take it or leave it lol)

[budkin]

11. It's all about turnout... people still hate Republicans

[Teaser]

13. probabilities like this should be used to prioritize contingencies

and not much else.

[fiorello]

14. Worth comparing: Obama had a 99% chance of victory at this point in 2008

Nate Silver's statistics normally show the probabilities going to near-100% or near-0% as election day gets closer - unless the race is REALLY close.

Nate had Obama a 60% chance in summer, 2008... rising to around 80% after John and Sarah started laying eggs... and hitting the upper 90's as election day approached.

Right now the election for congress is still REALLY close.

[wisteria]

17. I am so angry at how the media has been manipulating these elections.

Just this morning in PA I heard on the radio that two polls give the Republicans unsurmountable leads against the Democrats and a majority in PA are angry at the Dems. This is all spin. And, it is being done to discourage Democrats from voting.