- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: Sensing a little bit of unease from Nate Silver today
Posted on 10/27/20 at 1:01 pm to Eat Your Crow
Posted on 10/27/20 at 1:01 pm to Eat Your Crow
quote:
Too many unknowns in polling nowadays.
That is the purpose of 538 for ABC News
To give the appearance of applying science and math and stuff to their narratives and make cool videos and graphics so Dems can feel good in their confirmations
Posted on 10/27/20 at 1:02 pm to tigerskin
quote:
Sensing a little bit of unease from Nate Silver today
It's like none of you understand the concept of odds or probability.
Posted on 10/27/20 at 1:03 pm to barry
quote:
It's like none of you understand the concept of odds or probability.
Nate Silver certainly didn’t in 2016.
Posted on 10/27/20 at 1:05 pm to Warfox
quote:
Nate Silver certainly didn’t in 2016.
Me: You have a 1/6 chance of rolling the dice and it being a 1
*you roll a 1*
you: lol you fricking nerd loser, see I rolled a one
Posted on 10/27/20 at 1:08 pm to barry
quote:People keep using an easily reproducible event like dice rolling.
Me: You have a 1/6 chance of rolling the dice and it being a 1
*you roll a 1*
you: lol you fricking nerd loser, see I rolled a one
It is not equatable to a single one time event like an election.
Explain to me how someone who predicts that Trump has a 30% chance to win is more accurate than someone who predicts that he has a 1% chance to win or even a 99% chance to win.
If Trump wins, they are all equally as right because you cannot reproduce the event to determine the actual probability, yet, the 1% and 99% would be viewed differently.
Probability is a false narrative in this situation.
It is an unprovable concept and no one can be wrong.
“I predict that Trump has a 50% chance of winning” Is the only truly accurate statement.
This post was edited on 10/27/20 at 1:10 pm
Posted on 10/27/20 at 1:09 pm to tigerskin
So if Trump wins, are people going to stop asking this dude questions?
Posted on 10/27/20 at 1:10 pm to barry
quote:
Me: You have a 1/6 chance of rolling the dice and it being a 1 *you roll a 1* you: lol you fricking nerd loser, see I rolled a one
Sigh. Point went right over your head. Nate has the actual odds WRONG. Presidential election isn’t a like picking a lottery ticket.
It was NEVER a 1/6 chance, that’s why he and these pollsters suck.
Posted on 10/27/20 at 1:10 pm to lsuhunt555
No, they’ll just claim there were warning signs again and promise they adjusted for them, again.
Posted on 10/27/20 at 1:11 pm to Scruffy
quote:
t is an unprovable concept and no one can be wrong.
Well, you look at their track record over multiple elections. They predict literally hundreds of races every year.
If people want to post 538's track record for predicting elections and trash them for that, its fine. To your point, to trash someone for ONE predictions is dumb.
Posted on 10/27/20 at 1:11 pm to Powerman
quote:
The problem is you are looking at it as an all or nothing approach. If someone says that x has a 66% probability of happening and it doesn't happen you think they're "wrong" which says everything about your own stupidity. Low probability events happen all the time.
Objective verifiable data is needed for that.
His source data is absolutely not that. It’s skewed; I believe intentionally skewed. Calculating probability off of that is meaningless.
Posted on 10/27/20 at 1:12 pm to Warfox
quote:
It's like none of you understand the concept of odds or probability.
quote:
Nate Silver certainly didn’t in 2016.
You're actually proving his point with a comment like this.
Posted on 10/27/20 at 1:14 pm to barry
quote:
It's like none of you understand the concept of odds or probability.
Probability as it relates to sampling requires representative sampling.
Posted on 10/27/20 at 1:19 pm to tigerskin
On Election Day 2016, his map had HRC winning Florida, NC, PA, MI, WI, Iowa and AZ. His shooting percentage on swing states was similar to Shaq at the free throw line.
Posted on 10/27/20 at 1:19 pm to Goldrush25
quote:
You're actually proving his point with a comment like this.
No, I’m not. Elections probability doesn’t work in the manner of the example he put forth.
Posted on 10/27/20 at 1:21 pm to barry
quote:
Me: You have a 1/6 chance of rolling the dice and it being a 1
*you roll a 1*
you: lol you fricking nerd loser, see I rolled a one
Another thread wherein the avg person’s complete lack of understanding stats is on display.
Posted on 10/27/20 at 1:29 pm to barry
quote:
It's like none of you understand the concept of odds or probability.
Somebody doesn’t. And then you prove it with the dice scenario.
Posted on 10/27/20 at 1:29 pm to moneyg
quote:
His source data is absolutely not that. It’s skewed; I believe intentionally skewed. Calculating probability off of that is meaningless.
This. I don't care about his probabilities %'s
The problem is the source of those predictors.
There is one type of people who participate in polls. There is a totally different type of people who refuse to, no matter the type of communication used to contact.
This is no longer the days of no caller is landlines of the 90's
These are the days of spam callers constantly...ignoring calls and texts from numbers you don't know. Websites requiring your email address to participate etc.
I'll let you guess which group is more apt to enter those types of communications
Posted on 10/27/20 at 1:32 pm to Flats
People keep using the dice scenario, but if we applied the actual concept behind that to the election, the probability is 50/50.
Posted on 10/27/20 at 1:33 pm to Scruffy
quote:
People keep using an easily reproducible event like dice rolling.
It is not equatable to a single one time event like an election.
Explain to me how someone who predicts that Trump has a 30% chance to win is more accurate than someone who predicts that he has a 1% chance to win or even a 99% chance to win.
If Trump wins, they are all equally as right because you cannot reproduce the event to determine the actual probability, yet, the 1% and 99% would be viewed differently.
Probability is a false narrative in this situation.
It is an unprovable concept and no one can be wrong.
this is where i parked my car
Posted on 10/27/20 at 1:33 pm to Scruffy
quote:
It is an unprovable concept and no one can be wrong.
No application of probability is an attempt to be right. I see that you're stuck on "right or wrong" and I don't know why. You have this very narrow view of the value or probability.
Predictive analysis is applied in a wide array of industries; science, business, medicine, to anticipate events that have not yet occurred with the exact same set of variables, and thus can't be backtested. Just because you can't ever be "proven wrong" doesn't mean that there's not value in attempting to anticipate future events based on past events.
This post was edited on 10/27/20 at 1:35 pm
Popular
Back to top
Follow TigerDroppings for LSU Football News