538's projections on undecideds

[they don't love you like i love you]

538 has drawn up a projection on how undecided voters will break. Thoughts?

DC    64.4%
MS    64.4%
GA    63.0%
MD    61.5%
SC    61.1%
AL    60.9%
NC    58.0%
VA    57.8%
IN    57.8%
IA    56.9%
AR    56.8%
OK    56.5%
WI    56.5%
DE    53.7%
AK    53.4%
WA    52.7%
FL    52.4%
TN    52.3%
CO    51.8%
MO    51.6%
MI    51.5%
KS    51.4%
OR    51.0%
LA    50.7%
UT    50.6%
HI    50.5%
MN    50.2%
NE    49.8%
TX    48.3%
IL    48.3%
MT    48.0%
OH    47.2%
NV    46.7%
WY    46.6%
SD    46.4%
AZ    46.0%
ND    45.5%
ID    45.4%
NJ    45.0%
PA    44.8%
CT    44.6%
NY    44.6%
VT    43.7%
KY    43.4%
CA    42.9%
ME    42.6%
NH    42.2%
MA    41.0%
NM    40.0%
WV    38.6%
RI    35.0%

Looks more realistic to me than the "OMG UNDECIDEDS WILL ALL BREAK 2:1 FOR MCCAIN!" the GOP hacks keep saying. The only state hwere that even comes close is Rhode Island, where of course it won't matter one iota.

[Bacon King]

In what crazy parallel world do NJ undecideds break for a Republican?

[they don't love you like i love you]

In what crazy parallel world do NJ undecideds break for a Republican?

It's based on a formula mostly based off the primary results. Should be tweaked a bit I'll admit, considering that John McCain is not Hillary Clinton.

[Lunar]

Completely a 100% stupid allocation.  Obama primary voters aren't undecided in the general election and McCain's primary results are either completely ignored or mostly ignored since most of his states are uncontested.  Not to mention it can measure the Palin effect.  Sure, in some states, it can measure black turnout increasing and so on, but c'mon.  538 dropped the ball yet again with the desire to calculate everything.

I think a 50/50 allocation is much more fair.

[daboese]

They have done it considering their deviations from the polling numbers.
Hence, the effects that they have taken into seem to match the numbers even better.
This is just another variable to play around it when fitting the data, and they felt like they should take it into account now. There's nothing wrong about that, possibly they will refine the model later.

[Swing low, sweet chariot. Comin' for to carry me home.]

fail

[Swing low, sweet chariot. Comin' for to carry me home.]

538 has drawn up a projection on how undecided voters will break. Thoughts?

DC    64.4%
MS    64.4%
GA    63.0%
MD    61.5%
SC    61.1%
AL    60.9%
NC    58.0%
VA    57.8%
IN    57.8%
IA    56.9%
AR    56.8%
OK    56.5%
WI    56.5%
DE    53.7%
AK    53.4%
WA    52.7%
FL    52.4%
TN    52.3%
CO    51.8%
MO    51.6%
MI    51.5%
KS    51.4%
OR    51.0%
LA    50.7%
UT    50.6%
HI    50.5%
MN    50.2%
NE    49.8%
TX    48.3%
IL    48.3%
MT    48.0%
OH    47.2%
NV    46.7%
WY    46.6%
SD    46.4%
AZ    46.0%
ND    45.5%
ID    45.4%
NJ    45.0%
PA    44.8%
CT    44.6%
NY    44.6%
VT    43.7%
KY    43.4%
CA    42.9%
ME    42.6%
NH    42.2%
MA    41.0%
NM    40.0%
WV    38.6%
RI    35.0%

Looks more realistic to me than the "OMG UNDECIDEDS WILL ALL BREAK 2:1 FOR MCCAIN!" the GOP hacks keep saying. The only state hwere that even comes close is Rhode Island, where of course it won't matter one iota.

I can see undecideds breaking for Obama if he was up 5% going into election day, or if he had good momentum and went from -3% to +2% the week before the election. 

BUT, if this race is stagnant, and Obama leads something like 47-45 going into election day, I don't see how the undecideds break his way.  Now, I've never studied how undecideds break, but I just can't imaging how a terribly underpreforming minority candidate, with a thin resume and a somewhat unidentifiable life story, running for an open seat in a stagnant election, has a chance of having the undecideds break his way.  I just don't see it.

And its not just me saying it.  Even the pollster guy on MSNBC believes the same and he has stated the view is shared by both the Obama and McCain campaigns.

[Sam Spade]

I would presume that a good pollster could first see who the undecideds were (demographically) and then push them into giving a tentative preference. 

I suspect the undecideds will break very differently depending on which state they are in.

[Alcon]

Terrible projection methodology, fortunately only worth a few points.

The idea that the swing voters in Mississippi will break that way is pretty hilarious.

[J. J.]

Obama is using a 3 out of 10 model, and for a number of reasons, I think he's right.

[daboese]

Anyone who had to do with modelling will see that 538 is just trying to adjust their models in order to predict the majority of the polls better.
Allocating the undecideds is a good idea at this point in time, and they have a reasoning to do so- even if they really take into account some other effects.

Their numbers base on:
--snip--
Specifically, Obama overperformed:

1. In states with high African-American populations;
2. In states that share a border with Illinois (no, Kentucky doesn't count);
3. In states with highly educated electorates;
4. To a lesser extent, in the South (as indicated by the number of evangelicals), even after accounting for the above variables.

Meanwhile, he underperformed his polls:

1. In the Appalachian states (as indicated by the number of respondents who identify their ancestry as 'American', a practice concentrated in the Appalachian region);
2. In states with low education levels;
3. And in states with a high number of Catholics.
--snap--