Wednesday, October 13, 2010

Understanding Under 50%.

Real Clear Politics has an article that further elaborates on the famous "under 50%" rule for incumbents. Considering Strickland's consistent failure to top 45%, it's particularly significant.

In it they highlight six important caveats when discussing the rule, and they are interesting to consider when thinking about the Ohio Governor's race. My comments on each as they apply to Ohio are in bold.
  1. The rule is weaker in races where the incumbent is just under 50 percent. [Does not apply]
  2. The rule is weaker in races that haven't really engaged yet, where candidates have huge leads. [Does not apply]
  3. Beware polls that don't push leaners. [Only Rasmussen does]
  4. The rule is weaker in races where the incumbent is an appointee or is succeeding a retiring governor. [Does not apply]
  5. The rule is weaker in races where the challenger is a de facto incumbent. [Kasich is still largely unknown, does not apply]
  6. It doesn't apply to races where there is a strong third party challenge for obvious reasons. [Does not apply]
You can click here for the full article and deeper explanations of the caveats, but it's clear that the caveats for the famous Under 50% rule don't apply to Governor Strickland. In other words, as long as Strickland is running under 50% and behind Kasich, it's bad news for Strickland.

Moving on to the latest polling news, Rasmussen has the Governor's race back into a toss-up with Kasich leading 48-45. The partisan breakdown isn't necessarily good for either candidate with Kasich winning GOPers 83-11 and Strickland winning Dems 78-17. Kasich wins Independents by just 47-42. That's a substantial change from two weeks ago when Kasich was ahead with Indies by 59-26. Considering Rasmussen's relatively small sample size, you have to take the Indie numbers with a grain of salt. The true number for Rasmussen is likely somewhere between the two.

Overall, the Rasmussen poll differs from other recent polls by showing more tightening of the race. Obviously, that's not good. That said, it's just one poll. Overall, this won't improve Strickland's standing in the RCP average above 43 points, if that. The same level it's been since January. And as the incumbent rule states, that's not good.

10 comments:

  1. Wow... you keep moving the goal posts:

    1. A month ago: All that matters is the aggregate. Overnight, Kasich lost over 60% of his lead in the aggregate.

    2. Two weeks ago: All that matters is that Strickland hasn't gotten to 45% or above consistently. Meanwhile four independent polls, including Rasmussen, confirm Strickland's internal numbers having him at 46%.

    3. Now: Never mind that, now there's nothing to worry about until Strickland polls above 50%. (Duh, then the race is officially over.)

    4: Two Weeks from now: Polls don't matter.

    Two months ago, you said the DGA was pulling out of Ohio. Today, its chair was campaigning in Ohio for Ted Strickland.

    You keep moving the goal post and losing.

    ReplyDelete
  2. Oh, and btw, Rasmussen now joins RCP in calling the race a tossup.

    ReplyDelete
  3. LOL! Wow. Soooo wrong.

    1. I've ALWAYS stuck with the aggregate. That's why I was posting Silver last week. That's why I mention Strickland's average in this post.

    2. "Under 50" has been around loonnnng before me. I didn't make it up. Also, I beg of you, what polls show Strickland at 46? Rasmussen has him at 45. All others in the past two weeks (your qualifier) show him at no more than 43.

    3. Huh?

    4. Wha'?

    1st off, I was questioning whether they were leaving or not based on several indications that they were. I never said they definitely were bailing.

    Other than the DGA chair, I haven't seen anything of what really matters - and we both know that's money.

    ReplyDelete
  4. Wow. You even deny that which you've written? The DGA has ads up in Ohio right now. They're helping fund the ODP's GOTV effort.

    But you still want to pretend they're not in Ohio on the same day its chairman is campaigning here? LOL.

    You've moved the goalposts, Keeling. Anyone who has come here and read this will see that I'm right.

    Then there's this outright lie:
    "Overall, the Rasmussen poll differs from other recent polls by showing more tightening of the race. Obviously, that's not good. That said, it's just one poll."

    There are only polling outfits that have consistent polled in Ohio to pull trend lines that have released results in the past month (Rasmussen, Quinnipiac, Fox News, Reuters).

    Of those all but Fox News has shown the race tighting. So actually, the opposite is true.

    Not only do you not know how to read polls, you apparently don't know the meaning of the word "consistently," either.

    In the past twenty-four hours, two polls have Strickland at or above 45%. Over the past month, FIVE out of TEN polling organizations have shown Strickland at or above 45%. Not "just one poll" HALF the pollsters have shown this. And yet, during that time, you have claimed that Strickland has "consistently" polled at 42%.

    You're just objectively wrong, Jon.

    ReplyDelete
  5. If the DGA is up with ads right now, then I retract. Once again, I only questioned - never objectively said they were leaving. As for the Chairman being here - whoopity do.

    Now let's go through your polls that you say have shown in the past month things tightening.

    1. Reuters. They have polled twice. The first time was in EARLY AUGUST.

    2. Quinnipiac went from 17 (which you told me I should ignore) to Kasich +9. Yes, a tightening, but most serious analysts would simply call it more realistic.

    3. Fox News shows things improving for Kasich.

    4. Rasmussen's poll released today shows a tightening that we haven't seen in a month.

    All that said, I do agree things have tightened. But not because of any substantial or consistent increase in support for Ted. And that's bad news for the Governor. See my post yesterday focusing on "42", until today's Rasmussen, every poll in the past two weeks had shown the Governor at no higher than 43. That's very bad news for your side.

    As for your five of ten stat, I have no idea what you're talking about.

    See here: http://www.realclearpolitics.com/epolls/2010/governor/oh/ohio_governor_kasich_vs_strickland-1078.html

    On September 25th, Reuters had Ted at 46.
    On September 20th, the Ohio Poll had Ted at 45.
    Today, Rasmussen had Ted at 45.

    Your continued ignoring of Ted's flatlined support tells me all I need to know about you.

    Now, if Ted expands on the Rasmussen poll and starts polling above 45 and closer to 50, then I'll get more worried. But not until then.

    You also say that "during the past month" I've claimed Ted has polled no higher than 42. That is objectively WRONG.

    I said in the past two weeks. And I was talking about his average. Four polls had him at 42. One at 41. One at 43.

    Now quit wasting my time with such objectively WRONG analysis.

    ReplyDelete
  6. Cranky much?

    Um, Rasmussen showed the race tightening last month, too. From double digits down to single ones. You dismissed that as "just one poll" again. You were wrong.

    1. Thank you for admitting that Reuters showed the race tightening, too.

    2. You mocked me for my analysis of Quinny at the time, now you admit it was valid all along? How honorable of you.

    3. It does? Fox News went from a six point lead to a five point lead. That's not a sign of improvement. Regardless, and this was my point in the prior comment, it is the ONLY poll that isn't showing the race tightening. In your post you claim the opposite was true.

    4. You keep focusing on the "past two weeks" because only then does the data support your conclusions.

    If you look at the past month, you see that half of the polling organizations that have released results show Strickland at or above 45%. HALF of them.

    Rasmussen shows support for Strickland is going up. So did Quinnipiac and Reuters. You keep pretending they don't because you engage in deceptive data selection to fit your predetermined narrative.

    You keep telling yourself all is well and this is just like the New Jersey race even though the polling data is far more favorable for Strickland than it ever was for Corzine.

    ReplyDelete
  7. More distortions from you. Awesome.

    Reuters is removed from the analysis because the 1st of 2 polls was in early August.

    You told me Quinny sucks. I think they've had sampling issues, but their new likely voter model should be paid attention to. I'm happy with 50-41.

    Fox was at 2 in late September. It improved to 6 and now is at 5. I consider that an improvement.

    Rasmussen's latest shows things tightening, but it's JUST ONE POLL.

    Are you looking at something other than the RCP link to all polls in the race? I honestly have no idea what you're talking about.

    Are you comfortable with Ted's average of 43% in RCP?

    I keep using 42/43 and the past two weeks because there is a diverse and substantive sample. It's also an average used by RCP AND Pollster.com AND Nate Silver. If you want to use qualifiers different from those guys, fine. That's up to you.

    It's just not serious analysis.

    ReplyDelete
  8. Keeling you're such a moving target it's pathetic.

    You harp on the average issue by focusing on a narrow two-week window, you dismiss Reuters showing a tightening race because the previous poll was in August, but you look at a wider range of data from Fox News... however, you AGAIN cherry pick the time period simply so you can claim that Fox News' poll shows Kasich surging (when anyone who's looked at the history of Fox News' polling would tell you that your +2 baseline appears to have been nothing more than a statistic hiccup result.)

    You, in your own words, wrote this:
    ""Overall, the Rasmussen poll differs from other recent polls by showing more tightening of the race."

    Yet the ONLY poll you've cited to support that proposition is the Fox Poll and you had to, yet again, engage in some careful statistic hijinx to make that argument.

    What are the other polls, plural, that are not showing the race tightening? There are none.

    Heck, even Fox News doesn't really help you. Besides picking a baseline that forces the data to conform to your predetermined and desired outcome, you ignore one simple fact: Fox News and Rasmussen use the same pollster and methodology! They're basically Coke and Diet Coke to one another. And the Rasmussen poll is the most recent data now. You want to pretend that Kasich's lost 75% of his margins since August isn't concerning? You're alone in that regard.

    If Strickland can manage another five-point swing in the NEXT Rasmussen, we'll hit your "official panic" level. Of course, by then, there will only a few days left. It'll be too late.

    You want to pretend that the Kasich campaign shouldn't be concerned that they went from double digits leads in August to virtually tied in October, fine.

    You keep talking like Silver and RCP have said that Kasich's already won. RCP, like Rasmussen, say the race is a tossup and, like Silver, they've all noted that the wind seems to be blowing into Strickland's sails this month.

    But you're the same guy that laughed off the mid-September Rasmussen results that showed the race going from 12 point to 9. Now it's down to three.

    And yet, here you are singing the same tune you said in November. It's. just. one. poll.

    ReplyDelete
  9. Keeling you're such a moving target it's pathetic.

    You harp on the average issue by focusing on a narrow two-week window, you dismiss Reuters showing a tightening race because the previous poll was in August, but you look at a wider range of data from Fox News... however, you AGAIN cherry pick the time period simply so you can claim that Fox News' poll shows Kasich surging (when anyone who's looked at the history of Fox News' polling would tell you that your +2 baseline appears to have been nothing more than a statistic hiccup result.)

    You, in your own words, wrote this:
    ""Overall, the Rasmussen poll differs from other recent polls by showing more tightening of the race."

    Yet the ONLY poll you've cited to support that proposition is the Fox Poll and you had to, yet again, engage in some careful statistic hijinx to make that argument.

    What are the other polls, plural, that are not showing the race tightening? There are none.

    Heck, even Fox News doesn't really help you. Besides picking a baseline that forces the data to conform to your predetermined and desired outcome, you ignore one simple fact: Fox News and Rasmussen use the same pollster and methodology! They're basically Coke and Diet Coke to one another. And the Rasmussen poll is the most recent data now. You want to pretend that Kasich's lost 75% of his margins since August isn't concerning? You're alone in that regard.

    If Strickland can manage another five-point swing in the NEXT Rasmussen, we'll hit your "official panic" level. Of course, by then, there will only a few days left. It'll be too late.

    You want to pretend that the Kasich campaign shouldn't be concerned that they went from double digits leads in August to virtually tied in October, fine.

    You keep talking like Silver and RCP have said that Kasich's already won. RCP, like Rasmussen, say the race is a tossup and, like Silver, they've all noted that the wind seems to be blowing into Strickland's sails this month.

    But you're the same guy that laughed off the mid-September Rasmussen results that showed the race going from 12 point to 9. Now it's down to three.

    And yet, here you are singing the same tune you said in November. It's. just. one. poll.

    ReplyDelete
  10. Keeling you're such a moving target it's pathetic.

    You harp on the average issue by focusing on a narrow two-week window, you dismiss Reuters showing a tightening race because the previous poll was in August, but you look at a wider range of data from Fox News... however, you AGAIN cherry pick the time period simply so you can claim that Fox News' poll shows Kasich surging (when anyone who's looked at the history of Fox News' polling would tell you that your +2 baseline appears to have been nothing more than a statistic hiccup result.)

    You, in your own words, wrote this:
    ""Overall, the Rasmussen poll differs from other recent polls by showing more tightening of the race."

    Yet the ONLY poll you've cited to support that proposition is the Fox Poll and you had to, yet again, engage in some careful statistic hijinx to make that argument.

    What are the other polls, plural, that are not showing the race tightening? There are none.

    Heck, even Fox News doesn't really help you. Besides picking a baseline that forces the data to conform to your predetermined and desired outcome, you ignore one simple fact: Fox News and Rasmussen use the same pollster and methodology! They're basically Coke and Diet Coke to one another. And the Rasmussen poll is the most recent data now. You want to pretend that Kasich's lost 75% of his margins since August isn't concerning? You're alone in that regard.

    If Strickland can manage another five-point swing in the NEXT Rasmussen, we'll hit your "official panic" level. Of course, by then, there will only a few days left. It'll be too late.

    You want to pretend that the Kasich campaign shouldn't be concerned that they went from double digits leads in August to virtually tied in October, fine.

    You keep talking like Silver and RCP have said that Kasich's already won. RCP, like Rasmussen, say the race is a tossup and, like Silver, they've all noted that the wind seems to be blowing into Strickland's sails this month.

    But you're the same guy that laughed off the mid-September Rasmussen results that showed the race going from 12 point to 9. Now it's down to three.

    And yet, here you are singing the same tune you said in November. It's. just. one. poll.

    ReplyDelete

No profanity, keep it clean.

Note: Only a member of this blog may post a comment.