DragonSworne
Banned
That's because they had a incompetent President to defend from the mean liberals. Trump needs no one to defend him.
Fox news has to go full RINO labeling on Trump if he becomes President.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
NBC poll: Trump continues to lead the GOP field after 1st debate
- Thread starter chadskin
- Start date
- Status
That's because they had a incompetent President to defend from the mean liberals. Trump needs no one to defend him.
Fox news has to go full RINO labeling on Trump if he becomes President.
Voysa Reezun
Banned
Single poll = SSS, plus as we probably should remember from 2008, ground game is key and doesn't necessarily show up in the polling.
TheLostBigBoss
Banned
Bam Bam Baklava
Member
Single poll? We've had three already if you include NBC, MorningConsult, and now PPP.
The scarier part to the GOP- Social liberalism + Tea Party hawkishness sells to GOP voters.
There are large factions of the party dedicated to making sure the social issues stay in the 17th century who do not want evolution here.
How sexist can this man really be?
RustyNails
Member
Ladies and gentlemen, the Leader of the Free World
We're at the point where we have to admit that we can't use 2012 as a blueprint to explain what's going on here. Trump's campaign is something new, something we haven't seen before so no one has any idea what to make of it or even how to report on it. He's basically a kaiju, and everyone else in the race is Tokyo.
Voysa Reezun
Banned
NBC's is not an Iowa poll, correct?
PPP is the first Iowa poll that I've seen.
Maybe our wires are crossed; I'm talking about Iowa, where I (sadly) expect Walker to win. I don't think a PPP poll showing cross tabs that Trump is winning everyone is necessarily reflective of reality, or at least completely.
On the other hand, if we see multiple polls leading up to the primary continually stick him ahead by 7-9 points over Walker, I'll have to become accustomed to Trump winning there, more than likely.
I still think the nature of the Iowa election, which is a caucus, gives a bigger boost to the people with the best ground game, however.
Veritigo_X
Member
Only a deviant would interpret his tweet as that.
It'll be Walker.
Voysa Reezun
Banned
This would be even worse than Trump, so I'm hoping Trump exposes him a bit/Cruz and Trump fans feel like they were funneled toward Walker by the establishment and raise hell.
Walker would be a disaster on a level that Trump only reaches in anyone's wildest nightmares.
Literally lol'd over this at work.
nah, Walker will be a much easier candidate to beat for the Dems than Jeb I think.
Surprised that Bush even has 12%.
titiklabingapat
Member
Dang, that's at least three national polling now and one Iowa.
Voysa Reezun
Banned
Maybe, but I don't feel comfortable at all with Walker coming any closer to running the country - or Cruz, for that matter. I think the Dems win anyway just because the numbers shake out in their favor, so if I have to risk a Republican leading the executive branch, I'd rather it be someone who would only partially destroy the country instead of totally destroy it, like a Bush or a Kasich.
friskykillface
Banned
Trump the god B^)
SwiftDeath
Member
Hmm isn't that one of the major polls?
Seems like Trump maintained or even gained a point or two
StopMakingSense
Member
I don't know what people were expecting. If he could survive and thrive after those McCain comments, last week wasn't going to do a damn thing to him.
Yes it is a major poll. As is the PPP Iowa poll posted earlier.
EatinOlives
Member
Trump's "campaign" isn't an inscrutable lighting-in-a-bottle stroke of genius. It's really easy to explain:
Old rich white male with a lot of money who doesn't need campaign funds and has no career in politics gets in front of a GOP megaphone and yells what uneducated people want to hear in the most boorish manner possible. People mistake his crass behavior as "anti-establishment" and ride the train, not focusing on his policy, lack of experience, dodgy business background, questionable career choices, corrupt dealings, etc., etc,., because he's the new political flavor-of-the-month, riding the wave during the GOP Primary Circus part of the cycle. The part of the cycle where none of this matters.
He's not a force of nature. He's appealing to the sect of conservative voters who were always going to vote Republican. He's taking the Tea Party mentality and cranking it up to 11. It's not a mentality that gets anybody into the White House.
Let's not pretend that his boneheaded approach to politics is any smarter simply because he has been sustaining his abhorrent bile for a longer amount of time than people thought possible. It's still just as ridiculous and unelectable as it was in his first day.
It's still the exact same rhetoric simply enclosed in a different candy wrapper. The "nominee of the week" thing was nominees telling staunch Republicans what they wanted to hear. The "gaffes" were only part of the reason why they fizzled out.
And no, when people say "nominee of the week" they don't literally mean one week. Herman Cain, Newt Gingrich, Rick Santorum all lasted longer than 1 week and people said the same exact thing: they're new, they're different, they're not like Boring Old Romney because he's just a tired bland politician while these guys are new and edgy. Trump is "newer" and "edgier" and has the shock value going for him, but we're still retreading old ground. Just because he survived the first debate largely by Fox News handing him a victory with their idiotic attempts at throwing "curveballs" at him doesn't mean he's bucking any trends.
Stumpokapow
listen to the mad man
interesting that they chose to report that "Bush trails" when statistically Bush and Trump are within the margin of error of each other.
Oh it's not smarter, it's just that traditional campaign logic doesn't apply to his brand of stupidity.
There is no reason to read it that way.
Dixon's Jet
Banned
Donald Trump is not a loser.
Jeb Bush is a loser. Nobody has time for that.
Jeb Bush is a loser. Nobody has time for that.
I don't think margins of error work like that.
Rubio also has the views of an unhinged crazy person. He is against abortion even in the instance of rape and incest for crying out loud.
Stumpokapow
listen to the mad man
When the substantive interpretation varies from "it's a dead heat" to "it's an unprecedented blowout", that speaks to the merit of reporting the information cautiously or not at all, not to conclude that because errors are centered on 0, 0 error is what has occurred.
Fenderputty
Banned
Agreed. I thought he was gonna take a hit for that at first.
CNN: Dead Heat
Sure, if this was an outlier it would be iffy but it is backed up by every other poll coming out. They aren't all wrong.
titiklabingapat
Member
This is the third national poll showing him in the lead though. The other one by that Morning Call or something outfit is more suspect but is also supported by NBC/Survey Monkey and Reuters/Ipsos showing similar leads.
You don't "add" margins of error, though. It's a statistically significant lead.
EDIT: "Statistically significant" may not be technically accurate, either. But it's definitely incorrect to say their numbers are within the poll's margin of error.
Stumpokapow
listen to the mad man
It is true that it is unlikely, given independent error distributions, that multiple polls would report the same substantive finding by random chance. I suspect Trump is in the lead (and there's no reason the debate would have changed that, nor the campaign drama of the weekend). I'm simply noting that it's not good practice, statistically, to report a result the way they did. If you subscribe to the p-value paradigm, the difference between their polling levels is not significant. If you don't subscribe to the p-value paradigm and instead favour substantive interpretation of margins of error, then context would be given (that the substantive effect observed could be anywhere from dead heat to blowout) which it isn't. Either way, not good statistical practice.
RustyNails
Member
FWIW, the NBC poll has a margin of error of 3.4%.
Margin of error falls within a curve, so while it's possible for Bush to be actually ahead of Trump while staying within the margin of error of that poll, it's not particularly likely.
I don't have the exact parameters of that poll, but assuming 95% confidence level I think it's pretty reasonable to say that the poll suggests the Bush trails trump.
p.s.
It could still be way off base, but bias is more likely to fuck with the results than sampling error.
EatinOlives
Member
Sure, there are untraditional elements to his campaign (largely that he doesn't have any big donors to answer to), but he's still treading very old ground: largely seeking popularity by firing up the crazies. McCain did it in 2008, albeit with a lot of dogwhistling. Many GOP primary candidates in 2012 later did it as well, with less dogwhistling. Now Trump is doing it with no filter whatsoever. While things are different, though, we certainly haven't reached a point where we can safely say he's bucking trends. He's still fitting an old pattern with a couple of new variations.
Stumpokapow
listen to the mad man
The margin of error exists for each candidate independently. Repeated sampling would indicate that Trump's support, 19 times out of 20, would be 17-30%. Repeated sampling would indicate that Bush's support, 19 times out of 20, would be 5-18%. You don't "add" the margins of error, but they exist for each candidate independently, so when you are comparing them, you are able to use both. The errors between candidates in a multi-candidate race should be uncorrelated (i.e. it is possible both that Trump has 30% and Bush has 18% or both that Trump has 17% and Bush has 5%).
One of those variations is that he doesn't give a flying fuck what the party looks like when he's done, which is very new.
Stumpokapow
listen to the mad man
Assuming a normal and uncorrelated distribution of errors, yes, it is more likely than not that these polling results show Trump ahead of Bush. But we don't report "more likely than not"
Either you are in the p-value paradigm and this is not significant, or you are not in the p-value paradigm and you don't care about a magic 95%, you care about the substantive interpretation, which implies that there's a non-trivial chance of any outcome from a total blowout to a tie. Saying something is "almost" statistically significant is not good practice. And it's mocked by serious social scientists and statisticians.
Again, I should stress my point is not "Trump isn't leading", it's that if people are going to report numbers as if they mean something, they should understand what they mean.
Livingskeletons
If I pulled that off, would you die?
Yeah thought it was pretty good, money in politics wouldn't have been brought up without Trump.
That's all true, but it's still not normal practice (at least in the news) to combine margins of error like that. What are the odds Trump is at 17% and Bush is at 18%? We can't do an actual t-test ourselves, but we both know it's almost certainly <05%.
Stumpokapow
listen to the mad man
The entire purpose of reporting the margin of error is to be able to do this.
Here is an example of NYT reporting in 2012:
http://www.nytimes.com/2012/09/11/u...ying-obama-bounce-is-sugar-high.html?_r=1&hpw
Why are they "combining" margins of error "like that"?
hydragonwarrior
Member
I want Trump to attack Walker's record again. Don't leave anyone unscathed!
EatinOlives
Member
Also true, which might affect the aftermath toward the end of his campaign whenever that may be, but isn't really affecting what's going on right now. What we're seeing right now isn't anything that we haven't seen before in some form or another.
- Status
Similar threads
- Locked
- Locked
- Locked
- Locked