political report v.1 i.5 January 31, 2008

four left….

In the final three days before the Florida primary, Rudy G couldn’t get more than 100 people at his campaign stops.  Yesterday morning Barack Obama came to Denver and drew 18,000.  Yes, eighteen thousand. I’m pretty sure that’s more than 50-Cent draws.  Bill Clinton spoke in the same arena last night and drew 3,000.  [Denver Post coverage: http://www.denverpost.com/politics/ci_8124935 and ci_8124939]

Last report was sent the morning of the Florida primaries.  I said Rudy would get out in short order and he did (shocking one super rightwing R Rudy supporter who gets this newsletter, rumor has it).  I gave a two week countdown on Huckabee but I’m going to change that to Feb 6, the day after SuperDuperTuesday.  And I said Ron Paul is going to stay in until May, but I’ll revise that: Ron Paul will stop actively campaigning soon but might not officially drop out until mid-April or later.

Edwards’ quick exit surprised me a bit in its timing.  I thought he’d stay in through Feb 5^th even though he and everybody else in the world knew he wasn’t going to pull any states.  But the reason he could stay is the difference between the D and R primary structure: the R primaries are winner-take-every-delegate in each state while the D primaries are proportional allocations.  This is why Obama really won Nevada with 13 delegates to Hillary’s 12 even though Hillary got more votes.  Hillary’s votes were concentrated in Lost Vegas while Obama showed well everywhere else in the state.  What this proportional allocation meant for Edwards is that he could stay in the race and pick up a significant number of delegates heading into the convention, and wield them.  I think his camp probably saw a scenario where neither Obama nor Hillary clearly rises out of the field on Feb 6^th, giving JE’s batch of delegates a political weight beyond their worth on paper.  But alas, JE apparently didn’t want to play that game.

As far as Rudy G, what I failed to anticipate on Jan 29^th is how obsequiously he would be endorsing McCain during his drop-out speech.  It seemed clear he’d favor McCain publicly, as the debates have sometimes been a Rudy-John lovefest.  But I didn’t expect Rudy to essentially get down on his knees on stage and beg McCain to make him Attorney General right there in January ‘08.  Geez.  Give it a rest, Rudy, you’re in already.  You’re in my cabinet.  Please get off my shoes now.

political report v.1 i.4 January 28, 2008

all chewy center, less crunchy outside…Last report had us bemoaning the loss of Gov. Richardson, wondering why Ron Paul is still around, bashing poll mathematics, stating (predicting? naw…) that Barack might sweep the south, and talking about perception vs. reality.  This report?  More of it.

In campaigns, the perception vs. reality management game is also wound up with expectation and trends.  As I said last time, Mitt quickly moved out of South Carolina a few days before the R primary because he was going to come in fourth (he did), and he wanted to manage the ugly perception conferred by a 4^th place finish in a state he spent money.  But Mitt’s big loss there was tempered by his enormous win in Nevada (he was really the only R campaigning there, though), and suddenly it seems pretty clear that the last two dogs standing on the R side are Mitt and McCain.  I still give richest odds to McCain, but Mitt’s Kerry-in-drag routine and billions in personal cash might actually pull it out.  That’s what the D’s are hoping anyway, because Romney has about as much chance as Dukakis in the general.

All In The D Family   More importantly on the perception-reality-expectations-trends front, though, is the Barack-Clinton battle.  Here’s what you need to know about last Saturday’s South Carolina D primary: a week before the election, the Clintons - mostly Bill but a little Hillary, too - started getting nasty.  Some observers even used the word “sleazy” and various synonyms thereof.  The “Clinton Machine” was starting to do anything it could to win.  Bill might have been the first black president (Toni Morrison’s words), but that voting segment was moving toward Barack.

At the same time as the Clinton Machine started rolling, for reasons coincidental and not, Barack’s lead seemed to start slipping somewhat quickly, and in the perception/trends game, that meant a lot.  If Barack won SC but did so by only a few points, giving up a sizable lead in the process, it meant that the Clintons had the momentum back.  What happened instead is that Barack turned a 10-12 point polls lead into a stunning 28-point win.  (I’m not using “stunning” lightly.)  He took 80% of the African American vote where just a month or two ago Hillary was ahead on that tally in the state.  And he took about 25% of the white vote with Hillary and Edwards splitting the rest evenly.

A few things to take home: First, turn-out in SC was enormous, as it has been in all other previous states.  People are revved about this election, and we don’t need the numbers to tell us that; I can hear it from everybody I talk to or overhear.  (Even the R’s seem to be more excited about this election than previous ones, although not nearly to the level that the D’s are.  That alone tells us how tough it’s going to be for the R’s in November, but the party preference among undecided/independents is also strongly in favor of the D’s right now.)  Second, momentum is clearly to Barack.  Trends are toward Barack, endorsements continue to fall to Barack, and the only thing blocking his roll to SuperDuperTuesday (Feb 5^th; it used to be Super Tuesday but a bunch more states are voting that day this year than in years past) is Hillary’s inertia.

political report v.1 i.3 January 17, 2008

Last report had us mulling Hillary’s last minute pull-up in NH.  The wise-o’s in DC, NY and LA have it all figured out: it’s the Bradley effect!  That is, when white voters who had previously told pollsters that they were going to vote for a black candidate actually get in front of the lever, they pull white.  It’s supposedly what happened to Tom Bradley in the 1982 California governor’s race.  (It then supposedly happened to Doug Wilder and David Dinkins.)

Come on people!  Let’s call BS when we see it.  It may have happened to Bradley but it certainly didn’t happen to Dinkins in NYC and even if there is a real “Bradley effect” out there, it didn’t happen to Barack Obama in New Hampshire.  The “Bradley effect” is not an effect but an artifact of bad polling.  Pundits who would have you believe in a Bradley effect would have you believe that exit polls make more sense than pre-election polls.

The best polls out there get a ~85% accuracy rate and the worst get about ~65%.  What does this really mean?  That the reported margins of error are meaningless.  That shouldn’t be surprising though, as margins of error are derived by mathematicians, not by political analysts or sociologists.  Margins of error are based on simple population-sample statistics.  The problem is, groups of humans don’t behave like groups of tadpoles or temperature measurements or high-energy cosmic particles.  So when a pollster calls 500 Colorado Dems and asks them for whom they are going to vote, as long as the sample was taken at random the poll results are assumed to scale up to the entire population with a margin of error that is set by the relationship of the sample size to the entire population size.  Only a mathematician would think this is anything other than utter BS.  The pollsters explain away their inaccurate results by citing sample biases: some groups are more likely to engage in polls than others, we didn’t get a truly random sample, etc.  In one analysis of New Hampshire, the weak and poor don’t engage in polling, and that’s who voted for Hillary in NH.  Of course, there’s also the fact that Clinton was listed at the top of the D ballot and Obama at the bottom, and a Stanford prof who studies voting patterns has shown pretty conclusively that that effect alone is good for 3 percentage points, which is how much Hillary won NH by.

political report v.1 i.2 January 9, 2008

Like I said in the last email:

At this point best odds are for an Obama presidency, but politics is like college basketball (wildly unpredictable with national polls and national media opinions often meaning nothing). If it wasn’t for a breaking sex scandal late in Obama’s Senate race that sunk his R opponent, we might not even know his name.

Thus Hillary took her last-second tear-up and made up 13 or so points in a few hours. Or did she? Despite the confetti and balloons, I’m not so sure Hillary reversed Obama’s momentum by eking out in NH. As I also said in the last email:

In my view of this, if Hillary loses the perception of the D primary voters that she’s the best bet to win Nov’08 then she is totally finished.

The NH exit polls have Obama winning on the Nov’08 electability question among registered Dems, and despite my slight disdain for exit polls, that should be a very concerning number for the Clinton camp. Just last month Hillary was winning huge on the electability question (54-22), but ABC News/WaPo had Obama winning that crucial question 44-35 yesterday. To my eyes that’s a shocking reversal. (For what its worth, Pollster ranks ABC News/WaPo as the most reliable of the polls.) Then again, primary voters (both sides) are either a- not always the most reliable judges of general electability, or b- don’t vote on that issue (your guess as to what happened with Kerry in ‘04).

I see a bunch of other distressing signs for her as well. The main one is that the top issue among NH voters (as it was for Iowa voters) was “change” (55% of voters said it was their top priority) and Obama beat Hillary on that 55-28. Obama has also pulled even on the “strongest leader” question. However, surprisingly Obama comes in low on the “empathy” vote, with Clinton and Edwards in the 40% range to his 20%, and that’s not a number to dismiss. And Clinton wins outright on the “best commander in chief” question.

political report v.1 i.1 January 7, 2008

[This was the original political report (volume 1 issue 1), sent to 10 people on Jan 7. Other than the steamroll in NH comment, it turned out to be right on.]

It looks like Obama is going to take his Iowa momentum into NH and steamroll Hil (the Gallup poll shows him making over 15% on Hillary in a month, which is incredible). I think that’s going to continue indefinitely and I wouldn’t be surprised if he takes every state from here on out. The public mood in the polling seems to suggest that by winning Iowa he passed a general Nov’08 electability test that Clinton was previously winning. (In my view of this, if Hillary loses the perception of the D primary voters that she’s the best bet to win Nov’08 then she is totally finished.) You can see it in the R’s, too, as they seem to be shifting into a mode of “who can beat Obama?” (The answer is likely nobody.)

On the R side it looks like the better part of reason is finally winning out as McCain — the only R who can win in Nov’08 — is back on top of Romney, who is utterly unelectable and not because of his religion but because he’s the 2008 R version of a pandering John Kerry. Independent voters (i.e. the voters who matter in November) show again and again that they don’t like panderers (except for the exceedingly rare and highly-skilled panderers like Bill Clinton), but party bosses aren’t always the smartest people and are blinded by money in the primaries.
The NH primary is a fascinating lab for the national electability test because independents there get to vote in either primary (but not both). It’s the ‘I’ vote that is pushing Obama so high over Hillary, which bodes very well for his Nov’08 chances. At the same time, McCain finds himself competing not just with Romney (Huckabee isn’t even anywhere close in NH, getting no boost from Iowa) but also now with Obama for the ‘I’ vote. The exit polling there will be a very good early indication of how competitive a McCain-Obama contest would be, as both are (perceived as) centrists and the Nov’08 election will tip on the independent vote (McCain is actually a lot more conservative than you think but perception is much more important than reality).

At this point best odds are for an Obama presidency, but politics is like college basketball (wildly unpredictable with national polls and national media opinions often meaning nothing). If it wasn’t for a breaking sex scandal late in Obama’s Senate race that sunk his R opponent, we might not even know his name.

Obama (or whoever the D nominee ends up being) likely does not need to pull the western mountain states to win. All (s)he needs to do is repeat Kerry’s states and add Ohio which, considering the political mood there, should not be difficult. But Obama could do very well in the more-purple-than-believed-by-the-national-politicos states of CO, AZ, NM, MT and NV by pulling in a western running mate who is pro-gun and pro-hunting. Then the steamroll would be on.

Fun times ahead….