I agree to some extent, but, at the same time, she does talk about polls fairly often and she sits in on MSNBC's electiom night coverage, so I'd hope that she would have already known this stuff or that she'd have numbers people on staff who would have explained it more thoroughly to her before now... I mean, she does try to present herself as relatively data driven, although not quite as wonky as someone like, say, Chris Hayes.

She is a performer. She has a way of talking that makes it sound like she knows of what she speaks. That is why they hired her, she's a good actor.

She excels on the show because she is largely scripted. At live events like election night she's not that good.

She doesn't really know or understand things any more than a random person does LDO. Math is obviously a weak point for her.

I think this really understates her role. She isn't just a performer she is a Rhodes scholar who develops the content of her show.

And her objective is to poke sticks at conservatives. Seems to be doing a fine job.

She is better at that than attracting an audience.

+1. Good stuff.

Yes this is definitely true. Its profound that she makes the error to see that adding 5 polls keeps the error still as the error of each poll (it ought to at least reduce it as suggested or a bit higher than suggested but certainly much lower than 4% still). This is one error. And the other error is to not adjust for the tails when calculating small probabilities far from 50% (ie multiple opponents polls, see my previous post).

Now one cannot hold it against her for not being aware of such adjustments if it werent for the fact she has participated in many occasions in election results and polling opportunities discussions (over 10 years?). But its ok, its a math issue. The first error may be even more important to make because she fails to notice that improving the number of people polled must improve the accuracy also. So lets not gang up on the errors but notice them anyway.

What DS says that is important (that definitely make is thread worthy) is that the ability of a person (however brilliant) to make correct statements and judge properly situations is impacted (more than this occasion here that ultimately is not a big deal itself) if they are unable to recognize making such error as here. What is unimportant here becomes very important elsewhere because of what it suggests its possible to happen there too. That of course is true for all of us when we make errors. You have to gauge both the error and what not noticing it says about the person (and still be a bit open minded and forgiving even then to avoid being absolute because it might have been a one time thing, bad busy day etc).

It will have an impact in a variety of critical issues that may be important to not make such errors in order to arrive at more appropriate more rational and fair/objective positions there. That can be issues that relate to major current topics from guns to police violence to racial tensions etc where statistics does make a difference in the argument and where if you are supposed to be the correct side most of the time you deserve to have clarity in such arguments that the ability to make such error may slightly compromise elsewhere.

Of course lets not even start with other people in other channels and situations. It is unfair to target her and ignore the others because she is one of the best in general, even if sometimes very partisan and overly enthusiastic, she is still in general good in her topics making valid points and the point she made here still stands true even after corrected , regarding the bottom 6-7 people and the inability of the poll to be clean about who is the true 10th.

I dont think people realize that people with a daily show are pretty much writing that show (with staff) all day long prior to the show. This includes Jon Stewart, Maddow, and Bill O, even Beck/Hannity.

It's a pretty hard job, and they are not simply reading 'cue cards' written by other people.

Maddow is among the smartest people on tv. If she wanted to be Nate Silver, she could be. She probably couldnt be Skalansky, but he couldnt be Maddow.

Her show mostly annoys me, but I thought her book Drift was an intelligent read.

I liked her when I was a liberal. I haven't seen much of her since but she seems at the very least to be an intelligent, well-meaning, empathetic person.

I skipped 3 years and went to an ivy league school and the mistake was not obvious to me until I googled MOE and read about it this morning. I have never taken a stat class and have never cared enough to look into MOE in detail, so it just didn't really cross my mind that x% MOE is a max and actual MOE varies across the spectrum. Sure it makes sense, my mind just never left one-dimensional thinking wrt it because I never gave 2 ****s about MOE beyond knowing it's measuring variance relative to sample size

Sklansky, you've in a way dedicated your life to numbers/stats so not really fair to judge people outside of the field as stupid for not knowing details of statistical metrics. It's an embarassing mistake for her, especially in making argument vs Fox, but it doesn't make her a babbling idiot as you're implying

Right, and then you look at that the fact that there are a few other people with slightly lower numbers and you're probably looking at a situation where the tenth spot is more likely to be determined by luck than by actually being ahead (it does seem close though).

Of course its easy to say something like "Its the tenth spot, **** em", and I really wouldn't have a huge problem with that. But I think a better idea would be not have a fixed set number of participants and have some sort of methodology for eliminating a group at the end. In this case it seems like top 8 would make a lot more sense since there's a bigger drop off from eighth to ninth.

Burn!

Cue cards? Dude teleprompters have been around since 1950 son.

Watching the segment now, the other thing it looks like she's missing, is that averaging polls impacts the standard error. Basically, averaging polls increases your N to be the total N of all the polls. So saying "all these polls have a standard error of 3" doesn't mean "when I average these polls the average will have a standard error of 3." It means "When i average these polls they have a standard error much less than 3."

edit: I think the pollster doesn't really correct her because her point is still well taken: using polling to decide places 8/9/10/11/12 is going to be largely arbitrary, and will likely have a huge impact on the fortune of the candidates in spots 11 and 12. And yes, she should have done her due diligence on standard error, but shows are produced daily and move fast etc. The idea that this should be taken as any evidence of her being stupid is laughable.

The "margins of error" on these are statistical fictions anyway -- I would guess the bias from weighting responses to fit demographic models exceeds sampling error.

I already covered that with like 3 words at the bottom of my initial post

Groucho: Look Einstein, here’s Coconut Manner. No matter what you say this is Coconut Manner. Here’s Coconut Manner, here’s Coconut Heights, that’s a swamp, and, uh, right over here by the road fork, that’s Coconut Junction.

Chico: Where you get Coconut Custard?

Groucho: Alright that’s on one of the forks. You probably eat with your knife so you won’t have to worry about that. Now here is the main road leading out of Coconut Manner, that’s the road I wish you were on. Now, over here on this side we’re going to build an eye and ear hospital, this will be a sight for sore eyes. Understand? Now, right over here this is the residential section.

Chico: People live there, heh?

Groucho: No, that’s the stockyard. Now all along here, this is the river front, and all along the river, all along the river those are all levies.

Chico: That’s the Jewish neighborhood?

Groucho: Well we’ll passover that. You’re a peach, boy. Now here is a little peninsula and here is a viaduct leading over to the mainland.

Chico: Why a duck?

Groucho: I’m alright, how’re you? I say here is a little peninsula and here is a viaduct leading over to the mainland.

Chico: Alright, why a duck?

Groucho: I’m not playing “Ask me Another”, I say that’s a viaduct!

Chico: Alright, why a duck?! Why a-why a duck? Why-a no chicken?

Groucho: Well, I don’t know vianochicken, I’m a stranger here myself. All I know is that it’s a viaduct. You try to cross over there a chicken and you’ll find out why a duck.

Chico: I wasn’t- (mumbles)

Groucho: It’s-it’s deep water that’s why a duck. Deep water.

Chico: That’s alright.

Groucho: Look, look, suppose you were out horseback riding and you came to that stream and you wanted to fort over. You couldn’t make it, it’s too deep.

Chico: Why do you want with a fort if you have a horse?

Groucho: Well, I’m sorry the matter ever came up. All I know is that it’s a viaduct.

Chico: Now, look, I catch onto why-a no horse, why a chicken, why this, why that, I no catch onto why a duck.

Groucho: I was only fooling, I was only fooling! They’re gonna build a tunnel there in the morning. Now is that clear to ya?

Chico: Yes, everything except the why a duck.

Groucho: Well, that’s fine, then we can go ahead with this thing. Now look, I’m gonna take you down to show you the cemetery. Ive got a waiting list of 50 people at that cemetery just dying to get in, but I like you.

Chico: yeah, you’re my friend.

Groucho: I like you and I’m gonna shove you ahead of all of them. Im gonna see that you get a steady position, and if I can arrange for it, it’ll be horizontal. Now remember, when the bidding starts, somebody calls out a hundred dollars,

Chico: I say two hundred.

Groucho: That’s grand, and if somebody says three hundred?

Chico: I say four hundred.

Groucho: That’s great! Now you know how to get down there?

Chico: Oh no.

Groucho: Now look here, you go down there down that narrow path there until you come to that little jungle there, you see it? Where those palms are? There’s a little clearing there, a little clearing with a wire fence around it, you see that wire fence there.

Chico: Alright, why a fence?

Groucho: Oh no! We’re not gonna go through all that again!


Best wishes,
Mason

Well, DS for one, because she's a smart, well-educated, wealthy, liberal lesbian who likely has slept with more women than he has (prostitutes excluded).

But this is Maddow's bread and butter, which is why DS made the op.

Someone plz give me the cliffs of mason's post

For those who don't like teh maths, here is a picture:



Each of the four plots is created by:

• Assuming we know the actual level of support among primary voters. (Either 1%, 4%, 10%, or 20% support, top-to-bottom and left-to-right.)
• Randomly sampling 1000 of these voters for a poll, and recording the observed level of support.
• Repeating such a poll 1000 times and plotting the results.

The red dotted lines show the true* 2-standard deviation margin of error about the actual support %. So, for example, if your true support is 4%, then 95% of polls will show your support falling in the interval (2.76%, 5.24%).

* Assuming convergence to the asymptotic normal distribution, for those picky about teh maths.

Not only that but even the results right now (even if the polls were done more properly) reflect only some sporadic volatile education of the greater public (ie heavily manipulated by superficial events not the substance) about the candidates. Very few have actually been studied properly and some of them are probably better thinkers and potential leaders (albeit still republicans lol) than the current poll leader and deserved better numbers if the system worked in a more representative, democratic, stick to important things and not fame through media manipulations, logic.

At this level attacking McCain and trashing immigrants without framing his issues more appropriately for example for Trump improved his numbers for sure even if it was the wrong thing at many levels. He is clearly a winner by manipulation as he has been doing all his life rather than a man of principles and consistent inspiring ethics. Then again its politics what can be expected as it works worldwide.

Anyways, David is obviously right, this is not a good look for Rachel.

But not imo because it makes her look "stupid" in terms of genetic endowment or whatever; rather because anyone who has ever paid close attention to how natural phenomena distribute will have an almost overwhelming intuition that, "The standard deviation is probably correlated to the mean"...since, well it probably is in real life.

So she comes across as inattentive. Which is not a good look for a news anchor or whatever she is.

Thoughts on math literacy of msnbc's core demo?