Should Coakley Cancel Her Globe Subscription?

This could be the week that Martha Coakley cancels her subscription to the Boston Globe.

Last Friday as literally thousands of Massachusetts’ six million residents awaited the Democratic Party convention, the Globe’s new Friday “Capital” section devoted to politics was in its second week of existence. As promised it delivered a poll following up on its inaugural survey of the prior week and here’s the lead:

Martha Coakley holds a commanding 35-point lead in the five-way race for the Democratic gubernatorial nomination and has broadened her edge in a potential general election match up with Republican Charlie Baker, according to a new Boston Globe poll.

So far so good and there was no change in the week against the Democratic rivals but there was versus Baker, thus:

The results indicated a notable shift in Coakley’s lead over Baker in a hypothetical general-election matchup. She moved from a 5-point edge last week — 37 percent to 32 percent — to an 11-point lead this week, 42 percent to 31 percent.

Here is what was missing from the story: any explanation of what could have moved Coakley from a five point lead to an eleven point lead over Baker. I can’t think of anything but let me know if I missed something.

For the sake of being diabolical though let’s say the lead never really changed and the true difference in the hypothetical Coakley-Baker matchup was just statistical noise.

Happily such random occurrences quickly return to a more accurate reading, but maybe not so happily in this case. Suppose that this Friday’s Globe poll shows that Coakley leads Baker by thirty seven percent to thirty two percent – why Coakley has dropped from an eleven point lead to a five point lead!

What is worse, there is a ready narrative explanation: Coakley, that terrible campaigner, underperformed at the convention, is no match for Steve Grossman and can’t possibly withstand the surging Don Berwick!

Maybe none of this will occur and perhaps if any of it does the Globe will not follow the easy narrative; but others will, pushed by her Democratic rivals. It is a reminder that no thirty-five point lead over a statewide office holder and an ideological insurgent stands up, but any five point drop is likely to play into the narrative of Coakley as a Democratic disaster. There will be stories of much wailing and gnashing of teeth by Democratic Party insiders. Coakley faces a real risk of the story snowballing.

The other major risk here is that “Capital” in its infancy becomes a major variable in the campaign narrative and not just a chronicler of Massachusetts politics.

The final risk is that I am all wrong but that pales against the risks to Coakley and the Globe. It does heighten my interest in reading “Capital” this Friday morning. And on the small chance I’m right it might be the last day that Martha Coakley subscribes to the Globe.





About Maurice T. Cunningham

Maurice T. Cunningham is Associate Professor of Political Science at the University of Massachusetts at Boston. He teaches courses in American government including Massachusetts Politics, The American Presidency, Catholics in Political Life, The Political Thought of Abraham Lincoln, American Political Thought, and Public Policy. His book Maximization, Whatever the Cost: Race, Redistricting and the Department of Justice examines the role of the DOJ in requiring states to maximize minority voting districts in the Nineties. He has published articles dealing with the role of the Catholic Church in Massachusetts politics and on party politics in the state. His research interests focus upon the changing political culture of Massachusetts.
This entry was posted in Mass Politics and tagged , , , , , . Bookmark the permalink.

11 Responses to Should Coakley Cancel Her Globe Subscription?

  1. Pingback: Headlines for Tuesday, June 17, 2014 » MASSterList

  2. John Walsh says:

    …..or maybe the first week was the noise and the second was true.

    …or maybe the swing shows the outer limits of the margin of error and this week will settle into the middle.

    … or early polls are dangerous things. Don’t feed them. They might trick you and then gobble you up and spit you out.

    Better idea for folks who care about who wins? Rather than GUESS at interpretations of early polls, talk to your friends and neighbors about your favorite candidate and IMPACT the late and important poll in November.

    #GOTV14 is in 20 weeks!

    • You are absolutely correct John about where supporters of every candidate should spend their time.

      Sometimes I think polls serve best the professional political pundits, paid consultants and political bloggers: gives them something to sound smart about.

      And the highest entertainment for us outsiders and volunteers is to compare and contrast the different interpretations written by the 200 insiders and wannabe consultants who actually read and understand the damn things.

    • Maurice T. Cunningham says:

      I’m also going to fully credit MassiNCPolling Group’s Steve Koczela for being able to get revealing insights into Twitter about the Globe’s methodology. Here it is:

      “Multiple waves in each week results make variation different than usual.”
      “Think of each week as total of 2 surveys. Changes are from both dropping oldest, adding newest”
      “So change from 5 to 11 pt lead could mean oldest was very close, or newest shows larger lead.”

  3. Maurice T. Cunningham says:

    John, I agree with your remark. But, I have in mind a study done by Boston University’s College of Communications “Project for Excellence in Journalism” that showed that after the Rasmussen poll shook up the Coakley-Brown race, coverage of Brown turned overwhelmingly positive, and that of Coakley negative. It is that sort of cascade effect that could bring the harm, and not polls which you are quite right will change over time. I hope Democrats follow your advice to get involved and work hard for their favored candidates – and Republicans too!

    • John Walsh says:

      Amen, to that!

      I remember the report and appreciate the perspective of history the Brown win gives us all – and haunts some of us.

      I think the report accurately chronicles the facts as they occurred but I cringe a little at the notion of the poll CAUSING anything. At is best, a good poll REFLECTS results that someone else caused. I guess if I were a journalist, I would be more worried if I thought a poll CAUSED a shift in the nature of the professional product I produced. Early in the report, the phrase “the press never saw it coming” touches on the dynamic of many elections but unfairly attributes this condition only to the press. Many others are often in the same boat – myself included weeks out of the special Senate election. (Of course the many understandable reasons for discounting Scott Brown’s viability have been on public display since then and particularly in his adopted political opportunity, I mean home, of New Hampshire, but the implications of that unforgivable mistake of taking an election for granted live to this day)

      Understanding what CAUSES a political election outcome is a worthy area of study to which you and your colleagues devote significant expertise and time. Thank you.

      To me, the answer is simple. In a healthy democracy, individual conversations between a candidate (or their supporters) and voters and conversations among voters cause election results. Ideally and inevitably.

      When candidates and supporters fail to do that, or fail to do it well, all kinds of less-effective tools can take that role and can have decisive, if less certain, impacts. I believe the tools of broadcast conversations are inherently and increasingly less effective and we are reaching a point where they begin to lose their one real alleged advantage: efficiency. Nice to have known you Congressman Cantor. (Actually, not so nice…)

      The benefit of putting your faith in something other than face-to-face conversations and investing your time and resources into building the infrastructure to facilitate and support those conversations are illustrated in many election successes. The difficulty of doing that work explains (I think) the reason any other strategy remains credible when proposed in the power points from consultants. The danger of trusting those tools are illustrated just as often.

      The parlor game of political interpretation is bun but it is much better employed looking in the rear view mirror. For an election that is on the horizon, talk to people if you want to impact the results.

  4. Jeff Semon says:

    Interesting that 20 weeks away was mentioned. Only 8 weeks after the primary. Professor Mo, thoughts on our late primary? It is definitely to Baker’s advantage this time around and will for sure hurt the Dem primary candidates, all of them.

    Could be a case of the Democrat’s efforts to protect their power (think special election for Senate vs. appointment) coming back to bite them?

    As for the Coakley poll discrepancy, the weather was warmer and sunnier from the prior week to this past one… good a reason as any :)

  5. Is this article referring to a future, yet unmaterialized, Globe narrative for Martha Coakley’s downfall in the polls?… Coakley could cancel her subscription, but that does not mean people will stop reading the Globe. So far, the Globe has been tough, but fair on Martha: (search for ‘Coakley’ in the article listing below)

  6. Maurice T. Cunningham says:

    Thanks to all for the thoughtful posts, I know it takes time from very busy days. I learn a lot from the posts though.

    Awhile back I came across a post over at, a polisci blog, “Imagine there’s no polling. It’s easy if you try.” It is a brief but telling post on a study conducted about the effects of polling on elections in European democracies — many of which ban polling in the final week to ten days or even a month of an election! The study shows that polls tend to reinforce the front runner — the opposite of the effect we have been discussing. But there is an effect. Here’s the link

Leave a Reply

Your email address will not be published. Required fields are marked *