Bob Krum's Scenarios
Paul A. Rahe ·
November 5, 2012 at 4:40pm
If you want to have fun tomorrow night, let me suggest that you run off, read, and keep ready to hand the five election-day scenarios laid out over the last few days by Bob Krum -- to wit,
Scenario 1: Nate Silver is right
Scenario 2: RCP is right
Scenario 3: Rasmussen is right
Scenario 4: Gallup tracking poll is right
Scenario 5: Gallup electorate poll is right
I am inclined to go with scenarios 4 and 5 -- with one proviso: I fear that we will not do as well in the Senate races as Krum in these scenarios estimates.
- Comment (14)
- · Quote
- · UnfollowFollow (4)
Comments:
Nov '12
Re: Bob Krum's Scenarios
An interesting history of the accuracy of state polling, The Margin of Wrongness.
Sep '10
Re: Bob Krum's Scenarios
As far as I know none of the polling organizations claim to be predictive! The data collected is seldom properly and completely reported. If journalist cannot even manage this rather mundane task why would anyone waste time with their analysis?
Sep '10
Re: Bob Krum's Scenarios
If Rove is not lying about the numbers he is getting from the Ohio Sec. of State office on early voting Romney will win that state by 3-4 points, which means he probably has a better than even chance to carry PA, WI, and MI. You were correct all along Prof.
Jun '10
Re: Bob Krum's Scenarios
Just for fun, and if the five scenarios laid out by Dr. Rahe aren't sufficient, you can add Michael Barone's presidential prediction to the mix.
Or, it that's not enough, George Will has one out on TV that's a lot like the Barone prediction.
We certainly do not lack a shortage of folks out on a limb.
Edited on November 5, 2012 at 5:17pmRe: Bob Krum's Scenarios
I hope.
Jun '12
Re: Bob Krum's Scenarios
Bob Krumm's closing sentence in prediction #5 is, to my mind, overly optimistic. Democrats contemplating that their ideas are out of touch? Don't think so.
Sep '12
Re: Bob Krum's Scenarios
I've caught several interviews with prominent Evangelicals who say they have their get-out-the-vote operations fired up and cranking out Romney votes. These are the same votes that put Dubya over the top in 2004 but didn't go to McCain in 2008. In 2012 they're back in the GOP column and will hopefully put Akin, Murdoch, Fischer, Allen, Rehberg et al. over the top as well. We need eight wins in the Senate or the next two years will be fascinating primarily for their futility.
As I recall, one of the main digs on Romney was that he wouldn't be able to rally the Evangelical base. Well, that's not what I'm hearing.
Aug '10
Re: Bob Krum's Scenarios
Clearly this post is intended to sap the productivity of the Ricochet Out On The Limb wing. Now Rob Long et al will spend the day making spreadsheets and writing electoral college flow charts, instead of working!
May '10
Re: Bob Krum's Scenarios
Romney, 341.....or more.
Re: Bob Krum's Scenarios
Sadly, I still figure Nate Silver is the most apt to be correct. Now we don't even really have a national poll average to fall back on. You can rely on Rasmussen and Gallup if you like, but when you're staking everything on two pollsters of many being correct, it's a very tenuous limb you're resting on.
Mar '11
Re: Bob Krum's Scenarios
In my experience, banking on the majority of any group to be correct is a risky undertaking. A majority voted for Obama in 2008. How wrong were they about the guy? So a majority of pollsters lean slightly Obama? I say we win this one--no matter what the majority of polls say--and when we do, I get to be a rebel and be right--that's about as American as it gets.
Mar '12
Re: Bob Krum's Scenarios
I am looking at scenario #3 as most likely with WI and OH nailbiters. If this scenario pans out (particularly in OH with provisional votes) we are looking at 2000 in spades, since either WI or OH would give R the victory.
If scenarios #4&5 become reality, I think all pollsters will have to rethink their methodology.
I am still looking at VA as the canary. If BO wins VA, Silver's model will look reliable. If R wins 2% or less, scenario #2 scenario will look good. If R wins VA by 2%-4%. Scenario #3 is in play. If R wins VA by 5% or more then look for #4 or #5 to be the result.
Re: Bob Krum's Scenarios
The key is partisan self-identification. As I said in my post last night (scroll down), the polls showing Obama ahead presume that the pattern of 2008 is written in stone. Both Rasmussen and Gallup show the Republicans even or with a slight advantage in partisan self-identification. If this is true, they win and win --as in 2010.
Sep '10
Re: Bob Krum's Scenarios
Schrodinger's Cat:
If scenarios #4&5 become reality, I thinkallpollsters will have to rethink their methodology.
A look at Gallup's historical poll results show you that they can be off by up to 3 percentage points in modern elections. The Bush Kerry 49-49 tie in 2004 ended up being a 50.7-48.3 Bush win. The Reagan-Carter Gallup went into election day 47-44 and ended up being 50.8-41. This year is more like 1980 than 2004 and the GOTV weighs in favor of the Republicans.
I doubt Gallup at least will rethink anything.
Edited on November 5, 2012 at 11:04pm