Sasha Issenberg: the new science of winning campaigns
November 8, 2012 2:12 PM   Subscribe

A Vast Left-Wing Competency: "How Democrats became the party of effective campaigning — and why the GOP isn’t catching up anytime soon." Sasha Issenberg, author of The Victory Lab, has been writing a series of posts on Slate that focus on different aspects of "the new science of winning campaigns".

Some highlights from the Slate series:
*Obama Does It Better: "When it comes to targeting and persuading voters, the Democrats have a bigger advantage over the GOP than either party has ever had in the modern campaign era." (October 2012)
*Why Obama Is Better at Getting Out the Vote: "Lots of door-knocking—and years of statistical analysis." (November 2012)
*Stuff Some Envelopes, Then Ask Questions: "Reporters should have to volunteer for a campaign before covering one." (September 2012)
*High Stakes: "Do campaign signs work?" (January 2012)
*Obama’s White Whale: "How the campaign’s top-secret project Narwhal could change this race, and many to come." (February 2012)
*The Death of the Hunch: "Campaigns used to guess which ads were most effective. Now they can prove it. How Obama’s embrace of empiricism could swing the 2012 race." (May 2012)
*It All Comes Down to Race: "Your opinions on health care reform, taxes, and even the president’s dog come down to racial bias." Previously. (June 2012)

Archive of all of Issenberg's pieces for Slate

*a review of The Victory Lab at The New Republic
*Issenberg discusses his book on PBSNewsHour (video interview, ~8 min.)
posted by flex (103 comments total) 44 users marked this as a favorite
 
Do it again in 2014, that'd be REALLY impressive.
posted by Artw at 2:13 PM on November 8, 2012 [18 favorites]


Ctrl-F "policies".

Nope. That's kinda a shame.
posted by Jimbob at 2:16 PM on November 8, 2012 [16 favorites]


With an eager pool of academic collaborators in political science, behavioral psychology, and economics linking up with curious political operatives and hacks, the left has birthed an unexpected subculture. It now contains a full-fledged electioneering intelligentsia, focused on integrating large-scale survey research with randomized experimental methods to isolate particular populations that can be moved by political contact.

BaloneyBall
posted by chavenet at 2:18 PM on November 8, 2012 [6 favorites]


The 'do campaign signs work' article is short, and actually pretty good. Bookmarking it as a go-to for when the question comes up again.
posted by gimonca at 2:18 PM on November 8, 2012


Not to worry, the Democratic Party is proven quite effective at snatching defeat from the jaws of victory.
posted by Old'n'Busted at 2:19 PM on November 8, 2012 [7 favorites]


With the two parties fighting for advantage with quadrennial advancements in voter targeting and similar campaign tools, I can't help but think of the dialogue about escalation in the last scene of Batman Begins.
posted by Bromius at 2:21 PM on November 8, 2012


The GOP has closed its decision cycle to outside input. That's why they were so surprised by their loss.
posted by wuwei at 2:21 PM on November 8, 2012 [19 favorites]


For a party that can't campaign effectively, the GOP sure does seem to hold a hell of a lot of state legislatures and governorships. Oh...He's just talking about the national level? Where they still hold the House by the balls?

Look...I'm thrilled as shit that Obama won, and a handful of Senate races went the Dems way. But this dancing-on-the-GOP's-grave stuff seems a bit premature.
posted by Thorzdad at 2:21 PM on November 8, 2012 [32 favorites]


The Obama campaign I was a part of in 2008 was highly impressive. This time it was astounding. A well-oiled machine snatching up votes.

Looking at the GOP reactions to this loss, it doesn't seem like they're going to substantially change their ways in two years - at least, not toward the center. If the economy grows and the public sees the GOP as the same old obstructionists, I feel good about Dem chances in 2014. (No sex scandals please Barry)
posted by East Manitoba Regional Junior Kabaddi Champion '94 at 2:23 PM on November 8, 2012 [3 favorites]


I really want to believe that perhaps, just perhaps, Obama won because he was the better person with a better vision, and most of the country could recognize that fact.
posted by HuronBob at 2:24 PM on November 8, 2012 [10 favorites]


Not to worry, the Democratic Party is proven quite effective at snatching defeat from the jaws of victory.

Yeah, I’m pleased with the results, but before I even read the articles I’m going to laugh at the idea that the Democratic party has become some sort of model of efficient organization. Let’s not pat ourselves on the back too hard.

"I am not a member of any organized party — I am a Democrat." -Will Rogers
posted by bongo_x at 2:25 PM on November 8, 2012 [4 favorites]


Do it again in 2014, that'd be REALLY impressive.

Look...I'm thrilled as shit that Obama won, and a handful of Senate races went the Dems way. But this dancing-on-the-GOP's-grave stuff seems a bit premature.

Yeah, that's something that struck me today when skimming through seemingly-endless 'How did he do it?' articles. Story time: In 2008 Democrats held Texas House District 23, which is primarily composed Democratic-leaning border counties. In 2010 the incumbent was ousted by a Republican, basically due to low voter turnout. This year, despite Republican redistricting to make it a safer district, Democrats won it, again, based again on turn-out in border counties.

If Democrats in Texas can only turn out voters when there's a presidential candidate they like, there's not really a good path to any functional state-wide progress. It's hard to groom candidates for state-wide office, which is something we see over and over again in Senate and Governor races (mediocre candidates with no name recognition and no particular skill at campaigning).
posted by muddgirl at 2:31 PM on November 8, 2012 [7 favorites]


The weirdest thing is all the early campaign postmortems suggesting no one in the Romney camp even saw the loss coming.
posted by 2bucksplus at 2:37 PM on November 8, 2012 [4 favorites]


Yay for psychohistory! All hail Hari "Nate" Seldon!

And the Prime Radiant is currently being built in a windowless building in Utah, don't ya know.
posted by RobotVoodooPower at 2:40 PM on November 8, 2012 [3 favorites]


Crazy idea, let's look at the links in this post.
posted by shakespeherian at 2:40 PM on November 8, 2012 [4 favorites]


Man, I know it feels good to pat yourself on the back after a victory, but I have some words for everyone: sic transit gloria mundi.
posted by mullingitover at 2:41 PM on November 8, 2012


The Democrats only added 3 seats in the house, and the Republicans only lost 3. In the senate, the Democrats only added 1 seat, and the Republicans lost 2.

This, and winning a presidential election against a rich, entitled politician during a time of economic crisis hardly seems like enough to make the Democrats experts at winning.
posted by TheyCallItPeace at 2:48 PM on November 8, 2012 [4 favorites]


It helps if you don't go out of your way to alienate every voter that isn't a member a single shrinking aging demographic.
posted by localroger at 2:48 PM on November 8, 2012 [3 favorites]


Barack Obama is good at winning elections. Transmitting that to the entire Democratic Party is another matter entirely, and 2010 suggests it doesn't work all that well.
posted by Holy Zarquon's Singing Fish at 2:49 PM on November 8, 2012 [9 favorites]


Didn't we used to hear about how great Karl Rove and the Republicans were during the George W. Bush years at "microtargeting" their potential voters?

I'm pleased that Obama had a well-run campaign that used cutting edge methods. However, the Republicans have plenty of smart people and are capable of creating well-run campaigns.

Also, one advantage the President had this time around was that he was already known by the voting public which made him relatively immune to the wave of negative ads. Next presidential election, that may not be the case for the Democractic nominee.
posted by Area Man at 2:52 PM on November 8, 2012 [1 favorite]


Another article declaring some blanket belief to be true. If I have to read one more "Democrats win forever" or "Republicans lose forever" article I'm gonna lose my shit. Get a grip weirdo writers.
posted by IvoShandor at 2:55 PM on November 8, 2012 [4 favorites]


I'm thinking that every election season will be like every game of Catan: different board, every time, and we haven't even begun to truly discover the number of permutations of the board.
posted by JoeXIII007 at 3:11 PM on November 8, 2012


What does it mean to say that The Democrats are not organized?

I have run a number of elections and worked on several more. The national party rarely did anything because few of these elections mattered to them (safe districts). The local party, however, is a different story. In New York I rarely ever had to work with them, they were an empty organization. And this is from someone who helped engineer a takeover of local committee seats; a full on party machine cog.

Contrast that with my current home in Oregon where I get frequent calls asking to work for candidates. I have an event calendar emailed to me simply because I requested to via their website.

If you want to really experience how organized or disorganized your party is I urge you to get involved in local party politics. Run for committee, or however your local party organizes itself. Not only will you learn more about the party, it is the first step in changing the party for the better.
posted by munchingzombie at 3:22 PM on November 8, 2012 [6 favorites]


Where was this analysis in 2010 when the Democrats got done?

Hasn't an incumbent only been beaten 3 times out of 11 attempts and without a serious challenger as in many of the losses how many times has it happened?

Bush, a draft dodger, beat Kerry, an actual war hero, after making the US insolvent and fighting two wars that at the time were going badly. That's at least as big an achievement as Obama beating Romney.
posted by sien at 3:25 PM on November 8, 2012 [4 favorites]


The Democrats only added 3 seats in the house, and the Republicans only lost 3. In the senate, the Democrats only added 1 seat, and the Republicans lost 2.

This, and winning a presidential election against a rich, entitled politician during a time of economic crisis hardly seems like enough to make the Democrats experts at winning.


Well, the other way of looking at it is that a black guy presiding over a sour economy and 8% unemployment won re-election with a margin similar to that of other recent presidential victories, and had coattails large enough to expand his party's majority in the Senate despite the majority of the seats up for election being held by Democrats, and reduce the strength of the opposition in the House. I think that's quite a sizable feat.
posted by LionIndex at 3:30 PM on November 8, 2012 [6 favorites]


It's seemed to me that there's only been one well-run campaign in each presidential election since 1992, and the candidate with the better-run campaign has won every time. Clinton's 1992 campaign inspired a documentary, so Bob Dole could've adopted some of the best practices, but didn't. Bush ran better campaigns than Gore and Kerry, and Obama ran better campaigns than McCain and Romney. A key example of the Obama-Romney campaign is when Romney announced the Ryan VP pick on a Saturday (!) and the Obama campaign had a response within an hour.
posted by kirkaracha at 3:32 PM on November 8, 2012 [1 favorite]


The two guys that have beaten incumbents recently--Reagan and Clinton--were both charismatic and their campaigns had positive, upbeat messages. Romney was uncharismatic and had no real message (or certainly not a positive one at least).
posted by kirkaracha at 3:34 PM on November 8, 2012


I think Obama is good at campaigning. The Democratic Party still sucks. They can barely get obviously competent elected while the GOP put Michel Bachmann in Congress, despite her being mad as a hatter.
posted by fshgrl at 3:35 PM on November 8, 2012 [1 favorite]


I agree with Thorzdad and TheyCallItPeace. I'm very impressed, again, with the Obama campaign, who are very, very smart and up to date. But I'm underwhelmed with the Democrats. Failure to win state legislatures and governor-ships has resulted in districts so gerrymandered that the Democrats couldn't recover House control at a time when approval of the House is around 20%. It's nice that the Senate has a couple more Ds, but that isn't a filibuster proof majority, and it mostly happened because at least two Senate R candidates shot themselves in the foot with idiotic comments about rape and pregnancy.

Everyone can learn a lot from the Obama campaign . . . and I'm sure the Rs will do just that.

I also wonder when, if ever, Democrats will take up some critical issues which are infringing the right to vote and distorting our politics: 1) the need for a constitutional amendment to repeal Citizens United and allow meaningful campaign finance reform and transparency; 2) the need to put districting in neutral, nonpartisan, professional hands and 3) making it possible for all voters to actually get to vote without extraordinary efforts like standing in line for hours and hours. I know Eric Holder has done some good work in this area, but it needs to be a priority for Democrats as a group.
posted by bearwife at 3:36 PM on November 8, 2012 [6 favorites]


As for the house and gubernatorial races - well, the GOP locked those down back when they had all the campaign machinery and the democratic party was still finding its ass, and then gerrymandered the hell out of their districts in order to ensure house dominance. I believe it will eventually come back around but that one's going to take some time. (The dems also horribly underestimated the effect that the tea party would have in 2010, not just in the races actual tea partiers won, but in the races won by more run-of-the-mill republicans made to look moderate and sane in comparison to the tea party.
posted by Navelgazer at 3:53 PM on November 8, 2012 [1 favorite]


I'm very impressed, again, with the Obama campaign, who are very, very smart and up to date. But I'm underwhelmed with the Democrats.

I suspect they will struggle after Obama just as they struggled after Clinton -- without a strong, charismatic leader in place, the party's internal structures seem to go in unproductive directions.
posted by Forktine at 3:55 PM on November 8, 2012 [1 favorite]


1) the need for a constitutional amendment to repeal Citizens United and allow meaningful campaign finance reform and transparency;

How about never? Does never work for you?

Seriously -- corporations really like that rule. Corporations are big funders of Democrats and Republicans alike. Going against what your backers want is a political death sentence.

This rule will not be changed, at least not by Congress.
posted by Malor at 4:05 PM on November 8, 2012


For a party that can't campaign effectively, the GOP sure does seem to hold a hell of a lot of state legislatures and governorships.

And the media.
posted by DU at 4:05 PM on November 8, 2012 [2 favorites]


It wasn't too long ago that GW Bush was handed the keys to a machine that would keep the Republicans in control for the next 25 years.
posted by klarck at 4:12 PM on November 8, 2012 [1 favorite]


What does it mean to say that The Democrats are not organized?

Will Rogers's remark was more about the awkward coalition the party had cobbled together in the middle of last century. It wasn't really about the massively microtargeted campaign because those didn't exist in his day -- it was about Jim Crow Deep South Democrats coexisting with Eastern labor Democrats and Midwestern farmer Democrats and so forth. Even at its best it could lead to bumptious politics.

There are times it has been amusing to repeat it up to the present day, of course, but the party is much less of an awkward coalition since the ideological separation of the parties post the Nixon-Atwater Southern Strategy.
posted by dhartung at 4:17 PM on November 8, 2012 [3 favorites]


The GOP may figure out that speaking positively about rape is bad for their election chances and just not do that next time. There are many reasons Obama won, but that was I think a big part of it.
posted by joannemerriam at 4:20 PM on November 8, 2012 [1 favorite]


This was a good election for the Dems. With 23 of 33 Democratic senate seats up for grabs they managed to ADD 2 more seats. That alone is a boon considering many were wondering whether the Dems would even cling to a majority.

Romney got less votes than McCain and Obama is the first Democrat to win both the popular vote and the electoral college since FDR.

Twice. Clinton never even won the popular vote.

The insular nature of conservative thinking did them no favors in this cycle as well.

Don't like the poll numbers projecting a win for Obama, unskew them. Thinking that calling women sluts for wanting birth control covered by insurance is a winning strategy? You might be a republican.

Descrying those pesky regulations that, due to republican cries for cuts to government spending, barely keep our food safe and our water clean? That's a winning strategy!

Keep in mind all the bloviating conservative pundits were predicting a LANDSLIDE win for Romney, control of the Senate etc. One big giant circle jerk of out of touch sycophants falling over themselves for a chance to ingratiate themselves to their benefactors.

It was a schooling plain and simple.
posted by Max Power at 4:38 PM on November 8, 2012 [8 favorites]


Here's a nice account of the big ball of fail that was Romney's GOTV system.
posted by octothorpe at 4:42 PM on November 8, 2012 [5 favorites]


Just keep in mind WHY they predicted a landslide. They projected it to their base, while knowing it was a lie, in order to foment dissent. The right's base is angrier now than they would have been if the pundits had told the truth leading up to the election.
posted by yesster at 4:44 PM on November 8, 2012 [4 favorites]


Not to worry, the Democratic Party is proven quite effective at snatching defeat from the jaws of victory.

As evidenced by having achieved some sizeable victories - in the White House, Congress, not to mention a couple important progressive proposals being passed - only days ago, and already there are people moaning about how much the Democrats suck at everything they do.

I mean come on. Maybe we learned something? Maybe we can embrace this and build upon it, how about it? Just throwin' that out there.
posted by Marisa Stole the Precious Thing at 4:48 PM on November 8, 2012 [3 favorites]


Clinton got 5 million more votes than Bush in 1992 and 8 million more than Dole.

If you get that many more votes than 2nd place surely you have 'won'.

Obama did get more than 50%, but that's because there was no third party candidate getting lots of votes. If Bloomberg had run as an independent as was possible at one point the results for Obama and Clinton would probably have been similar.
posted by sien at 4:51 PM on November 8, 2012


Oh...He's just talking about the national level? Where they still hold the House by the balls?

But these articles are talking about getting out the vote. The Democrats got more Americans to vote for their congressional canditates than the Republicans did. The only reason Republicans control the house is because of the ruthless cynicism with which the Republican state parties have gerrymandered the electorates.
posted by yoink at 4:52 PM on November 8, 2012 [2 favorites]


We'll see what the GOP does in the next two and four years. Even they don't really know what their strategy will be among their own. The will either a) double-down and keep railing against birth control and abortion and any number of social issues. Michelle Bachmann, Newt Gingrich, et al will be the leaders. Or b), they make a gentle return to what conservatism used to be; the Richard Nixon who championed the EPA and Ronald Reagan who raised taxes multiple times. There are a number of conservatives who aren't Bachmann-esque crazy that could win lots of votes without the Tea Party/Libertarian/Fundie crowds. Someone like Jeb Bush, who I could see running in 2016.

If they double down and refuse to change their messages, the GOP is toast. Their messages only play to old, white people (namely men), and those numbers are shrinking. Maybe not rapidly but consistently. In two years there will be less of those old white men and more minorities who overwhelmingly chose Democratic. Which is always true, but even more this time around, and that number is consistently growing.

BUT: The Citizens United billions poured in the GOP this year was mostly wasted. Those billionaires are pissed right now, and they will be much more circumspect in two and four years. That money will be spent more smartly next time.
posted by zardoz at 5:11 PM on November 8, 2012 [2 favorites]


Yeah, next time around they will parachute the chapped-lipped upper middle class white women straight to your house to wonder aloud how they can explain to their six year olds about jorbs.
posted by robocop is bleeding at 5:42 PM on November 8, 2012 [1 favorite]


BUT: The Citizens United billions poured in the GOP this year was mostly wasted. Those billionaires are pissed right now, and they will be much more circumspect in two and four years. That money will be spent more smartly next time.

Not necessarily.

"A lot of people feel like Christie hurt, that we definitely lost four or five points between the storm and Chris Christie giving Obama a chance to be bigger than life,” said one of Romney’s biggest fundraisers, who requested anonymity to speak candidly."
posted by clockworkjoe at 5:46 PM on November 8, 2012


I really found the main article interesting - thanks for posting this, flex.

I haven't read all the other links yet, but I'm looking forward to them.

From the articles, it certainly sounds like the Democrats are doing some interesting and apparently effective work with behavioral science and testing outcomes, and I think that's great.

If it's been honed at the top, for the national election, but can then be increasingly adopted by state and then local races, that

I'm a fan of avoiding wasted effort in any area of endeavor. If these tactics can reduce wasting money and wasting resources, and instead increase effectiveness in getting people to actually vote and participate in the hard, frustrating work of self-government, I'm all for it - and I'm very heartened to learn that people are at least trying to put these techniques into practice.
posted by kristi at 5:59 PM on November 8, 2012 [1 favorite]


The GOP has closed its decision cycle to outside input. That's why they were so surprised by their loss.

That's not really true; Daron Shaw and some other conservative/Republican political scientists worked with a Perry campaign to conduct controlled experiments on the effects of ads (among other stuff I've forgotten). They found that ads work, but not for long.
posted by ROU_Xenophobe at 6:07 PM on November 8, 2012


The GOP doesn't have much to offer anyone besides their racist, homophobic, sexist, religious ideological crap. They've done a great job in this campaign in disenfranchising LOTS of voters. Rape, immigration...same sex marriage etc. Those hard lines, sadly, considering that Romney continuously came across as a poorly programmed robot, are what caused the failure.

They will likely learn from this and swerve to the left a little bit. A little bit will be all it takes.

That said, Obama can seal the deal for 2014/16 by befriending the red states.

1) He needs to invite Honey Boo Boo and family to the White House for an impromptu beauty pageant with the first daughters, have some beer and play horseshoes.

2) Hunt, capture and eat a 'gator.

3) Legalize home distilleries (moonshine) and make his own batch with his mash recipe posted online.

Obama has plenty of potential for securing DEM dominance other than my jokey stuff. Will he use it? Will he use his pulpit to reach out to red state voters/citizens? In his shoes, I would. That gap really needs to be bridged. He can do it. He has the heart and love of America for it.

Good luck Obama!
posted by snsranch at 6:19 PM on November 8, 2012 [2 favorites]


The reality Democrats need to accept is that they are not super geniuses at targeting their base and narrowcasting messages to specific voter blocks to sway independents. Karl Rove thought be was a genius in 2004 with the same bs. The hard truth is that it is just really difficult to unseat a sitting President. Most of the smart politicians wait for the end of the second term for this reason. The party out of power usually ends up running some b list guy such as John Kerry or Mitt Romney or the governor of a small southern State. Sometimes te perfect storm of events conspire to push the incumbant out, but it still takes an gifted nominee from the other party to get the win. Reagan and Clinton had incredible political talents. They ran against men of moderate political ability who faced significant economic and political headwinds. Both of them barely defeated the incumbent President.
posted by humanfont at 6:25 PM on November 8, 2012


Can this genius cluster please roll down to the South and start organizing again? Can we have substantial, well-funded and effective Democratic Party organizations in Louisiana, Mississippi, Alabama, Tennessee, etc. again? Could these folks do some studies down here and figure this out? RUN SOME ACTUAL CANDIDATES FOR CONGRESS? Holy crap. There are dozens of studies to use to help them figure this out, centers where southern culture and folklore are studied, etc., professors who specialized in southern politics, etc. Clinton at least won Arkansas and Tenn. both times, and Georgia in '92. Maybe you could talk to him. He likes to talk and explain.

(Whispered: I have social science degrees, btw, so you could always hire me for your team.)
posted by raysmj at 6:35 PM on November 8, 2012 [9 favorites]


Much of the Obama OFA infrastructure operates parallel to the official Democratic party apparatus. A major criticism of Obama immediately after the 2010 Tea Party sweep was that he failed to make us of his legions of personal supporters to mitigate Democratic losses in midterm races. Even this cycle, the Obama endorsements were few and far between. Once Obama is out of office, what becomes of his personal campaign network and how much the Obama People become Democratic Party people will determine whether the Dems maintain this level of organization and effectiveness.
posted by T.D. Strange at 6:41 PM on November 8, 2012 [5 favorites]


I hope they keep whining about Sandy throwing the election, because it completely misses what happened. The polls were already moving in Obama's favor, and Romney was already increasingly likely to lose.

Time was running out in the fourth quarter with Obama leading by a touchdown, then Romney threw an interception that Obama ran back for a TD. Romney was going to lose anyway; Sandy was just icing on the cake.
posted by kirkaracha at 6:54 PM on November 8, 2012


Definitely agree, raysmj. How the hell are we letting people run for congress UNOPPOSED?
posted by goHermGO at 7:23 PM on November 8, 2012 [4 favorites]


Thanks for posting these articles. I spent a lot of time on local campaigns this cycle, and found myself asking a lot of questions about how effective the canvassing and phone calling really is. One of the campaigns here (which defeated the best-funded candidate in local history) had good answers, and numbers, and clearly understood the science of it. I was impressed with their rigor and not that surprised that the beat the candidate with allllll the money.

I find the slate mailers and the phone calls and the door knocking really obnoxious when I'm on the receiving end, but I'm also someone who knows who I'm voting for well before election season and always vote. It's good to know that they go to some useful ends. The article about the signs was especially interesting.
posted by gingerbeer at 7:38 PM on November 8, 2012


I imagine that as Obama people filter out to the rest of the Democratic Party organizations, it should vastly improve senatorial and congressional campaigns. We'll see how they do in 2014. The economy should be going gangbusters by then.

It's unfortunate that there is really no designated successor that could just pick up where Obama left off. Biden and Clinton aren't it.
posted by empath at 8:49 PM on November 8, 2012


How the hell are we letting people run for congress UNOPPOSED

Lets say that Nate silver told you there's a 99.999999% chance you were going to lose a campaign. How much time and money is it worth to spend on recruiting and supporting a candidate? When you can use those people in other races, and spend that money on races you can actually win?
posted by empath at 8:52 PM on November 8, 2012


Lets say that Nate silver told you there's a 99.999999% chance you were going to lose a campaign. How much time and money is it worth to spend on recruiting and supporting a candidate? When you can use those people in other races, and spend that money on races you can actually win?

Sometimes, the shoo-in candidate self destructs (like Akin and Mourdock), and it's worth it to have a candidate ready to take the seat. Also, it's worth building up a national fundraising base.

I'm not sure exactly why the Democrats abandoned the 50-state strategy (and Dean, for that matter), since it was clearly pretty successful.
posted by robla at 10:05 PM on November 8, 2012 [5 favorites]


So basically the Dems won because seriously smart people migrated towards its campaigns, as opposed to the Repubs, who still had these institutional-thinking types directing things from the top.
posted by the cydonian at 10:40 PM on November 8, 2012


Lets say that Nate silver told you there's a 99.999999% chance you were going to lose a campaign. How much time and money is it worth to spend on recruiting and supporting a candidate?

It's called building name recognition, at least. an investment in the future. You build from the ground up. I'm not saying put the most money in it, just start a longish term plan. There are many reasons the Dems. lost the House in 2010, and lost Senate seats on top of that, and one was a complete lack of organization in the South. Build an organization in these states, meanwhile, and you won't (due to pressure) have governors who decide, Oh, we won't accept funds for Obamacare, we'll sue the federal govt. over it, etc.
posted by raysmj at 10:49 PM on November 8, 2012 [4 favorites]


As evidenced by having achieved some sizeable victories - in the White House, Congress, not to mention a couple important progressive proposals being passed - only days ago, and already there are people moaning about how much the Democrats suck at everything they do.

I mean come on. Maybe we learned something? Maybe we can embrace this and build upon it, how about it? Just throwin' that out there.


Yeah. I mean, jeez guys. Exactly what does carping about your own party accomplish? Nothing but discouragement. It undermines its chances.

Seriously, if you're in favour of the Democrats, there are two useful things to do: come up with better ideas for them, or support them. This endless grousing just makes liberalism look like a party nobody sensible wants to join.
posted by Kit W at 10:54 PM on November 8, 2012


I haven't read all these posts, but I really love everything I have read by this guy. As someone who's worked on my share of campaigns, he's really right about how off-base many political journalists are about how campaigns and elections actually work. Though I do think it's getting better - political journalists seem to have figured out that campaigns actually hire staffers other than press secretaries and fundraisers.

You know how annoyed scientists get by most science journalism? Yeah, that's how most campaign staffers feel about political journalism.
posted by lunasol at 11:04 PM on November 8, 2012 [6 favorites]


Lets say that Nate silver told you there's a 99.999999% chance you were going to lose a campaign.
Is this about baseball?
posted by fullerine at 6:04 AM on November 9, 2012


The Democrats only added 3 seats in the house, and the Republicans only lost 3.

After Republicans controlled redistricting; it's actually pretty impressive. I think on the old map, the Dems might have been able to swing 20 seats.
posted by spaltavian at 6:25 AM on November 9, 2012


It wasn't too long ago that GW Bush was handed the keys to a machine that would keep the Republicans in control for the next 25 years.

That machine was real; he just drove it into a ditch. Imagine if Bush had been able to force his party to accept immigration reform and didn't invade Iraq. The crash of 2008 might have still put a Democrat into the White House but it wouldn't have been Obama, and it's doubtful she (because, we know who it would have been) would have been reelected.
posted by spaltavian at 6:34 AM on November 9, 2012


What are "the public opinion firms of the left" referenced in the first link?

lunasol? anyone?
posted by univac at 6:41 AM on November 9, 2012


Two things. One, I don't think people have very good intuition about how leaders can drive electoral results, a 2% swing in the electorate could have massive consequences for the results but would basically be invisible to the "naked eye." Two, given that the election results basically mirrored the polls, I'm skeptical that special GOTV efforts had a massive impact on the outcome.

And it is true that Rove was widely heralded as a super-genius after the Bush elections, but in 2012 he looks like a grifter conning rich people.
posted by leopard at 6:50 AM on November 9, 2012


The gotv efforts were mirrored in the polling.
posted by empath at 6:52 AM on November 9, 2012


For a party that can't campaign effectively, the GOP sure does seem to hold a hell of a lot of state legislatures and governorships.

The one consistent truth I find is that the Republican electorate is stunningly consistent, while the Democratic one fluctuates wildly.

Most Republicans who voted this year will vote next year. The same cannot be said of Democratic voters, many of whom are not even aware that there IS an election next year. Republicans are, on the whole, diligent voters while Democrats, on the whole, are political dilettantes.

I love it when people complain about how elections are run and then don't show up to vote for county-level elections - the one time they have a chance to do something about it.
posted by snottydick at 8:49 AM on November 9, 2012 [3 favorites]


I've been saying the same thing for years now, snottydick, both as a way to improve state and local elections and as a way move the party as a whole to the left. Sadly, the responses both here and among my acquaintances has ranged from mild approval to outright denial of it's efficacy.
posted by zombieflanders at 9:01 AM on November 9, 2012


outright denial of it's efficacy.

This is why I have a hard time taking it seriously when Democrats complain about who's counting the votes, who's making decisions about what voting machines to buy, and who got to redraw legislative districts last year.
posted by snottydick at 9:11 AM on November 9, 2012


I won't believe the media "gets it" about how elections work until the debates feature, instead of panels of undecided voters turning a knob to express how much they like either candidate, panels of supporters turning a knob to express how much what their candidate says makes them want to get off their ass and vote.
posted by straight at 10:08 AM on November 9, 2012 [1 favorite]




The Republican GOTV software, code-named "Orca", was a complete mess.

It might have been lousy enough to throw the election to Obama.

(sorry for the link to Breitbart, the whole article was a bit too much for me to inline)
posted by Sauce Trough at 11:57 AM on November 9, 2012


I went down the rabbit-hole of conservative website internet comments, following that ORCA story, and I've come out a different woman.

There were many areas of similarity in partisan comments - allegations that voters are stupid; or complaints about voter fraud (although conservatives seem to favor a relatively simple-to-detect 'just add 3 votes for Obama for every 1 for Romney' conspiracy). But I don't remember a lot of liberals misquoting that Drowning Pools song as a call to violent action in the wake of the 2004 election. It's frankly un-American (and also very illogical, considering we've already survived 4 years of Obama presidency and will likely survive another four, considering we handily survived 12 years with that Marxist FDR).
posted by muddgirl at 2:29 PM on November 9, 2012


ORCA

Way to go Microsoft!

In all honestly, 11 database servers and 1 frontend server seems entirely backwards to me. I don't think that detail is correct.

I also don't believe they didn't check for bugs at all. I hear shit like that all the time too "I found a bug! Didn't you idiots test this! I know how to do your job better than you do!". Of course we tested it, thats why you only found 1 bug.

In short, it may have sucked, but blaming your loss on software is bullshit. The fact that according to Ace Of Spades, where this story broke, they didn't tell volunteers to get poll watcher certificates is far far worst.
posted by Ad hominem at 4:56 PM on November 9, 2012 [5 favorites]




The Obama campaign knew they'd win Ohio 45 minutes before the television-networks called it. Also, there was this:
Meanwhile, the Romney campaign was openly dismissive of the Obama ground game. Why are they wasting so much money with neighborhood offices, they asked? (In Ohio, for example, Obama had almost 100 more offices than Romney.) In retrospect, the Romney team is in awe and full of praise of the Obama operation. “They spent four years working block by block, person by person to build their coalition,” says a top aide.
posted by the cydonian at 9:42 PM on November 9, 2012 [1 favorite]


In short, it may have sucked, but blaming your loss on software is bullshit.

In data-driven organizations, I often find that the way you execute your largescale IT projects reflects on your overall work-culture quite a bit. At the very least, they didn't value their data-sending operations as much as they hyped them up.
posted by the cydonian at 9:53 PM on November 9, 2012 [2 favorites]


As far as I understand it, the analysis by the likes of Drew Linzer and Nate Silver suggested that the outcome of the election could have been quite well predicted from Obama's approval rating in June and the fact that there was economic growth in the period since. (As per the link, Linzer actually did predict it in June, and his prediction never changed much over time.)

So it's possible that the story about superior analytics and campaign strategy is just another plausible narrative that people put on events, much like the it-was-down-to-Sandy story, but it doesn't necessarily explain what were truly the most important factors driving the result.

In a way that would be quite heartening because, if true, it boils down to that the most important thing was that enough people on balance thought Obama was doing an ok job.

And the message for politicians would then be: "First get done what the voters most care to get done. The rest of it, what happens during campaigns etc, is not going to move the needle much compared to that."

Come to think of it, all this praising of the Democrats' prowess sounds a lot like praise of New Labour's supposed mastery back when Tony Blair was racking up election wins. And we all know how that went.
posted by philipy at 10:49 PM on November 9, 2012


Well, as I noted in the motherthread, you could also have predicted an Obama victory as long as two years ago using the performance-oriented Keys to the White House system of Allan Lichtman (and he did); the system has now correctly predicted the popular vote winner in every election since 1984. Similarly, this system de-emphasizes almost everything that happens or is done in a campaign.

What I saw from OFA here was close integration with the local Democratic campaigns. The staff they sent were tireless, smart, responsive, and dependable, a backbone supporting local volunteers. We in Wisconsin had built up a significantly well-oiled machine during the year of recall elections and that probably helped. But I can't escape my nagging suspicion that GOTV is important, but provides only a smidgen of a boost. Even with my close connection to the office I got a metric butt-ton of GOTV calls from both sides -- all of it essentially wasted effort. They can't know that I'm in the can, after all. I think of all the diehard partisans and the disaffected pox-on-all-pols people who got all these calls and have to wonder.

Demographics are important. Organization is important. But it still depends on having a candidate people will turn out for. Obama is clearly that candidate; so was Clinton. But what if we get a Kerry next time (who I actually think wasn't that bad, but faced severe prevailing winds), or God forbid a Dukakis or Mondale (again, some of their defeats is down to circumstance, but indubitably neither was gifted with the charisma gene). What if there's a Black Swan event like 9/11 or the financial crisis? What if a miracle occurs and a brilliant GOP populist emerges who can connect with all types of voters the way Clinton can and manage the expectations of both moderates and conservatives? It's really easy to predict victory going forward based on the stuff you have going for you in this election, but that's what happened to the Republicans after 2004. I really don't want to get overconfident. I definitely believe that being on the outs is a terrific incentive for the out party to find someone who fulfills their need and focuses their message where it needs to be.

I honestly don't know what to say about Wisconsin, having given Obama and openly gay Tammy Baldwin sweet victories just two years after Scott Walker's wrecking crew was elevated along with Ron Johnson unseating Russ Feingold. Red, blue, purple, or maybe none of that is the explanation at all. Prognostication is a risky game, that's for sure.
posted by dhartung at 1:28 AM on November 10, 2012 [1 favorite]


Mondale had to face Reagan, a Holywood actor who oozed charisma. I've heard Mondale speak a number of times and have been in rooms and elevators wth him. He does have charisma.

As for Wisconsin, the futility of the recalls and protests should remind everyone of the importance of actually voting during the regularly scheduled elections.
posted by Area Man at 4:48 AM on November 10, 2012 [2 favorites]


as long as two years ago

I guess the likely problem with predictions that far out is that a lot can happen, in terms of the economy getting worse, a war going horribly wrong and so on.

I'll check that model out, that would be very interesting if it holds up rather than maybe being a case of "overfitting". (In stats speak, making a complex model that is great at "predicting the past", i.e. matching the patterns in the data it was derived from, but not holding up so well when predicting the future.)

having a candidate people will turn out for

Charisma isn't a part of the models either though, so our hunches about this stuff may prove to be off base.

The models basically seem to say "It's the economy, stupid". But not necessarily the economic variables that everyone assumed. According to the models it looks like "growing" counts for more than "bad".

If I was going try my hand at modeling I might look at what you can predict if you took a state-by-state look at: "How does the economy feel now compared to when the President came in?" The intuition behind that being something like: "If you manage to rescue the auto industry, you win Ohio and Michigan. Charisma or no charisma, superb analytics or mediocre analytics."

Btw those of you who know more about politics than tech might like to read about the Tech Hype Cycle. Analytics and big data is taking its turn around that maybe.
posted by philipy at 7:47 AM on November 10, 2012 [1 favorite]


I've heard Mondale speak a number of times and have been in rooms and elevators wth him. He does have charisma.

To be fair charisma on a stage and on TV isn't always correlated with charisma in a room or an elevator. You hear that about a fair number of politicians: "Oh, but in private he's so warm and funny. If people only knew him as I know him..."

From what I hear Obama is the reverse, he comes alive on big stages and with set piece speeches, he's not so great at talking with people in a diner or on the golf course.

Bill Clinton was great at both.

Still it looks like charisma might have more impact on how people feel about leaders rather than whether they vote for them. e.g. From my limited knowledge of French politics, Sarkozy is rather charismatic, Hollande far from that. But probably it doesn't count for much if the economy sucks and people think you're to blame.
posted by philipy at 8:15 AM on November 10, 2012


It's really easy to predict victory going forward based on the stuff you have going for you in this election, but that's what happened to the Republicans after 2004. I really don't want to get overconfident.

Yeah. Now that I've sobered up from the delirium of winning as big as the Dems did this week, I'm noticing that there are a lot of parallels to 2004, just with the parties switched. An odd gaffe inflated by the media took out the more ideologically pure candidate in the primary, leaving the way open for a milquetoast flip-flopping robotic "electable" candidate from Massachusetts; a seemingly overwhelming and unbeatable majority for the opposition with no end in sight. Even the infuriated reactions of conservatives pretty closely match how I felt in 2004 with feeling completely unmoored in society and believing there's a huge media bias towards the incumbent, and now we've even got a whackadoo conspiracy about when the President knew about a terrorist attack. I think the only thing that's really different is that in 2004, people were still scared from the 9/11 attacks and voted out of that fear. This year, people seemed to vote for Obama despite their fears about the economy, so there may be potential for even bigger wins down the road if the demographics hold up.
posted by LionIndex at 8:44 AM on November 10, 2012


Yeah, I poked my head in on that, it just seemed a bit awkward so I dropped it, despite it's proto-Vertigo roots.
posted by Artw at 9:44 AM on November 10, 2012


I'll check that model out, that would be very interesting if it holds up rather than maybe being a case of "overfitting". (In stats speak, making a complex model that is great at "predicting the past", i.e. matching the patterns in the data it was derived from, but not holding up so well when predicting the future.)

The model is:

There are 13 'keys'. A 'true' answer is favorable to the incumbent. If the incumbent has 5 or less false answers, they win. If the incumbent has 6 or more false answers, they lose. Extremely simple, explicitly accounts for charisma, probably not overfitting too hard (Though there aren't that many data points). Also, note this model isn't just, 'It is the economy!'.

The 13 Keys to the White House: Standings, October 2012:

Key 1 (Party Mandate): After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than it did after the previous midterm elections. (False)

Key 2 (Contest): There is no serious contest for the incumbent party nomination. (True)

Key 3 (Incumbency): The incumbent-party candidate is the sitting president. (True)

Key 4 (Third Party): There is no significant third-party or independent campaign. (True)

Key 5 (Short-Term Economy): The economy is not in recession during the election campaign. (True)

Key 6 (Long-Term Economy): Real per-capita economic growth during the term equals or exceeds mean growth during the previous two terms. (False)

Key 7 (Policy Change): The incumbent administration effects major changes in national policy. (True)

Key 8 (Social Unrest): There is no sustained social unrest during the term. (True)

Key 9 (Scandal): The incumbent administration is untainted by major scandal. (True)

Key 10 (Foreign/Military Failure): The incumbent administration suffers no major failure in foreign or military affairs. (True)

Key 11 (Foreign/Military Success): The incumbent administration achieves a major success in foreign or military affairs. (True)

Key 12 (Incumbent Charisma): The incumbent-party candidate is charismatic or a national hero. (False)

Key 13 (Challenger Charisma): The challenging-party candidate is not charismatic or a national hero. (True)
posted by enkiwa at 10:24 AM on November 10, 2012 [1 favorite]


Mondale had to face Reagan, a Holywood actor who oozed charisma.

So what you're saying is ... there may be elections in which we face a charisma deficit. That's my point. Obama has energized Democrats, and so did Bill Clinton, but both Gore and Kerry -- able campaigners who both increased the numeric vote for Democrats, as I recall -- fell short in the perceived charisma department. (I've seen all four live.) I'm just saying we can't count on having great campaigners every four years.

maybe being a case of "overfitting". (In stats speak, making a complex model that is great at "predicting the past", i.e. matching the patterns in the data it was derived from, but not holding up so well when predicting the future.)

The model was devised in 1981, with one minor tweak after 1984, and has correctly predicted -- as I noted -- eight Presidential election popular vote winners. I don't know how many more elections it will take before a numbers-obsessed guy like Nate Silver will at least grudgingly respect its crude form, but it seems to have had some success, and tellingly seems to suggest that the numbers aren't the only thing worth looking at. If the model has any other utility, at least that.

(I think this is the only thread fully devoted to the system: The Thirteen Keys to the Presidency.)

I guess the likely problem with predictions that far out is that a lot can happen, in terms of the economy getting worse, a war going horribly wrong and so on.

The model actually accounts for that -- there is a foreign policy failure key and both a long-term economy and short-term economy key. The real problem with the keys isn't really a statistical overfitting as much as a question of judgement in whether a key has in Lichtman's somewhat inept jargon "fallen". What constitutes a foreign policy failure, or success? What is a scandal, or a policy change, that tips a key? How do you measure whether a candidate is charismatic? It's a dream model for political scientists who hate (or, say, distrust) numbers. My main problem with the numbers models is that they measure too few things -- the GDP, or the polls. But I've been fascinated with how well this model -- which I've personally "used" to make my own predictions on my own judgement since 1992 -- has continued to be accurate and how far in advance that accuracy carries. Essentially, and this is the underlying point I"m making rather than getting bogged down in the weeds about the system's characteristics, the model measures presidential performance and says that that matters, and a good deal more than the horse race crap the media throws on the street year after year. I think the polls say something about the electorate in aggregate, but they tell us little about our self-governance or, in a sense, ourselves. I ask what, indeed, all of a year of reading Nate Silver has taught us about how to decide who should run the country. I know he is a brilliant statistician and has a very effective communication style for somebody who lives inside spreadsheets, but I'm his audience -- I could eat that up -- and eventually my eyes glaze over from the whole approach. Sure, the electorate thinks this today, but what does that tell us if your whole model is ready to concede that they may not think that tomorrow? It's a slave to fluctuating opinion, whereas the Lichtman model is more about things in the real world and the concrete political process. The numbers, in a sense, may be too precise to give us any useful information.
posted by dhartung at 12:21 PM on November 10, 2012 [2 favorites]


dhartung, does Nate Silver ever do any regression modelling of the poll data, with explanatory variables? I know Australia's own version, Possum Comitatus tends to do that at times. Not just boring models against GDP, but more practical, politically interesting things that might explain why the polls do what they do, like does the majority party get a bounce in the polls when parliament is sitting?, or measuring whether the appearance of a third-party has a short-term or long-term impact.
posted by Jimbob at 2:15 PM on November 10, 2012


I got as far as looking up the model and reading the Nate Silver comments on it which mentioned overfitting and subjectivity.

I'd need to look into a lot more thoroughly to say anything really worthwhile though.

But some general comments on how something like that is likely to stack up against say using just approval rating and GDP growth...

1) In this model you "subjectively" decide what is a major achievement, scandal, foreign policy triumph or setback etc. If I was deciding those things, my impressions would most likely be based on the narrative portrayed in the media, albeit the more sophisticated end of the media. Those same things would likely be feeding into the approval figures also, and most of the time my "subjective" call and the approval ratings will correlate. But those times when they don't, it's likely to be because the public as a whole rates the event differently than I would. e.g. I'd probably think healthcare reform is a major achievement, but apparently the public is a lot less impressed, maybe even overall hostile. So in terms of predictive power the approval rating is likely to be better.

2) If you were interested in knowing why the public approve or don't approve of the President, and whether they're likely to change their mind, the Linzer type model doesn't help much. Using my personal ratings of the administration's performance in the different areas with a framework like Lichtman's might well help with that. But it would be even better if you could use the public's ratings of performance. I don't know if anyone's tried that. It might work, but it might be subject to the overfitting problem. These days you could maybe do a textual analysis of tweets to find out if Obama is considered charismatic or Romney dull.

3) The "slave to fluctuating opinion" problem isn't really improved by this type of model, because you're instead a "slave to fluctuating events" instead. Which events are probably the explanation of the fluctuating opinion anyway. But another lesson from the stats seems to be that the polls don't fluctuate as much as we think they do. Taken in isolation polls have a quite wide range of possible error, and different polls might have tendencies to err in different directions as well. Maybe poll X is within +/- 2%, and tends to overestimate Democrat votes, poll Y is within +/- 3% and tends to overestimate Republican votes. But when you carefully aggregate all of them, you get a combined measure that is reliably within +/- 0.5%, and doesn't fluctuate nearly as much as you'd think from plotting the spread of all the individual polls over time.

4) While things like economic growth sound "objective", they're actually still groping around for indicators that might tell you whether people feel good about their lot in life or not. For example it can be that while there's growth as economists measure it, that can sometimes mean the rich are doing well, but most people aren't any better off at all, and I would guess will then vote accordingly. So it's conceivable that your "subjective" read of events is a better reflection of how people are feeling than the economic stats are. Because you are maybe a decent barometer of the national mood, and there might be more connection between one person's mood and others' than between economic indicators and moods.

These models including Lichtman's are certainly very interesting, andI'd love to look deeply into them if I had the time. The less obvious "keys" are intriguing ones to consider.

Personally there were only a few weeks here and there in the last four years when I seriously questioned if Obama was going to be re-elected. But it was for the kind of reasons I mentioned here rather than any models or frameworks. (In a nutshell that the Republicans were going through too extreme a phase to win a national election and phases like that don't change quickly.)
posted by philipy at 2:18 PM on November 10, 2012 [1 favorite]


Australia's own version

If anyone knows interesting models for other countries, I'd love to hear about them.
posted by philipy at 2:20 PM on November 10, 2012


I think it's plausible that GOTV efforts made a difference in this election.

Both candidates were pretty weak. The economy isn't doing very well, and a lot of Obama's base condemned him for not being liberal enough. The Republican party is divided by factional infighting, and a lot of Republicans considered him a Republican in name only.

The Romney campaign was counting on low turn-out by Democrats and concentrated on going after independents. It also seems to have been poorly managed. The Obama campaign was well organized and focused on getting out their base. The enthusiasm gap failed to materialize.

It's not the only factor though. Probably the most striking difference in the exit polls from this election compared to 2008 is how strongly Hispanic voters turned against the Republican ticket. Really pissing off a major ethnic group that's not politically monolithic and wasn't previously biased against you is not really a smart thing to do. That probably cost him Nevada, Colorado and Florida.

While Republican pundits, bloggers and talk show hosts will be pointing fingers and wailing about how the Democrats stole the election in various ways, Republican consultants will be studying what the Romney campaign did wrong and what the Obama did right and learning from it. This doesn't translate into a permanent advantage for the Democrats. It might translate into an advantage in 2014, though, if the Republicans haven't gotten up to speed yet, but they definitely will have by 2016.

A lot of this depends on how well organized campaigns individual campaigns and candidates are. There's no guarantee that 2016 won't pit a Republican campaign that's learned a lot from this election and 2008 against a poorly organized Democrat.
posted by nangar at 2:26 PM on November 10, 2012




That "analysis" seems to use "return on investment" to mean how much money went to candidates who won out of total money spent. That is a very strange definition, e.g. I could get 100% "return" by spending all my money supporting Obama in California. On a saner definition that would more likely be considered a complete waste of my resources.

What you'd care about, but which is hard to determine, is to what extent was a result influenced by the spending. If the criterion is getting favorable candidates elected, money spent on candidates who lost is wasted. But money spent on candidates who would have won anyway without your money is also wasted. The money is only clearly not wasted in the pretty hard to prove edge case where the candidate was otherwise just short of winning, and your money is what pushed them over the line.
posted by philipy at 3:01 PM on November 10, 2012 [1 favorite]


I have to agree with philipy here. I mean, I sent five bucks to the fund to kick Todd Akin to the curb, so that means I got 100% ROI. It's an analysis that favors smaller players with narrower targets. Comparing Planned Parenthood to the Koch brothers is kind of silly.
posted by localroger at 3:22 PM on November 10, 2012


3) making it possible for all voters to actually get to vote without extraordinary efforts like standing in line for hours and hours. I know Eric Holder has done some good work in this area, but it needs to be a priority for Democrats as a group.

A Perfect New Mission for Eric Holder: The attorney general has the chance to build a lasting legacy as a defender of voting rights. How? By leaving his current post to take up a new one.
posted by homunculus at 6:18 PM on November 10, 2012 [1 favorite]




The 13 keys does seem to be a structured way in which Lichtman himself thinks about the broader political conditions and makes accurate predictions, but I wonder whether it would be so accurate for anyone other than Lichtman.
posted by Area Man at 6:10 AM on November 11, 2012


It's fascinating seeing how the media on this (England) side of the Atlantic cover the election. We don't have the quality analysis* that you have in the UK, so the reporting tends to be long, dreamy sequences (BBC newsnight), Leftie joy (Guardian) or Right wing "fall of America" (Telegraph).

One of the things that may be fascinating over the new two decades is, even with all these economic and social changes, whether US political dynastic families will produce more US presidents. The news reports here are already fixated on Hillary Clinton and Jeb Bush (possibly as, in fairness, they are the more likely for the next presidential election). But they are missing a few (at the end of self-typed ramble).

* - yeah, I know a lot of Americans think their media is "terrible" but it seems IMHO to be "very variable", with the sometimes remarkable quality of journalism and writing in some broadsheets something we simply don't have in the UK. I would love to have a regular New York Times, or Washington Post, or Toledo Blade, for where I live.

Des Moines Register; maybe not so much.
posted by Wordshore at 6:30 AM on November 11, 2012


Uh, missed the edit window. Should read "...that you have in the US, but..."
posted by Wordshore at 7:02 AM on November 11, 2012


Mark Liberman has collected multiple links about ORCA. (The article also mentions Issenberg's article about Narwhal.)
posted by nangar at 11:57 AM on November 11, 2012 [1 favorite]


Another thought occurred to me about models and prediction accuracy.

Namely, that if we're talking about predicting who wins, rather than by exactly how much, I would have called almost every election right just based on my gut feel, even when I wasn't much more than an interested child.

I'd have called them all from Carter onwards, apart from expecting George Bush Sr to be re-elected against the then unheard-of Bill Clinton, and I expected Gore to win in 2000.

There's a few things that fact might tell us.

Firstly, calling them all right with the help of a complex framework is not as impressive as it sounds, because the baseline for calling them right from nothing more than gut feel is pretty high.

Second, there are probably people whose gut feel is going to be a better barometer than most, not because they're astute or knowledgeable but because their own personal feelings happen to mirror the feelings of swing voters well. The kind of things that make me cringe are the kind of things that would make a lot of independents cringe also. I of course get turned off in a big way if someone seems to think that 47% of the population are irresponsible scroungers. But I'd also be uncomfortable if partisans on one side seem to write off 47% of the population as dumb good-for-nothing racists. Note that for the purpose of this discussion it doesn't matter if they even are or not, the point being my instinctive reaction, a poll of one person effectively, is nonetheless a pretty good measure of how swing voters as a whole will react, and therefore quite good at predicting how the election will go.

But where I don't call it right is quite an interesting thing to look at as well. For example, global warming is pretty much the #1 issue for the planet as far as I'm concerned. That is one place where my feelings and the feelings of swing voters seem to diverge. So in 2000 I find it hard to imagine that it can even be close between Bush and Gore.

A corollary of that observation would be: The more partisan you are, and the further from the median voter you are, the less accurate you will be in calling elections from gut feel, and the less able to understand why they went the way they did.

Now that leaves the Bush Sr - Clinton election to consider. That is one place where the 13 keys model might well have given me food for thought. For one thing, even now I'd be hard pressed to say what effect Ross Perot had on the result without looking deeply into it. At the time it never occurred to me that a third party candidate intervention might be enough to tip the balance. For another, it might have made me weigh up my assessment of George Bush Sr as a man against other factors. I rated him quite highly, he seemed a decent, capable, thoughtful, pragmatic guy. He was the one who once called the idea of slashing taxes to boost growth "voodoo economics". He'd done a fine job of building a coalition in the first Gulf War, and got that job done in short order, going exactly as far as the international community had agreed and no further. I would have thought he was a cinch for re-election.

What wouldn't have occurred to me is the weight people give to the economy, and the extent to which they credit or blame the president for it. To me, but maybe not to those average swing voters, there are always economic cycles with booms and busts, those problems are long in the making, long in the solving, and the guy in the White House when the economy tips is not necessarily the root cause.

I don't know if the 13 keys model would have helped me call this one correctly, but it would at least have suggested that maybe re-election for Bush Sr wasn't a sure thing like I imagined. In its favor, it would have got me look at the fact there was a third party player and that the economic numbers were bad, without me adding my ideas about root causes to the mix. Against it, the economic keys are only a couple out of twelve, and are given equal weight with things like foreign policy success, which looked very good for Bush Sr. A statistical model would probably have called this one better than I would have using the framework, because it would have given a lot of weight to the growth numbers compared to anything else.

A final thought from the world of chess. A Grandmaster with the assistance of a chess program is much stronger than either a Grandmaster alone, or a program alone. While it's not that close of a parallel to politics, something similar may hold. Programs don't make the kind of mistakes humans make, but they have other limitations, and there are times when a human that has a deep understanding of the game and of the program's strengths and limitations can use their insight and judgment to overrule them and do better. But if you're not a strong player and you're also clueless enough to think that when a program makes a bizarre-looking move it must be wrong, it'll kick your butt in no time flat.
posted by philipy at 10:11 AM on November 12, 2012


The All-Seeing Campaign - "Some astonishing details are trickling out about the Obama campaign’s sophisticated infrastructure. They were polling, watching, targeting, microtargeting, “microlistening,” and in general doing everything possible to stay on message, turn out voters, and economize."
posted by the man of twists and turns at 8:44 AM on November 16, 2012


When the Nerds Go Marching In: How a dream team of engineers from Facebook, Twitter, and Google built the software that drove Barack Obama's reelection

Yay, nerds!
posted by tonycpsu at 9:59 AM on November 16, 2012 [1 favorite]


« Older It is a time of freedom and of celebration when...   |   Beer Suits! Newer »


This thread has been archived and is closed to new comments