04/22/2014 05:54 pm ET

HUFFPOLLSTER: NYT's New Forecast Predicts An Even Chance Of Senate Takeover

Getty Images

The New York Times debuts 'The Upshot,' its data journalism project, and forecasts a tossup for control of the U.S. Senate. A political scientist explains what produced the Democrats' advantage among younger voters and why it might not last. And a GOP pollster gets even with a client who didn't pay her bills. This is HuffPollster for Tuesday, April 22, 2014.

NYT'S THE UPSHOT LAUNCHES... - The New York Times says The Upshot will "help people understand the news." David Leonhardt: "[W]e believe many people don’t understand the news as well as they would like. They want to grasp big, complicated stories — Obamacare, inequality, political campaigns, the real-estate and stock markets — so well that they can explain the whys and hows of those stories to their friends, relatives and colleagues. We believe we can help readers get to that level of understanding by writing in a direct, plain-spoken way, the same voice we might use when writing an email to a friend….One of our highest priorities will be unearthing data sets — and analyzing existing ones — in ways that illuminate and explain the news….Perhaps most important, we want The Upshot to feel like a collaboration between journalists and readers. " [NYT]

...and so does its Senate forecast - Upshot: "According to our statistical election-forecasting machine, it’s a tossup. The Democrats have about a 51% chance of retaining a majority. Every day, our computer churns through the latest polls and reams of historical data to calculate both parties’ chances of winning control of the Senate. Although the Democrats currently have a 51 percent chance, that doesn’t mean we’re predicting the Democrats to win the Senate — the probability is essentially the same as a coin flip." The Times has also made their model "open-source" by publishing the computer code used to generate it. [NYT Senate model, methodology]

Other highlights from the site's first day:

Nate Cohn on endangered Southern Democrats - "To take the Senate and consolidate their control of Congress, Republicans need only extend their stranglehold on Dixie to Arkansas, Louisiana and North Carolina, all of which voted for Mitt Romney in the 2012 presidential election. Yet the hopes of these three states’ incumbent Democratic senators — Mark Pryor, Mary Landrieu and Kay Hagan — are still alive. That may be surprising in light of the region’s lurch to the right, but it shouldn’t be: Incumbency is powerful. In the South, Democratic incumbents have won 85 percent of the time since 1990, and 77 percent since 2000….The Upshot’s Senate model, gives Ms. Landrieu and Mr. Pryor a real chance of winning re-election." [NYT]

Lynn Vavreck on swing voters - "If you want to understand the 2014 midterm elections, remember this simple fact about American politics: There just aren’t that many swing voters….The 2010 midterm elections highlight the relatively small number of swing voters. After winning with a wide margin and extraordinary enthusiasm in 2008, the Democrats suffered one of the largest losses of seats in any midterm two years later….But only a small percentage of voters actually switched sides between 2008 and 2010. Moreover, there were almost as many John McCain voters who voted for a Democratic House candidate in 2010 as there were Obama voters who shifted the other way….Turnout in midterm elections is always down from presidential elections, and Democrats routinely fight to return more of their voters to the polls than the Republicans….The 2014 fight is not over swing voters. It’s for partisans." [NYT]

-Jonathan Robinson (D): "@vavreck argues opposite of @ElectProject [Michael McDonald]...Not about swing voters...about mobilizing base. Food for thought." [@jon_m_rob referencing McDonald's recent blog post]

-Michael McDonald, responding to Robinson: "@vavreck is addressing dropoff between 08 & 10. I've written how it's a problem for Ds, but is overblown when you compare 06 & 10" [@ElectProject]

Whither the Times Polling Standards? - Via Twitter, Associated Press Polling Director Jennifer Agiesta asked, "Does @UpshotNYT follow the NYT's polling standards? The @vavreck article citing YouGov suggests not." She linked to the most recent version of the Times' standards, which, as of this writing, was still linked to by the Times' Poll Watch page. Respondents to YouGov's polls -- including both those cited in Vavreck's article and those conducted for HuffPost -- are selected from an "opt-in" panel of individuals who have volunteered to participate in their polls, rather than from a traditional random probability sample. The Times polling standards expressly forbid reporting of polls using an opt-in panel methodology: "Opt-In surveys are conducted with respondents who volunteer to answer the questions. Some polling companies pay respondents or offer other incentives for people to complete online questionnaires. In order to be worthy of publication in The Times, a survey must be representative, that is, based on a random sample of respondents. Any survey that relies on the ability and/or availability of respondents to access the Web and choose whether to participate is not representative and therefore not reliable." The New York Times did not immediately respond to HuffPollster's request for comment. [@jennagiesta, NYT Polling Standards, NYT Poll Watch]

Today's Upshot articles are not the first apparent exceptions made to the Times' standards. Vavreck did similar analysis of YouGov's panel data for a series of posts in 2012 on the Times' Campaign Stops blog, and Nate Silver incorporated both automated telephone and opt-in internet polls into his forecasting models and reporting during Fivethirtyeight's tenure at the Times. [NYT Campaign Stops]

PARTISAN LOYALTY STARTS EARLY - Dan Hopkins: "In the last decade, voters over 65 years old have become more Republican, even as the electorate as a whole has been trending in the opposite direction….It’s not simply that people get more conservative as they age, despite whatever Winston Churchill is alleged to have said. As late as 2000, there was little difference in presidential voting between the old and the young….When voters enter the electorate, still in their adolescence, their attachment to the political parties isn’t especially developed. They are less likely to think of themselves as strong Democrats or Republicans, and also less likely to see the world through that prism. That means that the political conditions at the time are especially important to the youngest voters. If the president in power is presiding over good times, young voters swing toward that party….The most Republican group is the one that came of age early in the Eisenhower presidency, with those turning 18 during the Reagan presidency coming in second...But the decades-old legacy of political conditions also highlights the problems facing our two major parties: Democrats have no lock on younger voters, and Republicans are bound by their advantage among older ones." [538]

WHY YOU SHOULD ALWAYS PAY YOUR POLLSTERS - Jon Ralston on a pollster for Sue Lowden's U.S. Senate campaign, who released a trove of negative data on the candidate after being stiffed by the campaign: "In a court filing two weeks ago to buttress his claim, [pollster Todd] Vitale submitted 150 pages of documents that seem like potential harbingers of the current GOP primary for lieutenant governor and tear back the curtain for a rare, granular peek inside a campaign, and a disintegrating one at that….Lesson, among many here: Don’t stiff consultants who have a lot of inside information about you….[T]hanks to the bitter parting, he has now released a road map for how to defeat her, showing she well knew – and knows – her own vulnerabilities." [Ralston Reports]

HUFFPOLLSTER VIA EMAIL! - You can receive this daily update every weekday via email! Just click here, enter your email address, and and click "sign up." That's all there is to it (and you can unsubscribe anytime).

TUESDAY'S 'OUTLIERS' - Links to the best of news at the intersection of polling, politics and political data:

-New York Gov. Andrew Cuomo (D) has a wide lead for reelection. [Siena]

-A Fox News poll finds independents more likely to back anti-Obamacare candidates. [Fox]

-Gallup sees Americans clustering into three groups on global warming. [Gallup]

-Benenson Strategy Group (D) finds Oregon's GOP primary nearly tied, with incumbent Sen. Jeff Merkley (D-Ore.) leading both Republicans by wide margins. [HuffPost]

-A Democratic poll finds Rep. Nick Rahall (D-W.V.) leading his Republican challenger. [The Hill]

-Ticket-splitting by district is at a new low. [WashPost]

-The Census may stop asking Americans about their toilets, commute, income and health. [Pew]

-James Ball wonders if data journalism is reaching a saturation point. [The Guardian]

-Eric Ostermeier reviews the history of first-term senators defeated in their run for reelection. [Smart Politics]

-Stuart Rothenberg regrets suggesting Rep. Cory Gardner (R-Colo.) could upset Sen. Mark Udall (D-Colo.) [Rothenberg]

-House Republicans are winning at Twitter. [WashPost]