arugula2 wrote:

willhud9 wrote:My mom works the polls every election and I dropped her off and voted. My only hope is that enough Sanders supporters in this state actually vote in the primaries.

What state, out of curiosity? Or region...

Virginia.

Eek, that one makes me nervous.

The_Piper wrote:Biden's is the brain that's cooked though. The man should go live out his golden years on a farm in upstate New York.

People were heartened by his SC outcome? Why? What could it possibly tell any prospective voter about him that was in doubt before? The world is doomed. Dooooooomed!

Added: I hope every one of those idiots behind him pooped their pants on the way home. And then voted sensibly.

Truths. The word you’re looking for is truths, Joe.

When I'm Biden's age, these video clips of his gaffes will still be circulating.
Don't get me wrong, everyone makes gaffes, but if you're running to be the president of the most powerful country on Earth, you should always bring your A game. Unfortunately we keep electing people who don't even have an A game.

edit- I forgot to add the word country after powerful.

arugula2 wrote:

OlivierK wrote:Well, you had a go at 538 having large error bars now, and you also had a go at 538 and Nate Silver's credibility generally (and I assumed your beef was the same as most people who scoff at Silver's credibility, who couldn't maths their way out a wet paper bag. Sorry if it's not.)

That was about aesthetics, presentation, not the underlying maths. I'm responding to them as a media company, not a statistical analyst... because they may functionally be the latter, but out where it matters they're media.

As a media company, they're talking inside baseball as they usually do at this stage. They're not out saying they know definitively from their model that Biden will edge Sanders by convention day, because that would be stupid, they're just publishing what they've got so far, as they do from the earliest polls, with appropriate acknowledgement that it could be very different by June, for stats nerds that want to look. Still worth more than just saying "let's imagine every state votes exactly like the average of February polls, and every state with no polling will go like the average of state with polling.", which is 270toWin's methodology, as far as I can tell. Neither approach is worth much after 4 states. But over time they play their part, and if projections based on poll aggregation are going to be out there, they might as well be rigorous ones, as 538's are, is all.

arugula2 wrote:

OlivierK wrote:It's not really important, in any case, but it just seemed (seems) odd that someone as into the detail as you would have such a negative opinion of a poll aggregator with a similar liking for detail. Especially if you then prefer 270toWin as a source, who take a broad, but lazy approach to the same task (for example not showing any confidence bars at all, and treating all polls as equally valid, regardless of their source, or more importantly, their methodology - LV vs RV, etc). Might as well just go straight to RealClearPolitics.

Yes, please! Big, sloppy voodoo numbers. That's what's appropriate for this moment in time. In fact, I'll trust a well-reasoned argument with an eye towards history and a sensitivity to trends that polls might be poor at highlighting... any day, even on election day. I want it to be a discussion, always, all the nuance on full display. And in defense of sloppy voodoo poll numbers... when you hold up those final 2016 poll averages next to the election results, they pretty much match.

Sure. We were pretty much having exactly that sort of discussion, without any delegate projections, until you quoted some, and 538's model was just sitting there, chugging away on early numbers, until you screenshotted it, and posted it here.

Anyway, results await, manoeuvring will be afoot, and by this time tomorrow, the delegate count projections might even have some better value.

I was zeroing in on the visual presentation because I’m putting myself in the shoes of probably the vast majority of 538’s consumers, for whom the nuance of the calculus is overridden by our weirdly visual-heavy cerebral makeup. 538 knows this, and they know how the vast majority of their consumers will interpret a visual like that, including a key 538 consumer base, the network media. I’m suggesting the honest thing to do would be to compensate for human visual bias, to more accurately project how they feel about the math. But it would be bad for business to do so.

arugula2 wrote:I’m suggesting the honest thing to do would be to compensate for human visual bias, to more accurately project how they feel about the math. But it would be bad for business to do so.

The pages I looked at include plenty of text explaining the underlying maths and what meaning they give to the visuals, but you know what they say about horses, water and drinking, right?

I’d rehash everything I said about visuals in media consumption, but it seems pointless. Yes, horses and water - except the media market is all about leading horses to the water you want them to drink. There’s still leading going on, is the point.

I’ve been following Biden for a couple almost three decades. When I was naive & less interested in domestic policies, I would’ve told you that he’s my favorite American politician - in his Foreign Intelligence days - not for any policy (this is pre-Iraq, btw) but because of the way he articulated nuanced ideas about the world, with a directness and an intellectual honesty. I loved his brain. When Obama picked him in 2008, regardless of how Iraq and some other things had changed the world, it got me interested. I wasn’t a voter back then, so that’s neither here nor there. I find villains fascinating, too, like Andrew Jackson - I wouldn’t mind picking his brain, in fact. I’m an equal opportunity brain picker. Obama’s candidacy suddenly looked interesting because Biden’s brain was interesting.

Fast forward to December 2016, closely following The Day of Blue Collar Reckoning... I spot Biden on Colbert’s Show:



It was probably the first protracted exposure since about the midpoint of his VP stint. Something had seemed weird in that choreographed skit at the 2014 WH Correspondents Dinner, the one with the yellow sports car? You know the one. Highly edited, and he still didn’t sound right. And he was notably absent from the dinner. He was absent from every dinner except 2016, apparently. Did anyone else notice either fact? I genuinely don’t remember him speaking even in 2016.

But on Colbert, something was glaringly wrong. His appearance was basically a 2020 test bubble. But he couldn’t even get through the rehearsed mini-skit(s) without completely losing his train of thought.

I haven’t checked for dates, but the long-rumored conversation in which Obama personally declined to endorse him, telling him it was too late, would probably have fallen in this post-election, pre-2019 period. Obama saw the obvious.

Now Obama is making phone calls to lesser candidates, and sending missives even to the New York billionaire, trying to carve a path for Biden. If anyone needed any further proof of the sheer cynicism of this man, look no further.

The same rumor mills that told us in November he was going to intervene if Sanders looked to surge; the same high-ticket functions in which he lectured young people in September against “wokeness” and the rash of “cancel culture” that seemed coincidentally to have singled out the crony capitalists this particular season; or patronizingly telling an audience in December that ‘aren’t we tired of old white men’ ruling the world... the ultimate “wokeness” douchebagery. Now Biden’s comment a few months ago that he’d be happy with a single term (who the fuck doesn’t cringe at that?) sounds like just another gross signaling of the feeble elite to themselves.

So let’s watch as Obama anoints - but coyly, in private, like an intellectual coward - the thing he has bluntly rejected for the past few years and cynically trashed only 2 months ago. Oh what a proud and stupid nation.

arugula2 wrote:I was zeroing in on the visual presentation because I’m putting myself in the shoes of probably the vast majority of 538’s consumers, for whom the nuance of the calculus is overridden by our weirdly visual-heavy cerebral makeup. 538 knows this, and they know how the vast majority of their consumers will interpret a visual like that, including a key 538 consumer base, the network media. I’m suggesting the honest thing to do would be to compensate for human visual bias, to more accurately project how they feel about the math. But it would be bad for business to do so.

It's a rigorous best guess at a very early stage, with wide confidence ranges acknowledged and accurately marked on the same graphic.

Remember, we got off on this tangent because you posted up some delegate projections without naming a source, and rolled your eyes when I noted the projection that you'd posted differed from 538's. You still haven't made clear what the source of your disdain for 538 is, but I find it hard to believe that it's that they don't sufficiently identify uncertainty in their projections, give your own nonchalant citation of a source that doesn't identify uncertainty in their projections at all, and that calculates those projections using a methodology with holes you could drive a truck through.

Anyway, this is all exceedingly dull. All we know as a result of this exchange is that you're not a fan of Nate Silver, because you think he's a media whore, or something.

More interesting things await. Let's get those Super Tuesday results, which will change everything or nothing, flowing.

Those delegate projections, like I stated at least twice already, were in *fun*. I don’t rely on them for anything but a diversion. I’ve typed some combination of these words more than once already.

My quibble with presenting statistical analysis with low confidence in a visual form for mass consumption, has everything to do with the nature of that consumption. Again... repeating. But since the first explanation (above) didn’t quite sink in yet, I’m not sure this one will have by the 4th (5th? 6th?) time. But let’s keep trying, I suppose.

Nah, let's not, let's just hope that people aren't stupid enough to put Biden over the top over the next few months.

I’m counting on Virginian suburbanites to not be as cynical as I think they probably are. And later, I’m counting on Michiganders to vote for their best interests - that one might be one of the votes which undercut the “moderate” resolve. In the meantime, 90% of the press, Obama, suddenly emboldened surrogates, and even how/when the early & provisional ballots are counted, are all against Sanders. Granted, California’s primary is much earlier this time, but it’ll still probably take weeks to get a real accounting of the outcome, which may proportionately favor the Bag of Regrettable Favors who just got the Obama treatment, even if just by making Sanders’s victory seem less dominant.

I almost forgot Beto. Pseudo-progressive Beto. Oh, and, TX is up to its anti-brown tricks again, of course.

arugula2 wrote:I’m counting on Virginian suburbanites to not be as cynical as I think they probably are.

It doesn't look like it's playing out that way: Biden doing very well early in VA, on track for about 70% of delegates with 40% counted.

The_Piper wrote:Biden's is the brain that's cooked though. The man should go live out his golden years on a farm in upstate New York.

Maybe he'll get to meet my old dog Shep that my dad took up to a farm when I was younger!

proudfootz wrote:

The_Piper wrote:Biden's is the brain that's cooked though. The man should go live out his golden years on a farm in upstate New York.

Maybe he'll get to meet my old dog Shep that my dad took up to a farm when I was younger!

OlivierK wrote:

arugula2 wrote:I’m counting on Virginian suburbanites to not be as cynical as I think they probably are.

It doesn't look like it's playing out that way: Biden doing very well early in VA, on track for about 70% of delegates with 40% counted.

It’s sadly an open primary meaning anyone can vote.

I’ve heard of anecdotal exit polls in rural areas in SC that suggest a bunch of anti-Shillary 2016 turnout just didn’t feel as motivated against Biden, and idk if SC is also an open primary, but if there are areas similar to that in other states, that could depress a non-trivial Sanders cohort. Though not so much VA... that’s a very blue-dog-leaning kind of base. I think Shillary did well there. It’s basically a DC satellite, but more wonkish (urban flight, etc.). Loooots of lobbyists.

This is being sold unironically at the joebiden storefront.