For a look back at 2008, Pollster.com has an excellent roundup. Did you know that 30% of all votes were cast early? That more than ever, the 2008 race was dominated by a uniform national swing of opinion? Great stuff. The last thrashing of the Coleman campaign is here; an exhaustive analysis is here (seems partisan but kosher). The events of Campaign 2008 showed that the Meta-Analysis is a useful tool to cut through the day-to-day noise of political commentary. I’m satisfied in proving my point that simple poll aggregation is a near-optimal approach (see some recent commentary by Pollster and Chris Bowers).

My remaining project is to write an article about what the Obama v. McCain race looked like with the noise removed. Looking back, the turning points were few. Political science / econometric models pointed toward a convincing Obama win. In that light, swings toward McCain were temporary deviations; the Republicans did as well as they could. Even the VP pick by McCain was fine. It was substantively awful, but considering the brief boost he got it was probably the best he could do. A Hail Mary pass, but with a significant repercussion for the Republicans: future prominence for Sarah Palin.

* * *

I’m considering a rearrangement of the site to reflect future interesting topics. But I’m not sure what those topics would be. Some of you have asked for 2010 midterms coverage. But averaging polls gets you nearly everything you need. For this, Pollster.com is your best source of relatively pure data. I’ll continue to offer their data stream – see the left sidebar.

The question is how to turn the spirit of this site into something interesting for the future. One possibility is to turn this site into a statistically-oriented look at other real-world events. I welcome your suggestions in comments.

For the moment, other projects beckon. I’m catching up on writing papers and grants, and the important people in my lab who make it all happen. I’m also promoting the paperback release of Welcome To Your Brain. And in the spring I’ll be guest-blogging for evolutionary biologist Olivia Judson at the New York Times. Look for me there!

And tomorrow I will watch Obama being sworn in. A win for the Democrats was highly likely from the start, but that doesn’t change the historic quality of the Inauguration. A recent ABC/Washington Post poll reports that 78% of people in the United States think our country is on the wrong track, and 33% approve of Bush’s performance. But an amazing 80% approve of Barack Obama’s performance so far. It’s a time of tremendous change for our country – change and hope.

I don’t ask for donations to keep this site going. Instead, I ask people to make efforts where they will have the greatest leverage. That’s the point of the “power of your vote” calculation in the right sidebar, which shows the power of individual voters – for instance, through get-out-the-vote (GOTV) efforts.

You might be surprised to know that the best place to put your donations is not the national race. After the jump is an argument for giving to knife-edge races. If you agree, go to my ActBlue page. My advice pertains to Democrats and Republicans alike, so here is the NRSC site.

Previously I recommended an ad by the Defenders of Wildlife Action Fund. However, the national race has moved decidedly toward Obama. I am not claiming that the race is over – quite the contrary. I am saying that donations at that level will make a nearly-unmeasurable difference in the outcome.

Your money is beter spent on Senate races. This year, each vote gained is a step toward a filibuster-proof majority (60 votes). In the other direction, the NRSC points out that the Republicans are just two votes away from regaining a majority – highly unlikely, but that’s what they say.

Furthermore, you should focus on races that are neck and neck, where your donation has the most leverage. Currently, the closest races are ones where the incumbent is a Republican: Oregon (D-Merkley vs. R-i-Smith) and Minnesota (D-Franken vs. R-i-Coleman). Assuming that other races go to the candidate with a >80% win probability, if Democrats win both races the Senate will be 58-42. If Republicans win, it will be 56-44.

I’ll update this link as races shift. Please give, and thank you for reading the site.

Update: I’m adding Georgia (D-Martin vs. R-i-Chambliss), which currently has Chambliss up by only 3%. I’m broadening the definition of toss-up to races that would become toss-ups after a 2% swing. A Democratic win would bring the Senate to 59-41.

Sam Wang

P.S. Also consider getting my book Welcome To Your Brain: Why You Lose Your Car Keys But Never Forget How To Drive. It’s science, not politics. But it’s a good read!

In a recent online chat at the Washington Times (yikes!), Nate Silver of FiveThirtyEight wrote that the Meta-Analysis presented here answers a “largely meaningless question.” Coming from him, I find the comment surprising. One should be careful not to throw stones in a glass house. My reply:

The Meta-Analysis is a high-precision snapshot of where the race stands today. The history shows you where swings occurred – and where they did not. The “Celebrity” ad campaign? Effective. Obama’s Europe/Asia tour? Not effective. McCain’s houses gaffe? Bad. If you like the national and state trendlines at Pollster.com, the Meta-Analysis is the next logical step. If you like seeing the state of play in the game, and not just an informed guess at the ending, tell Silver in another online chat he and Charles Franklin are doing at the Washington Post.

And now, my own stone…

His calculation starts from the same data, then adds adjustments that include a major guess about future movement and an asymmetric assignment of undecided voters. This approach raises questions.

The latest change to Silver’s model is an adjustment to assign undecided black voters to Obama. He calls this a “reverse Bradley effect.” As many readers of this site know, I am not a fan of “adjustments” unless they are extremely well grounded. His rationale is that Obama overperformed his polls in the primaries. This is thin ice. In recent work on the disappearing Bradley effect, Dan Hopkins pointed out that front-runners often do not do as well as expected from polls. If we think of Hillary Clinton as the front-runner, then this could explain the anomaly. Another possibility is that the discrepancy came from first-time voters. In that case people who voted in that primary have, by doing so, made the transition from registered voter to likely voter.

In other ways, Silver’s methods are generally acceptable, though some of the details add uncertainty more than they improve accuracy. He is a fiend for numerical information and has a good feel for what’s interesting in polling data. For example, today he noticed that even though the debate did not lead to a jump in Obama’s popular lead, it did strengthen enthusiasm for him among his supporters.

In 2004, the median Electoral Vote estimator got noticeably spiky in October. The following graph uses the last-3-polls rule alone:

in large part because individual polls passed through the average quickly.

At the very end of the campaign we made a last-minute switch to a last-three-polls or last-7-days (counting back from the date of the most recent poll), whichever gives more data for a state. This performed extremely well on Election Eve 2004, nailing the result. We’ll start using the new rule tomorrow morning. The rest of the procedure – from calculating the median and SEM onward – will stay the same.

To reflect the posting schedule of our data partner, Pollster.com. the EV estimator will now be updated four times a day at 8:00am, noon, 5:00pm, and 8:00pm. We will stop doing the midnight update since Eric Dienstfrey and Mark Blumenthal haven’t been putting up data at night. (Good for them – they need their rest.)

In the closing weeks we’ll continue to more technical topics. Possibilities include

• Further analysis of systematic error
• A shift to prediction, including nonpresidential races and quantitative oddsmaking
• The psychology and neuroscience of political opinion formation

Stay tuned.

The ad is persusasive to independent voters and moves opinion toward Obama, as shown by this survey. Independent voters expressed a degree of skepticism intermediate between Republicans and Democrats, but all groups found the content to be disturbing. Scores for other ads are found here; the second place scorer was McCain’s “Pump” ad.

You can sponsor the Defenders of Wildlife ad to be shown in Ohio and Florida. Use this link.

In terms of brain mechanisms, this ad has two notable features that I’ve written about as being good for making an idea stick. First, it appeals to disgust, an emotion that plays a powerful role in memory formation. Second, the low skepticism score among independents suggests that it may escape the phenomenon of biased assimilation, in which we tend to reject evidence that disagrees with our existing worldview.

For those on the other side, here is a link to the National Republican Senate Committee. This is a relatively good target for donations since the Senate is a point of strong leverage for Republicans, even as the minority party, which seems likely. And as the site says, more optimistically, “Two seats to recapture the Senate.”

Later I’ll have further suggestions for how to decide where your donations are most effective. One central idea is to identify races where the win probability might be movable by donations. I wrote about this in 2004. Therefore I will be recommending support for toss-up races; these will be the same races for either side.

The map graphic itself, now visible on the main page, shows state-by-state win probabilities for Obama and McCain, as well as states that are safe for them as of today. It’s static, and clicking it gets you a bigger version.

However, if you click on the lines below it, you get something more interesting: an interactive pop-up window. You can click on the map to change states, and right-click to return the results to their starting values – or to the 2004 results.

The “With +2% for Obama” and “With +2% for McCain” lines show what would happen if state margins all moved by 2% toward either candidate. This is a swing that can easily happen in a few days. It’s also the kind of hidden effect that could exist with certain kinds of polling error. If you are convinced that such an error exists, you can see for yourself where such an error would put the race.

The voter Influence tab is now better documented. This is an index of how much one person’s vote is worth in terms of swinging the overall election outcome. It’s a useful tool for you to ascertain where your efforts are most efficiently spent, for instance for voter canvassing. It applies equally to Obama and McCain supporters, and should be useful to any reader of this site.

Finally, the electoral vote history graph has been updated slightly to include the Democratic convention and the Republican convention. Because state polls stopped during that period, only the net effect is visible in the history itself.

There are likely to be more features (or adjustments) as Election Day approaches. We’re still cogitating on what they might be…