The Victory Lab: The Secret Science of Winning Campaigns

Very detailed history of the modern data-focused campaign, using microtargeting to make sure each potential voter is “touched” in just the right way, and yes it is in some sense that creepy. The detailed portrait of entrenched resistance to better means of finding relevant voters is useful in showing yet again that even a committed partisan can have trouble accepting new facts if those facts will threaten his (essentially always his) income stream, since it’s the consultants here who array themselves against innovation (in the form of other consultants). Issenberg tracks the Republican data/mobilization advantage of 2004 and its dissipation in 2008; much of this story has been told in the popular press, but Issenberg at least show more contextualizes it. One takeaway is that you can’t work miracles with data alone—though consultants are looking for the best way to phrase appeals for funds, they are also working off a support score (how likely you are to support a particular candidate) that is responsive to changes in the salience of public issues. People will still vote differently in a recession than in good times; targeting is mostly about making the most of your possible universe of voters. Issenberg leaves to others the rest of the story: the way in which issues and rhetoric are shaped by electoral demands. show less

If you've ever wondered about those super-personalized emails you get from political campaigns, or been intrigued about just how those pollsters know which demographics are "breaking" for one candidate or the other, you may want to pick up journalist Sasha Issenberg's new book The Victory Lab: The Secret Science of Winning Campaigns (Crown, 2012). By reading and exploring decades worth of research in political science, behavorial analysis, sociology and other fields, and by interviewing the men and women behind the experiments, Issenberg digs deep into the tactics now used to run political campaigns.

For a political junkie, this makes for absolutely riveting reading. Those with a more casual interest in such things may easily get bogged show more down in numbers and minutiae, but I loved every page. Issenberg discusses the real-time experiments campaign consultants run to figure out exactly which mailers or emails or phone calls are working best, and the demographic slicing and dicing campaigns can now deploy in order to get the most bang for their buck. He even shares what studies have revealed to be the most effective "get out the vote" technique yet discovered: sending lists of peoples' own voting records along with those of their neighbors and suggesting that an updated list will be sent after the election (most campaigns don't use this one, since people don't actually seem to like it very much at all, for some reason ... ).

If you haven't had enough of politics yet this year, and want a crash course in how the game is played these days, grab this book and settle down. show less

Issenberg offers a compelling account of how some campaign professionals and political scientists have worked to make the activity of winning votes a much more scientific enterprise. Behavioral science underpins much of this work and political science and our democracy will be better for it. Issenberg has done an admiral job of writing a book that makes this number crunching interesting and accessible to popular readers. Moneyball for campaigns indeed.

Really peels back the curtain on some of the statistical and data techniques modern campaigns use to isolate voters — as well as some of the historical battles data-minded politicos waged to win acceptance for their methods. (In Issenberg's book, first published in 2012, the apotheosis of the data-driven campaign is Barack Obama's 2008 and 2012 efforts, but his book makes it clear that each new development in tactics or methodology never remains cutting edge for more than one or two election cycles.) Lots of interviews with people in both parties who have pushed these methods, most of them obscure. Pretty well-written and a fast read. My highest tribute: reading about this stuff makes me want to run some regressions myself.

Two-thirds of the way in, this is very readable (reflecting Issenberg's background as a journalist) and interesting survey of the role (and lack) of experimental exploration in political science generally and in political campaigning specifically. It also traces the growth of "micro-targeting" or the practice of leveraging large data sets to predict which voters are likely to vote and who they are likely to vote for, in order to better target individuals and tailor the campaign message to those individuals. In the process, it provides an interesting perspective on the presidential races of the past few decades.

The quote from politico on the back of the book - "Moneyball for politics" - is an unfair comparison. Issenberg is a solid show more writer but he doesn't match Lewis's prose. I expect that this book will be wildly popular within the Beltway and policy wonk circles, but thus far I don't see it as appealing to the wide audience that _Moneyball_ did. It lacks the compelling overall narrative, partly because this is a bigger story with many more characters. It also buries the lead: all of the techniques described are used to gain a competitive advantage. In a few asides, Issenberg mentions the competitive dynamic (for example, he notes that in the 2004 race, Republican micro-targeting strategies made Democrats scratch their heads when Bush campaigned in traditionally Blue districts; the Dems didn't realize that the Republican data mining effort had managed to identify potential voters hidden in the mass of Kerry supporters) but he never takes the next step of explicitly exploring how these innovations affected the larger competition.

My provisional conclusion is that this would be an excellent book to give an undergraduate political junky. I expect that it will help create a new crop of political data-analysts. show less

The Victory Lab will appeal to political types. Sasha Issenberg's book traces the intersection of digital technology and behavioral science as they apply to elections and the marketing of candidates. There are interesting profiles of some of the many behind-the-scenes players in our political landscape, and perspectives on how current pols like Rick Perry embrace the scientific approach to a segmented electorate.

The most significant finding, perhaps, is that people are motivated more by negative than positive appeals: publicizing your voting record is more apt to get you to vote next time than asking you to do your civic duty. This has some disturbing implications for those who think negativity in campaigns affects the potential for show more bipartisanship in government. Keep an eye on your mailbox and listen for your telephone for the next salvo. show less

This book does a good job of vividly presenting what could otherwise be a dull topic. While discussing the statistical rigor of political polling and political science, the author keeps a great deal of focus on the people involved. While I would have liked to a more in-depth analysis on the stats end of things (most points are rather broad), the book describes the history, through both Democrat and Republican efforts, of how science has crept into an otherwise anecdotal and gut-based field. Non-partisan and well-written, I would recommend this book to someone interested in learning more about how a political campaign works (especially in light of the recent emphasis on Nate Silver and other statisticians joining the crowd, though this show more book was published before that happened). show less