You're reading way too much into this.

This is easy math for her to do, she just probably never looked into it before.

At first it was like "Haha look at the cable news anchor who devotes her life to this **** yet doesn't know how margin of error works" but now it just looks like you're trying to claim some superiority over something she just hasn't researched enough.

You don't think the average SAT score of Physics Phds isn't at least 785? Certainly at the top twenty schools they are. Remember that these people have to fully understand and be adept at advanced calculus, group theory, tensor analysis and I think, topology. They have to completely understand both Theories of Relativity and how they were derived.

Wookie could provide some valuable insight since he actually has a PHD in Physics IIRC. I think Max does too, not sure. I know borodog does but he isn't around.

I would think 785 is too low actually.

OK, OK. Let it not be said that in the face of overwhelming evidence such as this I won't admit that I am probably wrong.

This is bull****. I'm doing a math phd. Yes, perhaps, I have some natural aptitude for this. But mostly, I am good at math because that has been my passion and interest for the last decade. If I hadn't done that, if my passion and interest had been in political science, then whether or not I caught an "obvious" mathematical error just isn't a good way to judge whether or not I would have been good at math and could do a math phd. I have no idea what kind of aptitude she has for such things, but I highly question your implication that this margin of error business is that illustrative of it.

Its not a lot of evidence. And it would be less still if she hadn't devoted a whole segment to this error which shows she thought about it and was still not suspicious of her conclusions. But it changes the price that she is really smart by quite a bit in my mind. (And for those who think I might be biased because of some of her attributes I will say that it is almost inconceivable to me that non math major Venessa Selbst would have made this error.)

Just noticed this. No way its near 99% If MOE means there is a 5% chance you are outside of it then its about one in 400 that two tails touch. Fifth grade math.

changes the price, sure, but very small. Consider our evidence. She is obviously incredibly smart, has a phd, risen to the top in her field, etc. And she made a math error this one time. To use this as meaningful evidence against the proposition that if she had dedicated herself to a math based profession she wouldn't have succeed is just ludicrous. I think you are revealing yourself as subscribing to the false belief that no matter how obviously smart and accomplished someone js, if they lack a sort of innate ability to be "good" at math they couldn't be successful in it.

Frankly, you are making a FAR bigger error here than she made.

Why is math different from sprinting, golf, acrobatics, or composing music?

trying to compare the difference between her obvious intelligence, accomplishments and phd in one discipline to another discipline to sports immediately disqualifies you from the conversation.

As good as you think you are your quaint little probability problems, coming from someone actually doing a math phd, I'd suggest your series of basic errors here is much more disqualifying of you from being able to do a math phd as Maddow is from her error.

Too bad there is no way to bet.

(By the way I find it interesting that some mathematicians denigrate gambling type probability questions. There is little doubt in my mind it is because to do them creatively and quickly takes pure smarts and no advanced type knowledge. So if you are well educated, knowledgeable, but only semi smart, you can't hide that third fact when tackling these questions as you can when addressing more obscure ones. I don't think it is a coincidence that two of the most educated math science guys never go to the Poker Theory or Probability forums to discuss those questions.)

you're on cracking form these days.

Btw I think you are too focused on a theoretical physics phd from a top uni.

This seems inconsistent with your opinion that you have to have pure smarts to do a phd in math/physics

Almost feels self-serving to me. At first Maddow shows stupidity (and the inability to be able to deal with a high-level mathematical education) because she screws up a statistical inference that you wouldn't. She clearly couldn't do a math phd. Then once some well-educated math/physics guys disagree with you, suddenly the highly educated mathematicians aren't actually that smart, and dodge probability questions because it would expose them not being that smart, and again, not being able to do something you can.

90% of the gambling questions you hear are things like "what are the odds of AA holding up" or "what are the odds of set vs. set twice in one game" and such that are routine, prob 101questions. I would guess that a for-reals math guy wouldn't be super interested in them compared to famous unsolved theorems.

My understanding of Rhodes Scholars from Wikipedia is that they have strong literary abilities and like sports. Math didn't seem at all relevant.

I was never exceptional at math in HS, but I got through a lot of advanced math throughout undergrad/grad school purely because I thought it was interesting and I was willing to put in the effort. A big part of the discrepancy is that the material is just so much more interesting to me at the college level as compared to HS math.

This threads always seem to devolve or revolve, I guess, to or around the amazing intelligence of people who do math, unsurprisingly.

Halo effect.

People are focused on the Rhodes Scholar thing, but she stayed at Oxford a few more years to get a PhD (DPhil), a more rigorous undertaking. Her 383 page thesis is titled "HIV/AIDS and Health Care Reform in British and American Prisons." Most social scientists these days do quite a bit of statistics, so she may be more more up on stats than even someone like The Donald.

"Asked by students what kind of major she looks for in a successful job candidate, Rachel Maddow, the popular television host and best-selling author, did not hesitate in her answer. "I look for people who have done mathematics. Philosophy. Languages."..."And really," she concluded, "History is kind of the king."

Here 1994 undergrad honors thesis on dehumanization of aids victims begins with a discussion of statistics, though in an oblique, literary way.





source: https://publicpolicy.stanford.edu/pu...dehumanization

However good the Wharton School of Business may be, I don't have Sklansky's faith there is something magical going on in their classrooms. Undergraduates are required to do two semesters of statistics, and the syllabii for these courses don't suggest to me that the content is any different than it would be for introductory courses in these topics at any university in the country. And what if Trump happened to not feel like showing up the week they covered MOE?

"And you know numbers are numbers and my numbers happen to be great."

No way to bet, but thankfully we CAN try to make coherent arguments to defend our positions. It is the latter thing you failed to do.

And your aside is just awful. Mathematicians - like myself - denigrate your quaint little gambling type probability questions because they are fundamentally uninteresting in comparison to the type of math researchers are working on. They might be fun little puzzles to play with, but they will never be anything beyond toys to anyone seriously into mathematics. Mathematicians aren't interested in just going around proving their intellectual superiority, they are interesting in discovering deep and powerful truths. It has nothing to do with educated mathematicians being unable to do probability - I can't believe I'm having to refute such a ridiculous position.

The basic theme that seems to keep coming up is that you like playing around with this class of problems and have done so for decades. This is fine. What is not fine is trying to use competency on this particular class of problems as some sort of bright indicator for someones intellectual prowess. It reaks of "i'm good at x, and if anyone else is not good at x, they are stupider than me". You are massively overvaluing your own intelligence here at the expense of skills accrued over a large amount of time playing around with these types of puzzles. So when one person who is obviously smart, obviously accomplished, and has a phd doesn't immediately grasp a particular kind of problem, and use this to want to bet on their inability to accomplish other things like a phd in mathematics or whatever else is more or less just entirely nonsense. No part of your cute story stands up. Your implied theory on human intelligence doesn't work, your just hilariously false view of why mathematicians might not care about these problems, and the oh so subtle patting your own back about how intelligent you are because you like this class of problems is just the icing on the cake.

Note that I've taught thousands of students in mathematics over a range of disciplines. Little probability problems as a class aren't particularly unique in the field of mathematics at being able to suss out someones sort of raw intelligence, or whatever you are inaccurately going for. We work hard to inculcate a sense of mathematical maturity that grows creative and rigorous problem solvers across a range of disciplines. Probability isn't in any way special here.

As (I think) the person in this thread with the most accomplishments in math, I am happy to defend the position that math is not some special bastion of the amazingly intelligence and instead DS's skill, such as it is, for quaint little probability and logic problems are not a function of his amazing intelligence anywhere near as much as he thinks they are and that his dedication to this particular skilled domain vs any other is a massively important factor.

Are these the polls she used or even more recent ones (probably more recent, you updated, right? Are they still about 3000 total and 5 of them?) ? Because in this chart assuming the width in each box for 95% is ~2*((p*(1-p)/N)^(1/2) and N~3000 (makes sense in your chart to be near that eg judging from Trump's data and width) there actually is no issue with who is 10th+ vs the lower ones. I think her original polls data had it as close as some 2.6 vs 2.2 and 2 and 1.8 type things chasing with errors around 0.6,0.5% each that would have had them overlap frequently, unlike what is going on in this chart above. In this chart she has no argument basically for who is not included in the 10 top candidates other than imply the polls are not done properly/being representative etc which is another issue.

The essential difference here is that goofy is representing the data with a confidence interval for each candidate instead of a global margin of error for the entire poll. It's much easier for a layman to see that there's no overlap between the 05% intervals for 10th place and 11th place.