Saturday, November 7, 2009

Scorecard for Election 2009 Predictions, What the DEMs Should Really Be Afraid of in 2010, Critical Health Care Vote Upcoming

Election 2009 -- How'd We Do?
The dust has settled on the 2009 "mini-midterms" and as always, I wanted to take a look back at the actual vote tallies versus my projections and see how we did from a prediction standpoint and what may have driven the differences. Note that in New Jersey, I will be focusing on percentages of the "three-party" vote (two two major parties plus Chris Daggett), in Virginia on the two-party vote. As I did not make a projection on write-ins or minor third party candidates, their tallies (which are very small percentage anyway) are ignored.

1. Virginia
My Final Projection: McDonnell 57.0%, Deeds 43.0%
Actual Two-Party Vote Tally: 58.7%, Deeds 41.3%
Average Miss per Candidate: 1.7%
Total Bias: +3.4% Democratic

Obviously, we projected the right candidate to win and win handily, but this site was hardly unique in that projection -- just about everyone was projection McDonnell to win. The error, while higher than I would like, is not as bad as I thought on election night, when the early returns showed an even strong McDonnell lead.

In total, I think part of the miss was just the random error associated with attempting to call a statewide race and part of it was due to lower Democratic and higher GOP turn out. With a race this lopsided, turn out is very difficult to predict, since most voter know that their vote is highly unlikely to make a significant difference.

The broader implications of this race are unclear. Democrats ran a bad candidate who had already lost a statewide race previously to McDonnell. The White House and national party provided little support to Deeds. But, still, this was a major butt-kicking by the GOP to a party that had started to dominate Virginia politics. So my take on the implications: this isn't damning to the DEMs in 2010, but it sure isn't good news.

2. New Jersey
Note: As of this writing, Monmouth county had, for some reason, not reported results and this therefore reflects only about 95% of the total vote in New Jersey. Monmouth is not anticipated to significantly change the percentages.

My Projection: Christie 45.0%, Corzine 44.1%, Daggett 10.9%
Actual Three-Party Tally: Christie 48.1%, Corzine 46.1%, Daggett 5.8%
Average Error Per Candidate: 3.4%
Total Bias (Error on the Margin): +1.1% Democratic

You can clearly see where I missed here, Daggett got far less votes than projected. We made the right call on the outcome of the race with Christie winning and the final margin was within a very respectable level of bias, calling the race within 1.1%.

The two potential causes of a Daggett fade that we discussed likely came into play: First, that in a close race, many voters would not want to "throw their vote away" and therefore switched to one of the two major candidates and second, Daggett was buried so deep in the ballot many voters who may have polled for him may have switch in the booth to one of the two major candidates.

While we can always discuss these possibilites (as we did), there is no statistical way to predict them happening, therefore in odd circumstances like this, there is always a good possibility for error.

The implications in this race are probably more severe for national Democrats than in Virginia. Unlike in Virginia, both the national party and President Barack Obama worked hard to re-elect Corzine. His loss shows that Obama is not a silver bullet for victory. And it also shows that even in blue states, people are very mad about the economy and taxes, the two primary issues in the New Jersey campaign.

3. NY-23
I couldn't make a statistical projection here, but I give myself credit for calling this race to be a close battle, whereas most pundits and projectors viewed Hoffman as having clear sailing. I didn't call the Owens win, but I certainly alerted you to the scenario that unfolded -- Scozzofava voters deciding Hoffman was just too conservative.

Democrats, with little else to celebrate this election, have tried to turn this into a predictor of 2010. It's hard to imagine that every house race in 2010 will feature the crazy drama that unfolded here. This race is unique and as such probably projects little. The one piece of solace that Democrats can take away is that even a re-eneregized GOP is fighting for its ideological soul and that was on full display in this race.

Be Afraid of Unemployment, Not 2009
So should Democrats be afraid in 2010? Absolutely. But not because of what happened on Tuesday, but because of what was reported on Friday. The unemployment rate in October spiked to 10.2%, a new 23 year record. There are now less people employed in this country than there were in 2000. And the reality is that it is unlikely to be pretty by November of next year.

I admit I seriously undercalled the lag between economic growth returning and unemployment dropping. Consider this -- even if unemployment were to begin IMMEDIATELY dropping by 0.1% per month (and few believe we have bottomed out yet -- most are saying we won't until at least December) -- we would STILL have a 9% unemployment rate next November.

Democrats will try to pin this all on George W. Bush and there is some fairness to the argument that macroeconomic changes don't happen over night. But voters don't want to hear it. It seems highly likely they will be feeling a lot like throwing all the bums out a year from now. And there are more Democratic "bums" than Republican "bums" at the moment.

Will the Democrats lose the Senate? I still can't envision a scenario.
Will they lose the House? Possible, as I've said, but still unlikely.
Will they lose seats? Oh yeah.

And if they are struggling to get major legislation passed now, with wide majorities, imagine what it will look like if there are, say, 20 more House Republicans and 5 more Senate Republicans.

We could be in for 1990s style gridlock. And maybe that isn't such a bad thing. Remember those balanced budgets and 3% unemployment rates in the 90s? Sounds nice right about now.

House Health Care Vote Upcoming
The House is just convening for a rare Saturday session to push towards a vote on Health Care Reform. A late compromise was struck between House Speaker Nancy Pelosi and moderate anti-abortion Democrats, who favored coverage but wanted assurances that funding wouldn't go to abortions. The moderates will be allowed to offer a floor ammendment to the bill that would specifically prohibit funding from the bill to be used for abortions. As of this writing, that is the only amendment that is likely to be allowed.

Pelosi had to give ground. The Republicans are united against the bill, meaning that Pelosi can only give away 40 votes. Counting the Blue Dogs who oppose the included public option and anti-abortion Democrats, the numbers were adding up more than 40. She had to give ground on one of the two to get a bill passed.

Passage is still not assured, nor is the vote timing. If the Democrats don't feel that they have the votes, they may delay a vote from Saturday evening until Sunday.

This is high drama to watch, but even assuming the Dems get is passed, the key roadblock is the Senate. Keep in mind that the House passed a Cap and Trade bill, the other key domestic policy priority for President Obama, months ago and the Senate has yet to act.

Harry Reid has indicated that the year-end "deadline" President Obama had set (actually, his original deadline was the August recess, but no matter) may slip as the Senate still grapples with how to get to 60. It only gets tougher the later this bill goes...primaries for 2010 start in February. The only way I can see Reid straddling the middle is to give in to Sen. Olympia Snowe's (R-ME) proposal to replace an automatic public option with a later-date "trigger mechanism" public option. This would bring along Snowe and Sen. Joe Lieberman (I-CT) and presumably Sen. Ben Nelson (D-NE) as well. Without Snowe and Lieberman, the DEMS are at best at 59 votes and possibly only at 57 or 58 (depending on how Nelson, Specter, etc. vote.)

I love watching congress work the weekend. We should make them do it more often.

Next post I'll cover the votes, plus get back to our normal polling updates. It is, after all, time to shift the focus to the 2010 mid-terms.

Thanks for reading. If you like this site, tell your friends. And if you have some thoughts on the 2009 election results, send me a note or post a comment.

Incidentally, we had 131 people view this site on election night and 170 view it in the 48 hours surrounding the election. This is a record for the year, although I suspect lower than November of 2008 (I did not have tracking in place at that time.)

No comments: