- 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: Nate silver: the slowdown in CV19 cases in NOLA is unlike anything we have seen elsewhere
Posted on 4/13/20 at 3:04 pm to Antonio Moss
Posted on 4/13/20 at 3:04 pm to Antonio Moss
quote:
No, what I am saying is that, clearly, I can assume that deaths will be significantly higher without social restrictions. I don't need a model to tell me that - I have one, the IC, but its really just belaboring the point
"significantly higher" doesn't tell me anything. Your idea of significant might not be equal to someone elses.
quote:
I have one, the IC, but its really just belaboring the point.
I don't think this is relevant if it's not what was used to justify the decision. Was the IC model used to justify the decision? I'm still interested. Can you fill me in on the estimated number of deaths with no mitigation from the IC model?
But, more importantly, what model that estimated deaths without mitigation was used to justify the shutdown. Obviously, there had to be some expectation of what would happen with no action. Do you know what it was?
1
Posted on 4/13/20 at 3:09 pm to CptRusty
quote:Of course you realize that, in your fantasy scenario, there would be both more deaths not from Corona and more deaths from Corona.
It's not a matter of the net deaths. We can reduce the number of people that die every year by locking everyone in their homes indefinitely. It's a question of cost vs benefit. The number of people that die from corona virus is not likely to be statistically significant
So you're ignoring every possible death, and then basing your projections of "cost" off of that. Wow.
Posted on 4/13/20 at 3:10 pm to Antonio Moss
quote:
That's not true. How is the most accurate model presented in an event the one with bad data?
They ALL had bad data. The polling data was flawed. Your opinion on the election is kind of enlightening to your opinion regarding this shutdown. If the only thing that needs to happen to justify a trillion dollar shutdown is that a model is more accurate than the others, then God help us.
quote:
Those aren't voting percentages, those are chance predictions. 77% in PA means that when he ran four models Clinton won three of them.
I know that those aren't voting percentages. But, that 77% is a HUGE favorite. We all know 77 percent means.
I'ts 77 out of 100. There is no value in reducing it to 3 out of 4.
Florida was 55/45. He was wrong on that state, but that was much more reasonable as to how close that state was.
Posted on 4/13/20 at 3:12 pm to moneyg
quote:
I know that those aren't voting percentages. But, that 77% is a HUGE favorite. We all know 77 percent means.
I'ts 77 out of 100. There is no value in reducing it to 3 out of 4.
Florida was 55/45. He was wrong on that state, but that was much more reasonable as to how close that state was.
So you really don't understand probabilities?
Posted on 4/13/20 at 3:12 pm to ashy larry
quote:
He gave Trump a 28.6% chance of winning which is better than other oddsmakers
Better than other oddsmakers is meaningless.
quote:
Conservative media outlets had Hilary winning
They were also wrong. Do we have to pretend they were right?
quote:
The Fox news poll released the day before the election had Hilary leading by 4% overall.
And, they were wrong.
quote:
The National Review had Hilary winning by a small margin.
Also wrong.
I'm not saying Silver wasn't less wrong than others. I'm saying he was very wrong.
Posted on 4/13/20 at 3:13 pm to HailHailtoMichigan!
It's a holiday weekend so reporting of the numbers has probably been delayed.
Posted on 4/13/20 at 3:16 pm to moneyg
quote:
We all know 77 percent means.
quote:So then you don't know what 77 (or 55) percent means.
Florida was 55/45. He was wrong on that state
Posted on 4/13/20 at 3:17 pm to Korkstand
quote:
It was based on the data available. "Bad" or not, the probability was based on data.
If this isn't a microcosm of what a lot of people are questioning with the decision to shutdown, I don't know what is.
A LOT of people were pointing out that the polling data was flawed. The pointed to oversampling of democrats, a hidden percentage that wouldn't admit to voting for Trump for whatever reasons.
quote:
You keep saying "he", as if the projections were simply an opinion of his. The probabilities were based on polls, and possibly other data. You seem to think Nate was just going around assigning random numbers to states based on his feelings.
I've said nothing of the sort. Get your feelings out of this discussion. Silver's model is sound. The data he used was flawed, and thus his conclusion was very far off the mark in several key states.
quote:
Edit: Oh, and I didn't see where he gave those states 100% odds of Hillary winning. Because that's the only way that a probability can be "wrong". Like I said, zero understanding of probabilities.
By that logic, any model or data input is valid as long as it doesn't result in 100% odds of winning. I'll let you stand by that idiotic statement.
Posted on 4/13/20 at 3:17 pm to madddoggydawg
quote:
So you're ignoring every possible death, and then basing your projections of "cost" off of that. Wow.
wtf are you talking about? I never outlined any "fantasy scenario" or gave any opinion at all of what I think should have been done. I have only called into question potential recklessness on the part of those whose job it is to be experts on these things and advise those whose job it is to make decisions based on their counsel.
Posted on 4/13/20 at 3:20 pm to SECdragonmaster
IDIOT.....if u think this is like a typical flu virus you need to learn how to think critically. This is a pandemic with a virus spreading without any vaccines available. Think about it.....flu can kill.....but not in these numbers.
Posted on 4/13/20 at 3:24 pm to JohnnyKilroy
quote:
So you really don't understand probabilities?
I understand probabilities better than you do. Let's get that out of the way.
Polling data that Silver based his model on, showed that there was a 77% chance she'd win Pennsylvania. That equates to close to a +4 in polling. Michigan was similar. Wisconsin was worse +5.
If your point is that Silver recognized Trump had a chance, I'm not disagreeing. I'm saying that thinking he had a 1 in 3 chance was way off the mark. He was probably a slight favorite heading into election night.
Posted on 4/13/20 at 3:25 pm to moneyg
quote:You've said numerous times "he was wrong".
I've said nothing of the sort. Get your feelings out of this discussion.
quote:No, by my logic a projection can only be "wrong" if the actual result is not possible according to the probability assigned to it. The only way to measure how "valid" or "accurate" a model is would be to run through reality multiple times. This is obviously not possible.
By that logic, any model or data input is valid as long as it doesn't result in 100% odds of winning.
Clearly you haven't gathered yet that the issue is with you using the word "wrong".
Posted on 4/13/20 at 3:25 pm to Korkstand
quote:
So then you don't know what 77 (or 55) percent means.
Posted on 4/13/20 at 3:26 pm to moneyg
quote:
I understand probabilities better than you do. Let's get that out of the way
Clearly
Posted on 4/13/20 at 3:27 pm to moneyg
I wondered how long before some butthurt Trumpkin who doesn't understand statistics came in the thread and complained about Nate Silver
Posted on 4/13/20 at 3:33 pm to CptRusty
quote:
if Dr. Birx and Fauci are up there haphazardly saying "best case scenario is 100k people die", then that's what the decision makers are hearing, and furthermore, that is what drives public perception -which in turn drives the expected response from the administration.
But from what I can tell, they didn't say that.
That's kind of the point.
It's the same thing last week when people were going nuts claiming that Birk said that any death of a person with COVID gets coded as a COVID death. She didn't say that. You can pull the transcript and read it. But it gets repeated so often by people both in and out of the media that it becomes gospel.
And then people takes these gospel truths that they think they know and attach emotional conclusions to them.
Posted on 4/13/20 at 3:35 pm to moneyg
quote:
Florida was 55/45. He was wrong on that state,
Do you know what the word forecast means? It's an estimate. Like when the weatherman said there was 60% chance of rain the other day and it didn't rain at my house. He wasn't wrong, bc he didn't say it WOULD rain. He said there is a 60% chance of it happening. How do people not understand this?
ETA the full quote which still doesn't make sense.
quote:
Florida was 55/45. He was wrong on that state, but that was much more reasonable as to how close that state was.
This post was edited on 4/13/20 at 3:51 pm
Posted on 4/13/20 at 3:40 pm to Antonio Moss
quote:
It's the same thing last week when people were going nuts claiming that Birk said that any death of a person with COVID gets coded as a COVID death. She didn't say that. You can pull the transcript and read it. But it gets repeated so often by people both in and out of the media that it becomes gospel.
And then people takes these gospel truths that they think they know and attach emotional conclusions to them.
See also all the people that melted over JBE's doctor that said LA was thinking about extending the stay at home order an additional 120 days.
The problem was that a 120 day extension wasn't stated, discussed, mentioned or implied by anyone in the press conference.
That didn't stop people like pride87 and others from melting about it like twats.
This post was edited on 4/13/20 at 3:41 pm
Posted on 4/13/20 at 3:43 pm to Antonio Moss
quote:
But from what I can tell, they didn't say that.
That's kind of the point.
There's a video in the link.
Her exact words are: "if we do things together, well, almost perfectly, we could get in the range of 100,000 to 200,000 fatalities"
Start around the 3:00 mark.
The journalist did not misrepresent her statement at all.
Posted on 4/13/20 at 3:43 pm to moneyg
quote:
"significantly higher" doesn't tell me anything. Your idea of significant might not be equal to someone elses.
Well, that's fine but there is no other way to answer the question.
quote:
I don't think this is relevant if it's not what was used to justify the decision. Was the IC model used to justify the decision? I'm still interested. Can you fill me in on the estimated number of deaths with no mitigation from the IC model?
quote:
But, more importantly, what model that estimated deaths without mitigation was used to justify the shutdown. Obviously, there had to be some expectation of what would happen with no action. Do you know what it was?
I guess maybe IC unless there were internal models that haven't been released.
Page 11 of 15
Popular
Back to top
Follow TigerDroppings for LSU Football News