What's new
By:
USCHO Fan Forum

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

  • The USCHO Fan Forum has migrated to a new plaform, xenForo. Most of the function of the forum should work in familiar ways. Please note that you can switch between light and dark modes by clicking on the gear icon in the upper right of the main menu bar. We are hoping that this new platform will prove to be faster and more reliable. Please feel free to explore its features.

Biden v Dump 1: If not now, when? If not us, who?

Status
Not open for further replies.
my absolute favorite thing on Twitter is all of the other election model makers who just incessantly b—tch about 538 with such massive inferiority complexes. It’s hilarious.

All they do do is try to pick holes in Nate’s model. 90% of their arguments boil down to one of those stupid alternating capital-lowercase replies. A lot of them argue in bad faith. Silver has started to block some of them. It’s funny because they crawl back to their own Twitter and act like complete martyrs. It’s really kind of incredible to watch really smart people act like such children.

Instead of bitching about 538, do better. Build a model that outpaces 538. Don’t worry if he’s wrong. If he is, he’ll be exposed. If he isn’t, shut the fuck up.
 
Well Nate is hardly perfect...but you can tell the people who know what they are talking about in their arguments against him. (most would not be whining on twitter about it) And in reality there is no model you can make that is going to nail it because there is just too much uncertainty in polls and people in general.

Most of the whiners dont even have the brains to realize you use Nate to look for trends not to get a true prediction.
 
I mostly agree with that. I think people like Bitecofer are fairly intelligent in their criticisms. But one of the people I follow is at The Economist and he’s a really intelligent guy. But he spends most of his time complaining about 538 or explaining why his model is better. Which is a stupid thing to say anyways.

Models aren’t trying to do the same thing. Even ones that are trying to predict the election results. Yes, in general, they are trying to predict the results. But there’s a narrative on the other side of that coin and it’s the story the model maker(s) are trying to tell. It’s the sum of their experiences. They factor in very different things with very different results. Nate’s recently had a 10% chance that the GOP wins the presidency, the senate, and the house (I think). Do I think there’s a 10% chance? No, that seems insanely high. Do I think Nate thinks that? I don’t think so either. I think he’s trying to use that number to say, and this was discussed here recently, shit can happen and we can’t predict that. He’s trying to tackle the unknown unknowns. Other models don’t try to factor that in. Or they might do it in a different way. Neither are wrong. It’s a whatever the artist decides to put on the canvas today.

Where I think Nate has some wisdom the other modelers lack, is honestly that 2016 race. Most of the name brand models went awfully quiet after that. Or they changed their game. 538 didn’t. And that’s why I think the uncertainty 538 adds is a better model. And I say this as someone whose job it is to create detailed, statistics-based estimates of complex multimillion-dollar projects with minimal information. That uncertainty is what kills you. What you know is easy to model and predict. It’s essentially useless.

it’s why I appreciated 538 throughout this cycle. You can tell that model has real world experience but it’s telling such a different story than the others.
 
Oh Nate and his crew are great I follow them for that just that reason. They break it down in ways even I can understand. (and I suck at this stuff) Sometimes I think he is a bit gunshy but I dont think that is out of fear of being wrong but more that some people seem to deify him and 538 to levels that are kind of insane. He is a stats based person so it is about the percentages not about who he thinks will win.

He caught some grief because his models showed Hillary was the "likely" winner in 2016 and the media ran with that and really pushed him out in front of the stats craze. The problem is they didn't actually break down his stats they just looked at the likely outcome and extrapolated it to to a conclusion which made no sense. (he also didnt always correct them which is my only real gripe with him) Just like in sports, even if a team is favored to win and all the models show they should win that isnt how it always plays out. So while his model shows 60/40 the Dems take the Senate that doesn't mean if the Dems dont he was wrong. It just shows one of the other outcomes happened.

What Nate shows about America (and Rachel Bitecofer too although he style is different) is that the average media person and most normal people don't understand how probability works.
 
Handyman said:
Just like in sports, even if a team is favored to win and all the models show they should win that isnt how it always plays out.
Click to expand...

There's a great meme going around where it says if your team has a 2/3 chance of winning and they lost, oh well, upsets happen, but if your candidate has a 2/3 chance of winning and they lost, MATH MUST BE WRONG!
 
I was looking at the Economist forecast and you have to go back to February to find a national poll that had trump in the lead. It is followed by literally hundreds of polls that show Biden in the lead, including from sources that clearly favor trump, versus more legitimate polls that are from sources that are just seeking to report. What that partially tells me is we need these polling firms to figure out more effective ways to poll locally. Until we get rid of the electoral college (something a Democratic government needs to figure out and do, by the way), national polling has proven to be a lot less effective than it used to be in predicting a winner of a presidential election.

Hard to believe that as recently as the year 2000, surveys consistently showed that in addition to most people, most republicans also favored the elimination of the electoral college and thought we should simply use a popular vote to elect the president, the same way we elect every single other office holder at every single level of government in this country. Actually now that I think about it it isn't hard to believe at all, because in addition to their other flaws republicans are disgustingly hypocritical. At the end of the day that is the trait that shows them to be the weak, cowardly little people they really are.
 
Jr. calls for 'army' of supporters to 'protect ballots' Just more normal stuff from this administration.

We need every able-bodied man and woman to join Army for Trump’s election security operation.
Click to expand...
This is where we are in 2020 as a nation.

Mark my words, we are going to see the Kenosha terrorists at polling stations in dozens of states. I believe we will see violence and violent intimidation at an unprecedented scale on election day in this country, and probably on the days leading up to it, when early voting efforts normally have ramped up in recent elections. And if you expect law enforcement to protect the rights of minorities to vote, I have a bridge to sell you.

This election is likely to be fraudulent all right, but like with every other case of projection we've seen out of the hypocritical republicans, it will be them committing the fraud. If they cannot intimidate, kill or otherwise prevent enough people from voting, there are people lined up in positions of authority in enough states to deny the rights of people to have their votes counted accurately and fairly. This election is going to be a joke because there simply are not enough people who will vote for Biden and other Democrats. And this is how a minority of the people are going to destroy this country, finally and once and for all.

I will say this though, since based on appearances alone I could easily be mistaken for a MAGAt, I am going to show up to vote wearing a Biden hat, a Biden t-shirt, and I'll probably put a Biden bumper sticker on my as s. I dare someone to try to intimidate me.
 
WeAreNDHockey said:
I will say this though, since based on appearances alone I could easily be mistaken for a MAGAt, I am going to show up to vote wearing a Biden hat, a Biden t-shirt, and I'll probably put a Biden bumper sticker on my as s. I dare someone to try to intimidate me.
Click to expand...

Careful with that - it's not permitted to wear a specific candidate's gear at a polling site in some states. You'll get turned away.
 
FadeToBlack&Gold said:
Careful with that - it's not permitted to wear a specific candidate's gear at a polling site in some states. You'll get turned away.
Click to expand...

As long as I take off the hat and turn the shirt inside out within a certain number of feet to the door, I'll be fine. I can't say the same thing about the AR-15 toting wannabees.
 
Handyman said:
What Nate shows about America (and Rachel Bitecofer too although he style is different) is that the average media person and most normal people don't understand how probability works.
Click to expand...

This is absolutely true. Someone on Fark was legitimately arguing that a 28% chance is next to nothing. When people pointed out that things with a 28% chance happen all the time (like flipping a coin and getting tails both times), he continued to argue.

Basically, for most people i think anything in the 50-59% range is a tossup, but >=60% is the same as 99% in their heads. They know there's a chance of the upset, but it's minor.
 
unofan said:
This is absolutely true. Someone on Fark was legitimately arguing that a 28% chance is next to nothing. When people pointed out that things with a 28% chance happen all the time (like flipping a coin and getting tails both times), he continued to argue.

Basically, for most people i think anything in the 50-59% range is a tossup, but >=60% is the same as 99% in their heads. They know there's a chance of the upset, but it's minor.
Click to expand...

It’s funny how the perception of statistics change depending on the situation.

Baseball probably does this more than anything else.

If you’re successful more than 30% of the time over your entire career at the plate, you’re in the top 200 all-time.

Only successful 20% of the time? You don’t have much of a career. We’re closing in on 20,000
MLB players, and while most pitchers haven’t hit for 40 years, that’s still only 1% of players all-time who have cracked the gold standard of .300.

And if anyone is ever hitting .400+ in a situation, like RISP, it feels like an automatic outcome even though they’re still more likely to fail.

But flip that around, if a .250 hitter comes up
with a walk-off, it’s baseball. If a candidate with a 25% chance wins, the math is somehow wrong.
 
Status
Not open for further replies.
Back
Top