|
Read the rules in the OP before posting, please.In order to ensure that this thread continues to meet TL standards and follows the proper guidelines, we will be enforcing the rules in the OP more strictly. Be sure to give them a re-read to refresh your memory! The vast majority of you are contributing in a healthy way, keep it up! NOTE: When providing a source, explain why you feel it is relevant and what purpose it adds to the discussion if it's not obvious. Also take note that unsubstantiated tweets/posts meant only to rekindle old arguments can result in a mod action. |
On December 13 2017 12:44 jalstar wrote:Show nested quote +On December 13 2017 12:41 mozoku wrote:On December 13 2017 12:33 jalstar wrote:On December 13 2017 12:29 mozoku wrote:On December 13 2017 12:08 jalstar wrote:On December 13 2017 12:06 mozoku wrote:I'm in agreement with Plansix on the value of this Upshot model. It's literally worthless besides for entertainment purposes and has a useful life of two hours. On December 13 2017 11:42 jalstar wrote:On December 13 2017 11:41 Plansix wrote:On December 13 2017 11:39 jalstar wrote:On December 13 2017 11:38 Plansix wrote:
Our model predicts the outcome of an election with closed ballets. This is a waste of Human Resources and time, since more accurately results can be obtained by waiting for the real count to come in. Well obviously it's 50-50, either Moore wins or Jones does! Someone already won! Predictions no longer matter!!!! It's a level of confidence, not a probability. Pretty basic statistics, there's no need to be obtuse here. This depends on your statistical assumptions and, empirically speaking, you're more likely than not wrong here seeing as most electoral forecasters aren't using frequentist methods. Or probably shouldn't be anyway. I would have ignored this if it wasn't for the snarky "basic statistics" comment. Most people are taught the difference between confidence and probability in stats 101 when being taught how to interpret confidence intervals. edit: and empirically it's not a probability because as Plansix said, it already happened. I don't understand what statistics has to do with what Plansix said. He was making a point about the utility of the model. As for the bold, you've got a bit to learn in this area. Quit while you're ahead. I don't have time to read all of that but typically the word probability is used in terms of predictions and confidence is used for accuracy of measurements of things that already happened, which is what we're dealing with here. And even if you find a counter-example, I did say typically, and a "win" for you would just be semantics. Your whole point was semantics lol. My point was "if you're going to be a technical smartass, at least be a correct one to avoid looking foolish." Fyi, this isn't a mere "counterexample." It's an entirely independent statistical framework, and one that's more commonly used in electoral forecasting than the one you're familiar with. Fine. It's not a frequentist "confidence". It's also not a probability, because the result already happened.
It's a frequentist confidence level or a Bayesian probability depending on the Upshot's work, which I'm not familiar enough with to be drawing conclusions from. I'd bet on the latter.
For practical interpretive purposes, they're the same though.
|
|
|
On December 13 2017 12:49 Karis Vas Ryaar wrote:
personally I think Jones should just go independent and caucus with the Dems. prob won't though
but... but...
BERNIE SANDERS IS THE ANTI CHRIST FOR DOING THAT!
Yet Jones could probably get away with it
|
Please court black voters democrats. They win ya all elections.
|
On December 13 2017 12:50 mozoku wrote:Show nested quote +On December 13 2017 12:44 jalstar wrote:On December 13 2017 12:41 mozoku wrote:On December 13 2017 12:33 jalstar wrote:On December 13 2017 12:29 mozoku wrote:On December 13 2017 12:08 jalstar wrote:On December 13 2017 12:06 mozoku wrote:I'm in agreement with Plansix on the value of this Upshot model. It's literally worthless besides for entertainment purposes and has a useful life of two hours. On December 13 2017 11:42 jalstar wrote:On December 13 2017 11:41 Plansix wrote:On December 13 2017 11:39 jalstar wrote:
[quote]
Well obviously it's 50-50, either Moore wins or Jones does! Someone already won! Predictions no longer matter!!!! It's a level of confidence, not a probability. Pretty basic statistics, there's no need to be obtuse here. This depends on your statistical assumptions and, empirically speaking, you're more likely than not wrong here seeing as most electoral forecasters aren't using frequentist methods. Or probably shouldn't be anyway. I would have ignored this if it wasn't for the snarky "basic statistics" comment. Most people are taught the difference between confidence and probability in stats 101 when being taught how to interpret confidence intervals. edit: and empirically it's not a probability because as Plansix said, it already happened. I don't understand what statistics has to do with what Plansix said. He was making a point about the utility of the model. As for the bold, you've got a bit to learn in this area. Quit while you're ahead. I don't have time to read all of that but typically the word probability is used in terms of predictions and confidence is used for accuracy of measurements of things that already happened, which is what we're dealing with here. And even if you find a counter-example, I did say typically, and a "win" for you would just be semantics. Your whole point was semantics lol. My point was "if you're going to be a technical smartass, at least be a correct one to avoid looking foolish." Fyi, this isn't a mere "counterexample." It's an entirely independent statistical framework, and one that's more commonly used in electoral forecasting than the one you're familiar with. Fine. It's not a frequentist "confidence". It's also not a probability, because the result already happened. It's a frequentist confidence level or a Bayesian probability depending on the Upshot's work, which I'm not familiar enough with to be drawing conclusions from. I'd bet on the latter. For practical interpretive purposes, they're the same though.
Show me a published study that uses Bayesian methods that refers to "probability" to describe an event that already happened. (not another Wikipedia article, just something I can ctrl-F)
Actually your whole argument, while hard to decipher, might be "probability theory is used to calculate confidence levels" in which case I'm done since that's so nitpicky it's silly.
|
Man black people are fucking stellar. Their numbers fucking shot Moore in the dick. This is so wonderful.
|
Then the Dems better lose their love for pro Corporate positions and inject new blood into the party.
|
I always like the idea of being a snowflake. Being from the Northeast, snows give me anxiety, it is powerful, it nearly killed me. And when snowflake band together, they can bury anything.
Tonight, we buried a child molestor in the deep South.
In 2018, it will be an avalanche.
|
|
|
They have been in the minority since 2010. New blood is the only option. But it's about good candidates.
|
|
|
I just hope they learn the lesson that focusing on the positives of your candidacy including being a civil rights activist who fought the KKK (admittedly when running against scum incarnated in human form) runs up a larger margin than many predict.
On December 13 2017 12:59 {CC}StealthBlue wrote:
I still have a feeling that Moore is going to speak and demand a recount. No idea why though.
The man's a crazy person, it wouldn't surprise me in the slightest.
|
I still have a feeling that Moore is going to speak and demand a recount. No idea why though.
|
|
|
"We lost voters because our base is is a complete fucking mess"
|
|
|
Jones was a legitimately good candidate on paper. Much better than you'd usually see in Alabama. If he's a talented politician he might be able to retain his seat, though he's definitely not a favorite (I'd say like ~30% chance in 2020). Like if he networks well enough and doesn't just say "fuck ya'll i'm in dc now." A lot of the people against him were doing so based on an extreme misrepresentation of a short sound bite about abortion - he now has 3 years to prove that people from Alabama can trust him on the issue. They literally claimed he was in favor of "full birth abortion", which is quite a way to describe cesarian sections.
I really am so fucking happy about him winning though.
|
Never thought I'd say it, but I'm sorta proud of Alabama for passing this super low bar. Was certain they would go full third-world state on us.
Also, it's a hilarious fuck you to Bannon, Trump, and pedos.
|
On December 13 2017 12:53 jalstar wrote:Show nested quote +On December 13 2017 12:50 mozoku wrote:On December 13 2017 12:44 jalstar wrote:On December 13 2017 12:41 mozoku wrote:On December 13 2017 12:33 jalstar wrote:On December 13 2017 12:29 mozoku wrote:On December 13 2017 12:08 jalstar wrote:On December 13 2017 12:06 mozoku wrote:I'm in agreement with Plansix on the value of this Upshot model. It's literally worthless besides for entertainment purposes and has a useful life of two hours. On December 13 2017 11:42 jalstar wrote:On December 13 2017 11:41 Plansix wrote:
[quote]
Someone already won! Predictions no longer matter!!!! It's a level of confidence, not a probability. Pretty basic statistics, there's no need to be obtuse here. This depends on your statistical assumptions and, empirically speaking, you're more likely than not wrong here seeing as most electoral forecasters aren't using frequentist methods. Or probably shouldn't be anyway. I would have ignored this if it wasn't for the snarky "basic statistics" comment. Most people are taught the difference between confidence and probability in stats 101 when being taught how to interpret confidence intervals. edit: and empirically it's not a probability because as Plansix said, it already happened. I don't understand what statistics has to do with what Plansix said. He was making a point about the utility of the model. As for the bold, you've got a bit to learn in this area. Quit while you're ahead. I don't have time to read all of that but typically the word probability is used in terms of predictions and confidence is used for accuracy of measurements of things that already happened, which is what we're dealing with here. And even if you find a counter-example, I did say typically, and a "win" for you would just be semantics. Your whole point was semantics lol. My point was "if you're going to be a technical smartass, at least be a correct one to avoid looking foolish." Fyi, this isn't a mere "counterexample." It's an entirely independent statistical framework, and one that's more commonly used in electoral forecasting than the one you're familiar with. Fine. It's not a frequentist "confidence". It's also not a probability, because the result already happened. It's a frequentist confidence level or a Bayesian probability depending on the Upshot's work, which I'm not familiar enough with to be drawing conclusions from. I'd bet on the latter. For practical interpretive purposes, they're the same though. Show me a published study that uses Bayesian methods that refers to "probability" to describe an event that already happened. (not another Wikipedia article, just something I can ctrl-F) Actually your whole argument, while hard to decipher, might be "probability theory is used to calculate confidence levels" in which case I'm done since that's so nitpicky it's silly.
This isn't something you're going to gain an understanding of from a conversation on an internet forum. This is my last post, and you'll have to learn the rest on your own if you want to know more.
There's actually more than one interpretation of Bayesian probability, but one interpretation is based on subjectivity. If I know the event already happened, than the probability would be 1. If the already event happened, but there's uncertainty in my state of knowledge (as was the case tonight), the probability would be between 0 and 1.
Your understanding of frequentist probability is correct.
Whether a Bayesian or frequentist interpretation is appropriate depends on how the Upshot came up with its number. Neither of us know that, so we can't come to a conclusion.
|
This man isn't going to get dragged before congress in the next 10 years to answer for the hot bullshit he peddles. He is like Alex Jones, but with less supplements.
|
|
|
|
|
|
EPT
